Molecular Phylogenetic Characterization of Microbial Comunity Dynamics Asociated with Freshwater Stream Environmental Organic Matter Sources by Moli Michele Newman A dissertation submited to the Graduate Faculty of Auburn University in partial fulfillment of the requirements for the Degree of Doctor of Philosophy Auburn, Alabama December 12, 201 Keywords: leaf breakdown; microbial comunity; succession; stream; macroinvertebrates; anthropogenic disturbance Copyright 201 by Moli Michele Newman Approved by Jack W. Feminela, Co-chair, Professor of Biological Sciences Mark R. Liles, Co-chair, Associate Professor of Biological Sciences Nedret Bilor, Associate Professor of Mathematics and Statistics B. Graeme Lockaby, Professor of Forestry and Wildlife Sciences ii Abstract The efects of various biotic and abiotic factors on microbial comunity dynamics during leaf breakdown were assessed through a series of in situ leaf breakdown studies within the Fort Bening Military Instalation (FBMI), Georgia, USA. Leaf liter microbial comunity composition was mainly controled by incubation time and, to a lesser extent, by leaf chemistry. Instream sediment disturbance and its associated efects on stream physicochemical conditions drasticaly altered bacterial assemblage composition during leaf breakdown. Chapter 2 described a protocol for purifying genomic DNA from environmental sources for use in a polymerase chain reaction (PCR). This protocol was necessary because many of the techniques utilized in this dissertation involved extraction and amplification of genomic DNA from organic mater and often required purification of the genomic DNA. The protocol involves embeding genomic DNA extract in an agarose plug and incubation within a formamide and saline solution. The purified DNA can then be extracted from the agarose and used as a template for PCR. A test of this protocol using red maple leaf genomic DNA yielded significantly more amplicons using ~20 ng of purified DNA compared to extracted DNA alone. Chapter 3 described a 128-d in situ leaf breakdown study within a single stream at FBMI to assess the efects of shreding macroinvertebrates on leaf iii liter microbial comunities. Contrasting mesh sizes (6.35- and 1-m mesh) were used to reduce shreder macroinvertebrate abundance, and microbial comunity composition was characterized over 9 dates. Macroinvertebrate results revealed no reduction in shreder abundance, sugesting that the use of 1-m mesh may be inapropriate in streams where the dominant shreders are fairly smal and slender (e.g., Polypedilum and Leuctra spp.). Chapter 4 described the diferences in microbial comunity composition betwen leaf species of strongly contrasting leaf chemistries and associated breakdown rates. Maple and oak leaf species were used due to their drasticaly diferent leaf chemistries (e.g., higher percent lignin and celulose in oak). Leaf chemistry diferences resulted in significantly diferent microbial comunity composition measured using both ribosomal intergenic spacer analysis (RISA) and phospholipid faty acid analysis (PLFA). These diferences in microbial comunity composition were strongest during early leaf breakdown and decreased over time. Time was the main factor found to structure leaf liter bacterial assemblages with early and later breakdown times having diferent bacterial assemblage compositions. Chapter 5 described a 64-d in situ leaf breakdown study at FBMI to quantify the efects of sediment disturbance on leaf liter bacterial assemblages. A variety of response variables were measured including several physicochemical conditions (streamwater temperature, pH, depth, current velocity), leaf breakdown, and bacterial assemblage composition (measured using RISA and bar-coded pyrosequencing). The main physicochemical iv condition measured in this study that afected bacterial assemblage composition of leaf liter was streamwater pH, which was correlated to disturbance intensity, which in turn correlated to sediment. Overal, results showed sediment disturbance significantly altered leaf liter bacterial assemblage composition and was associated with a shift towards an assemblage capable of surviving harsher environmental conditions (e.g., increased pH, decreased dissolved oxygen). v Acknowledgements I would like to thank my mentors, Drs. Jack Feminela and Mark Liles, for their continued guidance, suport, and friendship during this dissertation work. I also thank my comite members, Drs. Nedret Bilor and Graeme Lockaby, for always being suportive and providing assistance throughout this dissertation. I am also very grateful to Dr. Joseph Kloeper for serving as university reader. Many thanks to the folowing for contributing to both my development as a graduate student and the research within this dissertation: Nancy Caps, Dr. Abel Carrias, Erin Consuegra, Ann Marie Gode, Emily Hartfield, Laura Heck, Dr. Brian Helms, Miler Jarrel, Kavita Kakirde, Dr. Ely Kosnicki, Dr. Andrew Land, Brian Lowe, Jahangir Hossain, Dr. Kely Maloney, Dr. Richard Mitchel, Shamia Nasrin, Dr. Larissa Parsley, Chao Ran, Susan Reithel, Brad Schneid, Stephen Sefick, and Malachi Wiliams. I would especialy like to thank Britni Bryant for her help and dedication in the laboratory. Funding from the U.S. Department of Defense?s Strategic Environmental Research and Development Program (SERDP) and an Auburn University Graduate Schol research award suported this research. Thank you to my entire family for their suport and encouragement during this endeavor. Most of al, I would like to thank my husband, Shelby Newman, for his continuous and loving suport and understanding during this dissertation. I could not have done this without you. vi Table of Contents Abstract..................................................................................................................ii Acknowledgements................................................................................................v List of Tables........................................................................................................ix List of Figures........................................................................................................x I. Chapter I: Literature Review...............................................................................1 A. Literature Review....................................................................................1 B. Sumary..............................................................................................22 C. Literature Cited.....................................................................................25 I. Chapter I: Purification of Genomic DNA Extracted from Environmental Sources for use in a Polymerase Chain Reaction...........................................42 A. Abstract................................................................................................42 B. Introduction...........................................................................................43 C. Related Information..............................................................................43 D. Protocol................................................................................................44 1. Materials..........................................................................................44 2. Method.............................................................................................44 E. Discussion............................................................................................46 F. Troubleshoting....................................................................................47 G. Literature Cited.....................................................................................49 vii II. Chapter II: Effects of Benthic Macroinvertebrates on Microbial Comunity Associated with Leaf Breakdown in Smal Coastal Plains Streams...............51 A. Abstract................................................................................................51 B. Introduction...........................................................................................52 C. Methods................................................................................................54 1. Study Site........................................................................................54 2. Experimental Design.......................................................................55 3. Microbial Lipids and Comunity Characterization..........................59 4. Mesh Control Study ........................................................................62 5. Hypotheses and Analyses ..............................................................62 D. Results.................................................................................................64 E. Discussion............................................................................................67 F. Literature Cited.....................................................................................71 IV. Chapter IV: Influence of Leaf Species on Liter Breakdown and Microbial Succession in a Smal Forested Stream........................................................84 A. Abstract................................................................................................84 B. Introduction...........................................................................................85 C. Methods................................................................................................88 1. Study Site........................................................................................88 2. Experimental Design.......................................................................89 3. Microbial Lipids and Comunity Characterization..........................92 4. Data Analyses ................................................................................96 D. Results.................................................................................................97 1. Physicochemical Conditions............................................................97 vii 2. Liter Breakdown..............................................................................98 3. Microbial Comunity Characterization............................................99 E. Discussion..........................................................................................100 F. Literature Cited...................................................................................106 V. Chapter V: Effects of Sediment Disturbance on Bacterial Assemblage Associated with Leaf Breakdown in Smal Coastal Plains Streas..............123 A. Abstract..............................................................................................123 B. Introduction.........................................................................................124 C. Methods..............................................................................................128 1. Study Sites....................................................................................128 2. Experimental Design.....................................................................129 3. Bacterial Assemblage Characterization.........................................132 4. Data Analyses ..............................................................................135 D. Results...............................................................................................138 1. Physicochemical Conditions..........................................................138 2. Liter Breakdown............................................................................139 3. Macroinvertebrates .......................................................................140 4. Bacterial Assemblage Characterization.........................................140 E. Discussion..........................................................................................145 F. Literature Cited...................................................................................151 ix List of Tables Table 3.1. Sumarized physicochemical variables (mean ? 1SE) and leaf breakdown rates for each mesh and leaf species treatment..............77 Table 3.2. Macroinvertebrate metrics (mean ? 1SE) for each leaf species and mesh treatment cobination during an 128-d incubation..................79 Table 5.1. Study streams at Fort Bening Military Instalation (FBMI) identified by their military compartment, UTM coordinates, predominant land use, and disturbance index value (proportion of watershed as bare ground and road cover). Disturbance intensity values from Maloney et al. (205)......................................161 Table 5.2. MID barcodes used to tag each PCR product during pyrosequencing of bacterial assemblages.......................................162 Table 5.3. Mean ? 1SE current velocity, depth, and temperature for each site and sediment disturbance treatment.........................................163 Table 5.4. Macroinvertebrate metrics (mean ? 1SE) for each leaf species and disturbance treatment combination during a 64-d incubation.........................................................................................165 Table 5.5. Mean (? 1SE) diversity, richness, and OTU estimates for each time and sediment disturbance treatment sampled during pyrosequencing................................................................................170 x List of Figures Figure 1.1. Stepwise process of leaf breakdown as proposed by Petersen and Cumins, 1974..........................................................................40 Figure 1.2. Comparison of traditional stepwise and more recent overlapping aproach to studying leaf liter breakdown, accounting for both (a) the factors acting on degrading leaf liter and (b) primary and secondary products produced during breakdown.............................41 Figure 2.1. Bar graph representing amplified environmental genomic DNA isolated from red maple (Acer rubrum) leaf liter..............................50 Figure 3.1. Ash free dry mass (AFDM) remaining over time during breakdown of red maple and water oak leaf packs of fine and coarse mesh treatent incubated for 128 d in Kings Mil Creek, GA, USA. Points on graph represent average AFDM remaining (%) ? 1SE.........................................................................................78 Figure 3.2. Relative abundance (%) of select bacterial lipid markers of red maple and water oak leaf packs over the 128-d incubation in Kings Mil Creek, GA, USA...............................................................80 Figure 3.3. Relative abundance (%) of select fungal lipid markers of red maple and water oak leaf packs over the 128-d incubation in Kings Mil Creek, GA, USA...............................................................81 Figure 3.4. Average linkage dendrogram based on Bray-Curtis dissimilarity measures betwen red maple leaf pack bacterial assemblage ribosomal intergenic spacer analysis (RISA) profiles of coarse and fine mesh treatments.................................................................82 Figure 3.5. Average linkage dendrogram based on Bray-Curtis dissimilarity measures betwen water oak leaf pack bacterial asseblage ribosomal intergenic spacer analysis (RISA) profiles of coarse and fine esh treatments.................................................................83 xi Figure 4.1. Mean (? 1SE ) % of ash free dry mass (AFDM) remaining over time during breakdown of red maple (o) and water oak (!) leaf packs incubated for 128 d in Kings Mil Creek, GA, USA...............117 Figure 4.2. Relative abundance (%) of bacterial lipid markers of red maple and water oak leaf packs over the 128-d incubation in Kings Mil Creek, GA, USA. Points on graph represent average relative abundance (%) ? 1SE. (* = p<0.01, ns = not significantly diferent)..........................................................................................118 Figure 4.3. Relative abundance (%) of fungal lipid markers of red maple and water oak leaf packs over the 128-d incubation In Kings Mil Creek, GA, USA. Points on graph represent average relative abundance (%) ? 1SE. (* = p<0.01, ns = not significantly diferent)..........................................................................................119 Figure 4.4. Dendrogram of ribosomal intergenic spacer analysis (RISA) electropherograms displaying bacterial assemblage similarities calculated using the Ward?s method based on Jaccard?s similarity coeficient.........................................................................120 Figure 4.5. Denaturing gradient gel electrophoresis (DGE) analysis of red maple leaf pack bacterial assemblages. Uper case leters indicate sequenced ribotypes. (Ribotype key: A = Comamonas, B = Sphingopyxis, C = Herbaspirilum, D = Nitrosospira, E = Mesorhizobium, F = Ralstonia, G = Colimonas)......................121 Figure 4.6. Denaturing gradient gel electrophoresis (DGE) analysis of water oak leaf pack bacterial assemblages. Uper case leters indicate sequenced ribotypes. (Ribotype key: B = Sphingopyxis, E = Mesorhizobium, H = Sphingomonas, I = Aquabacterium, J = Citrobacter, K = Thiobacilus)....................................................122 Figure 5.1. Ash-free dry mass (AFDM) remaining (%) over time from low (solid) and high (holow) sediment disturbance sites for both maple (circles) and oak (triangles) leaf litter. Exponential decay parameter estimates were derived as the slope of the regression line of ln(AFDM remaining) against time.........................................164 Figure 5.2. Shanon diversity of macroinvertebrates (H?) and associated sediment (g) in maple (solid circles) and oak (holow circles) leaf packs. Spearman?s rank correlations were used to describe the macroinvertebrate diversity to sediment relationships. Trend lines shown indicate significant relationships (p<0.05)...................166 xii Figure 5.3. Macroinvertebrate taxon richness (S) and associated sediment (g) in maple (solid circles) and oak (holow circles) leaf packs. Spearman?s rank correlations were used to describe the macroinvertebrate taxa richness to sediment relationships. Trend lines shown indicate significant relationships (p<0.05).........167 Figure 5.4. Nonmetric multi-dimensional scaling (NMDS) plots based on Bray-Curtis siilarities betwen maple leaf liter samples from low- (!) and high- (!) sediment disturbance sites for days 8, 16, 32, and 64.......................................................................................168 Figure 5.5. Nonmetric multi-dimensional scaling (NMDS) plots based on Bray-Curtis siilarities betwen oak leaf liter samples from low- (!) and high- (!) sediment disturbance sites for days 8, 16, 32, and 64.......................................................................................169 Figure 5.6. Nonmetric multi-dimensional scaling (NMDS) plots derived from pairwise unweighted UniFrac distances betwen maple leaf liter samples from pyrosequenced low- (!) and high- (!) sediment disturbance sites for days 32 and 64. Stress level = 0.081............171 Figure 5.7. Comparison of relative abundance of the most comon bacterial phyla found in pyrosequenced maple leaf pack samples grouped by days in stream and sediment disturbance (high vs. low)...........172 Figure 5.8. Comparison of relative abundance of Proteobacteria classes found in pyrosequenced maple leaf pack samples grouped by days in stream and sedient disturbance (high vs. low)................173 Figure 5.9. Conceptual model ilustrating the predicted shift in leaf pack bacterial assemblage composition under increased sedimentation and catchment disturbance intensity (see Maloney et al., 205) toward a bacterial assemblage dominated by taxa capable of surviving harsher instrea environental conditions (e.g. decreased dissolved oxygen and increased pH)............................174 ! 1 CHAPTER I A. LITERATURE REVIEW Allochthonous leaf liter inputs represent a significant energy source for stream ecosystems, especialy within smal, forested watersheds. In their classic study, Fisher and Likens (1973) demonstrated that ~99% of the energy fueling Bear Brook, at the Hubard Brook Experimental Forest, New Hampshire, was from alochthonous sources. Of these inputs, leaf liter in particulate form composed 44.2% of the energy entering the stream, either through liter fal or wind transport. Leaf liter entering streams may act either as a nutrient source or sink, depending on ambient nutrient levels and the demands of stream organisms (Tate and Gurtz, 1986). These and other early studies (e.g. Minshal, 1967; Cumins, 1974) ilustrated the importance of alochthonous leaf liter inputs to stream energy and nutrient flow, and sparked numerous subsequent studies investigating factors afecting the processes of leaf breakdown, and thus the release of energy and nutrients to the recipient ecosystem. Leaf breakdown, or decomposition, is ?the combined result of physical and biological mineralization and transformation processes, resulting in the generation of CO 2 and other inorganic compounds, dissolved and fine-particulate organic mater (DOM and FPOM, respectively), and decomposer biomass? (Hieber and Gessner 2002). Historicaly, breakdown has ben viewed as a ! 2 stepwise process consisting of 3 temporaly distinct phases (Fig. 1.1), leaching, microbial conditioning, and fragmentation (Petersen and Cumins, 1974; Webster and Benfield, 1986; Boulton and Boon, 191; Abelho, 201). However, in recent years, some have demonstrated that the phases of leaf breakdown actualy overlap and are not as distinctly separate as previously thought (Gessner et al., 199). Leaf litter (hereafter ?liter?) typicaly enters the stream as coarse particulate organic mater (CPOM) and begins leaching, where soluble organic and inorganic components are released. During this portion of breakdown, liter often is a nutrient source, releasing soluble sugars and polyphenolic compounds (Nykvist, 1961; Suberkrop et al., 1976). Typicaly, leaching occurs during the first 24-48 h of incubation within a stream but can occur up to 7 d depending on leaf species and other environmental variables (Nykvist, 1963; Canhoto and Graca, 196). Leaching can account for a rapid loss ranging from ~4-42% of initial mass (Canhoto and Graca, 196; Maloney and Lamberti, 195). Leaching efects on mass loss varies greatly among leaf species; thus, some researchers have atempted to pre-leach liter to reduce variation in initial leaf breakdown from leaching, although this practice often produces atypical breakdown rates (Boulton and Boon, 191). Oven drying of liter also can afect leaching through its efects on leaf cuticular structure, which often increases breakdown rate (Taylor and Barlocher, 196). Instream leaching of soluble liter components is folowed by microbial colonization and conditioning. At this point, liter is typicaly low in N but high in C, ! 3 which can serve as a substrate and fod source for stream icroorganisms (Cumins, 1974). Microbes consume this C-rich source, which, in turn, increases microbial biomass and liter quality for macroinvertebrates and other secondary consumers. As microbial biomass increases, nutrients are assimilated and transformed leading to increases in N content and nutrient quality of microbes (Kaushik and Hynes, 1971). As they grow and reproduce, microbes also produce several extracelular enzymes that can mediate microbial degradation of CPOM (Sinsabaugh et al., 191), thus ?conditioning? liter. Sinsabaugh and Morhead (194) showed that enzymes involved in lignocelulose degradation and the cycling of nutrients, including N and P, were the most important to microbial conditioning of liter. The contribution of microorganisms to leaf mass loss is relatively low (~22-27%) compared to stream acroinvertebrates (~51-64%) (Hieber and Gessner, 202). However, macroinvertebrate preference of liter is largely afected by microbial conditioning (Petersen and Cumins, 1974; Wright and Covich, 205). Thus, although overal litter mass loss due directly to microbes is low, microbial conditioning is fundamentaly important to liter colonization and high mass loss from macroinvertebrates and, thus, nutrient / energy release to higher trophic levels. Macroinvertebrates dependence on microbial conditioning as wel as nutritional content in liter has long ben known. Cumins (1974) analogized this dependence as the ?peanut butter-cracker? relationship, highlighting the nutritional importance of microbes (?peanut buter?) on nutritionaly low-quality leaf liter (?cracker?). Presumably because of their ability to transform nutrients, microbes have ben shown to play a role in ! 4 regulating macroinvertebrate feding and nutrition (Cargil et al., 1985; Arsufi and Suberkrop, 1989). Regardless of whether microbes influence breakdown by transforming nutrients, alter physical leaf structure, or simply consume liter for their own C source, it is clear that they play a major role in breakdown and degradation of leaf liter and the subsequent cycling of energy and nutrients. Fragmentation is typicaly thought to occur after liter has ben conditioned, which can be caused by shreding from stream acroinvertebrates (Walace and Webster, 196), as wel as by physical abrasion from stress exerted by flowing water. Overal, fragmentation results in the physical conversion of residual CPOM to FPOM (Cumins, 1974; Allan, 195). CPOM- consuming shreders often typicaly consist of macroinvertebrates in the aquatic insect orders Plecoptera, Trichoptera, and Diptera (Tachet et al., 1987; Walace and Webster, 196; Graca, 201). Shreding macroinvertebrates can cause overal liter mass loss of up to 63% (Hieber and Gessner 202) for some leaf species. Aside from the nutrient content obtained from degrading litter, shreders also may derive key limiting nutrients by consuming associated liter microbes, which have a higher nutritional content than unconditioned litter (Moriarty and Pulin, 1987), and can actualy show preference for conditioned liter. Arsufi and Suberkrop (1989) demonstrated macroinvertebrate preference for fungal- colonized liter by Diptera, Plecoptera, and Trichoptera, indicating that microbial assemblages can influence macroinvertebrate colonization and subsequent breakdown. As macroinvertebrates fragment leaf liter, it is further broken down into smaler particulate organic mater as wel as digested and released back into ! 5 the stream environment as FPOM (Cumins, 1974). An aditional source of FPOM comes from the flocculation of dissolved organic mater (DOM) folowing leaching and microbial assimilation (Cumins, 1974). In terms of the factors afecting the timing relationship betwen liter leaching and microbial conditioning, there is likely some critical concentration of inhibitory compound(s) that, once leached, alows for colonization by previously inhibited microbes. Yet, other microbes may be inhibited by a diferent concentration of the compound and colonize at a later or earlier time. This diferential response creates a process where microbial colonization and liter conditioning overlaps with leaching (Gessner et al., 1999); in contrast, traditional studies may have viewed microbial colonization as a portion of a stepwise process that always folowed leaching (e.g. Cumins, 1974). A similar situation likely is true for macroinvertebrate colonization and fragmentation where some, but not all, macroinvertebrates are capable of digesting certain leaf liter only after liter reaches some critical point of palatability. Then, based on the species- specific limitations of each macroinvertebrate group, successional colonization would occur with some, but not al, macroinvertebrates colonizing only after microbial conditioning. An example of this preferential colonization by stream macroinvertebrates folowing leaching was observed in a study by Pereira, et al. (198) where they observed arthropod colonization folowing stabilization of polyphenol content. Leaching, microbial conditioning, and fragmentation al occur during leaf breakdown; however, these processes likely not only occur in an overlaping (vs. sequential) progression (Fig. 1.2a), but they also may account ! 6 for production of primary or secondary products that may regulate breakdown (Fig. 1.2b). Given that microbes (including bacteria and fungi) contribute sizably to overal leaf mass loss during breakdown (Hieber and Gessner, 202), studies have explored the separate contributions of microbial consumers to the process. Hieber and Gessner (202) found that fungi contributed to ~15 and 18% of overal mass loss in alder and wilow leaves, respectively (see also Gessner, 197). In constrast, Hieber and Gessner (202) found bacteria to contribute a leaf mass loss contribution of only ~13 (alder) and 9% (wilow). These estimates were made after taking into account bacterial cel turnover rates as previous studies only measured bacterial biomass at a single point in time, and thus likely underestimated bacterial biomass lost from substrate (Hieber and Gessner, 202). Regardless, most studies conclude that fungi exert a greater contribution to liter breakdown than bacteria (Suberkrop and Klug, 1976; Baldy et al., 1995; Hieber and Gessner, 202). There is a wealth of evidence sugesting time-dependent colonization of instream liter by fungi and bacteria. Several studies have shown that fungi dominate microbial biomass during the initial phases of breakdown and decrease in biomass over time (Triska, 1970; Kaushik and Hynes, 1971; Suberkrop and Klug, 1976). Colonization of liter by fungi also has ben shown to exhibit succession (below, see also Suberkrop and Klug, 1976). Although much lower than fungi, bacterial biomass tends to increase as fungal biomass decreases (Suberkrop and Klug, 1976; Weyers and Suberkrop 196). This interaction ! 7 betwen fungi and bacteria has ben described as an antagonistic relationship. By inoculating microcosms with either fungi or bacteria alone and a combination of bacteria and fungi, Mile-Lindblom and Tranvik (203) observed higher biomass accumulation for both fungi and bacteria alone compared to in coexistence. Fungi apear more capable of colonizing freshly senesced liter because of their ability to form hyphae, which can penetrate the resistant leaf cuticle (Suberkrop and Klug, 1980) and provide a colonization benefit over bacteria. This advantage alows fungi to contribute more actively to breakdown during early phases; in contrast, bacteria are generaly more active during later stages once fungi have degraded the leaf cuticle and other structural components. Research on specific fungal taxa involved in liter breakdown has typicaly revealed dominance by aquatic hyphomycetes (Ascomycota) over terrestrialy derived fungi. Early studies by Suberkrop and Klug (1976) found Flagelospora curvula Ingold and Lemoniera aquatica DeWild to be the dominant fungal species during the first 4-6 wk of breakdown of white oak and pignut hickory leaves. After 6-8 wk of incubation, there was a succession from early fungal species to Alatospora acuminata Ingold, Tetracladium archalianum DeWild, and also Anguilospora sp. and Clavariopsis aquatica (Suberkrop and Klug 1976) in liter packs of white oak and pignut hickory. More recently, Nikolcheva and Barlocher (204) found fungi of the division Ascomycota dominated linden, maple, bech, and birch liter. The fungal divisions Basidiomycota and ! 8 Chytridiomycota also were present, as wel as Oomycota and Zygomycota in some seasons (Nikolcheva and Barlocher, 204). Comparatively less work has ben conducted to study specific bacterial taxa present during liter breakdown than fungi, likely because bacteria contribute less to the breakdown process than fungi (Baldy et al., 195; Hieber and Gessner, 202). Early studies on microbial comunity structure found most bacterial cultured isolates were Gram-negative species from the genera Flexibacter, Flavobacterium, Cytophaga, Achromobacter, and Pseudomonas (Suberkrop and Klug, 1976). More recent work also has isolated bacteria from the Cytophaga-Flavobacterium-Bacteroidetes group as wel as members of the phyla Proteobacteria and Actinobacteria (Das et al., 206). Suberkrop and Klug (1976) was unable to culture members of the Actinobacteria phylum in their study, but this discrepancy could be because of methodological limitations. For example, the above study used dilutions plated onto PYG agar with an incubation of 2 wk for growth of viable bacterial cels, which was a very comon medium and at the time was known to yield the highest viable cel counts and diversity. However, many bacterial taxa require >2 wk for cultivation (Janssen et al., 202) and often have complex metabolic requirements unknown at the time of the previous study (see below, Aman et al., 195). In adition, methodological limitations, including those mentioned above, may have prevented identification of other liter microbial comunity members. Quantitative methods used to examine stream icrobial comunity composition within liter have varied greatly and, in so doing, provided ! 9 researchers varying degrees of resolution on questions involving comunity structure. Early on, studies focused on quantifying overal microbial respiration and enzymatic activity and estimating microbial densities by culturing and direct cel counts (Witkamp, 1963; Suberkrop and Klug, 1976; Walace et al., 1982). These traditional aproaches allowed direct quantification of microbes and made possible examination of leaf species diferences in microbial density at single- and multiple-points in time during breakdown. However, direct cel counts provide no information about cel viability, whereas use of microscopical and staining techniques alowed a beter visualization of microrganisms within the liter substrate. For example, use of 4',6-diamidino-2-phenylindole (DAPI) and 5- cyano-2,3-ditolyltetrazolium chloride (CTC) stains coupled with fluorescence microscopy alows viable and nonviable cels to be distinguished, and thus provides a more accurate estimate of cel densities and biomass in liter (Simon, 200). Historicaly, determining identity of fungal and bacterial taxa has relied on culturing techniques to grow the microorganisms present coupled with tests to characterize and identify them (Suberkrop and Klug, 1976). Within the past 10 y researchers have recognized that ~9% of microorganisms have complex growth requirements and unpredictable nutritional demands that render them incapable of culture (Aman et al., 195). This limitation has led to developing culture- independent techniques for analyzing microbial comunity structure and composition. These culture-independent methods alow for examination of microbial comunity structure and composition at various levels of resolution, ! 10 including from DNA sequencing to whole comunity similarity/dissimilarity (Findlay, 2010). In a recent review of stream icrobial ecology, Findlay (2010) described the varying levels of resolution provided by a ful suite of methods for quantifying microbial comunity structure and composition. For broad levels, such as shifts in microbial comunity structure, DNA fingerprinting techniques including denaturing gradient gel electrophoresis (DGE), temperature gradient gel electrophoresis (TGE), ribosomal intergenic spacer analysis (RISA), and terminal restriction-fragment length polymorphism (tRFLP) provide genetic profile-based information about comunity similarity paterns. Most of these methods provide minimal to no sequence information regarding the taxa present. The major benefit of these fingerprinting techniques is to characterize broad scale shifts in comunity structure without specific prior knowledge of taxa. Many studies have used these modern techniques alongside traditional methods for assessing microbial diversity and activity. This combined aproach has helped adress questions about seasonal, leaf species, and environmental efects on microbial comunities in many aquatic habitats (Das et al., 206; Hular et al. 2006; Res et al., 206). Relative to the above methods, sequencing of smal ribosomal subunits (such as the 16S rDNA gene in bacteria or the 18S rDNA gene in fungi) has provided even finer resolution of taxa composition in microbial comunities (Hular et al., 206). Using methods including cloning as wel as next-generation sequencing, researchers can now sequence and identify al members within a ! 11 microbial comunity, from species- to division-level, with the degree of confidence in taxonomic assignment increasing with sequence length (Huse et al., 207; Womack, Bhavsar and Ravel, 208). In this context, Findlay (2010) pointed out that division- or taxa-specific probes can be designed to screen for microbes within samples and also estimate their relative proportions. By combining traditional and contemporary (molecular-based) techniques, there is high potential to adress questions regarding microbial comunity succession and response to contrasting environmental conditions that heretofore have not ben possible. Many of these above techniques are based on polymerase chain reaction (PCR) amplification of the environmental sample. Here, deoxyribonucleic acid (DNA) is directly extracted, with a specific portion of it being amplified with a specific primer pair. Depending on the sample, there may be substances co- extracted along with DNA that can inhibit further enzymatic-dependent downstream processes. The polymerase enzyme involved in PCR often can be inhibited from the presence of polyphenols, humic acids, and polysaccharides; each of these substances can occur in degrading leaf liter. In addition, many studies involving genomic DNA extraction from environmental samples comonly use DNA extraction kits (Rose-Amsaleg et al., 201). Several of these often involve alcoholic precipitation of DNA (Porteous et al., 1997) and can co-precipitate humic substances, which have ben shown to be highly inhibitory towards polymerase enzymes (Tebe and Vahjen, 193). ! 12 Genomic DNA extracted from liter usualy requires further purification prior to PCR because of humic and phenolic substances and other low- molecular-weight contaminants (Widmer et al., 196; Widmer et al., 197). Presence of these co-extracted contaminants in extracted genomic DNA requires sample DNA purification folowing extraction. Several purification methods exist in the literature (reviewed by Rose-Amsaleg et al., 201). Some of these methods involve adition of products, such as polyvinylpolypyrrolidone (PVPP), sodium ascorbate and hexadecyltrimethylamonium bromide (CTAB), during cel lysis that bind to humic substances, thus reducing their impact on downstream processes (Rose-Amsaleg et al., 201). Often, folowing these treatments, it is stil necessary to dilute genomic DNA and its co-extracted contaminants to an optimum that reduces contaminant concentration to a non-inhibitory level. However, in many instances these products and subsequent dilution are not suficient and further purification is necessary. Some comonly used purification strategies include cesium chloride (CsCl) density gradient centrifugation, chromatography, electrophoresis, and dialysis and filtration. Samples high in humic substances, such as soils and organic mater-rich samples, often are electrophoresed in low-melt agarose so that faster migrating humic substances wil separate from genomic DNA (for review, see Rose-Amsaleg et al., 201). This separation of contaminant and genomic DNA alows for the genomic DNA to be extracted from the agarose and used in subsequent analyses. Many of the above methods result in a decrease in DNA yield, which can reduce eficacy of downstream processes, including products of PCR (Lear et al., ! 13 2010). In adition, a single purification does not completely remove al contaminants, which then may require use of two (or more) purification methods. Using a combination of purification methods even further increases potential loss of DNA and its concentration. Selection of a purification method is often based on balancing the conflicting demands of increased processing time and reduced DNA yield (Rose-Amsaleg et al., 201). Thus, developing new PCR-based techniques capable of higher resolution detection is necessary to further purify extracted genomic DNA without the concomitant loss of DNA product. In addition to the role of microbial comunities in litter breakdown in streams, several other factors, including abiotic and biotic factors, can influence breakdown rate. Concentrations of dissolved nutrients, such as N and P, can alter leaf breakdown rates. Laboratory stream icrocosm studies have shown NO 3 -N and PO 4 aditions stimulate microbial activity, and increase leaf fragmentation and breakdown rate (Howarth and Fisher, 1976). In situ studies have also shown strong efects of nutrients on liter decomposition. Streamwater P and N have ben positively correlated with increased decomposer activity and leaf breakdown (Elwod et al. 1981; Suberkrop and Chauvet, 195; Gulis and Suberkrop, 203). Increased concentrations of dissolved N and P potentialy stimulate microbial activity and increase biomass, which, in turn, increases breakdown (Suberkrop and Chauvet, 195; Gulis and Suberkrop, 203). Breakdown also can be influenced by streamwater pH, often being higher in circumneutral (vs. acidic) stream water, where it is thought that low pH inhibits microbial growth (Webster and Benfield, 1986; Ripinen, et al., 209). ! 14 High streamwater temperature has historicaly ben associated with increased breakdown (Webster and Benfield, 1986), although this efect may be because of higher stream etabolism (e.g. gross primary production [GPP] and ecosystem respiration) (Irons, et al., 194; Young et al., 208). Increased temperature typicaly leads to increased GPP, which provides energy for primary consumers (Larsson and Hagstrom, 1979) and increased microbial respiration. Dissolved oxygen concentration within stream water may exacerbate the efects of temperature on metabolism, which if low, may decrease breakdown. Marmonier et al. (2010) found lower breakdown rates in sites with decreased dissolved oxygen; they hypothesized that lower breakdown rates in their sites could be explained by decreased oxygen availability that suports heterotrophic activity and alters invertebrate assemblage composition. Aside from the above physicochemical conditions, inherent chemical and structural properties of the leaf litter itself also can afect breakdown. Melillo et al. (1983) found leaf breakdown to be slower in liter with lower N content, which was inversely related to the initial lignin content. In general, high concentrations of refractory structural components such as lignin have ben considered the primary factors determining breakdown (Gessner and Chauvet, 194; reviewed by Gessner, 205). The efect of high lignin:N ratios, as wel as high C:N and C:P, in decreasing leaf breakdown was also reported (Enriquez et al., 193). Structural components of liter, such as lignin, celulose, and hemicelulose, often are indigestible for stream acroinvertebrates who lack the apropriate digestive enzymes (Wright and Covich, 205). For this reason, liter ! 15 high in structural components, particularly lignin, is generaly unpalatable and of low quality to macroinvertebrates (Cumins and Klug, 1979). For leaf liter high in lignin, a large portion of the initial breakdown is from stream icroorganisms that produce enzymes, such as celulases, capable of lignin degradation (Pusch et al., 198). Microbial conditioning increases liter palatability, and accumulation of microbial biomass increases nutritional quality of liter alowing further fragmentation from acroinvertebrates (Wright and Covich, 205; reviewed by Graca and Zimer, 205). Secondary plant compounds, including polyphenolics such as tanins, also may decrease breakdown. Studies have examined if this decrease occurs by colonization inhibition of stream acroinvertebrates and/or microbes. Canhoto and Graca (195 and 199) found that leaf consumption by the cranefly Tipula lateralis was negatively correlated with polyphenolic content. They also noted that those T. lateralis consuming leaves high in polyphenolic content, including eucalyptus and oak, did not grow. Macroinvertebrates may avoid consuming liter high in polyphenolics as part of their life history strategy to enhance growth. Canhoto and Graca (199) showed that the inhibitory efects of secondary compounds on macroinvertebrate feding could be transferred from eucalyptus leaves to fast-degrading leaf species. They also showed that aquatic hyphomycete growth decreased with increasing concentrations of the secondary compounds eucalyptus oil and tanic acid (Canhoto and Graca 199). Previous studies also have sugested that secondary compounds inhibit microbial colonization of liter (Stout 1989), although this efect does not seem to occur in ! 16 the tropics. In a Costa Rican stream Ardon and Pringle (208) found that secondary compounds were rapidly leached from 8 leaf species and were believed to be less important than structural compounds in determining breakdown. In general, few studies have examined the efects of leaf species (and their associated structural and secondary components) on microbial comunity composition. Recently, Das et al. (207) compared overal diversity of fungi, bacteria, and actinomycetes on 2 leaf species (sugar maple and white oak), and found time of exposure to be the major factor controling microbial comunity composition. However, they used only DGE banding paterns to characterize the comunities, which unfortunately alows for co-migration of DNA fragments from diferent taxa to the same position on a gel, and they obtained no sequence information on dominant taxa. More definitive sequence-based research neds to be done to examine diferences in microbial comunity composition in response to leaf chemistry diferences at a finer level of taxonomic resolution. Having finer-scale resolution of the specific taxa present throughout the breakdown process would alow researchers to explore whether specific microbial taxa often were associated with faster- or slower-degrading species. In adition, it would alow for a direct comparison with previous studies describing the dominant taxa associated with degrading liter. As mentioned previously, feeding and fragmentation of instream liter by shreders and other macroinvertebrate consumers can afect leaf breakdown (Barnes et al., 1986; McArthur and Barnes, 198; Walace and Webster, 196). ! 17 Whole-stream insecticide treatment at the Coweta Hydrologic Laboratory, North Carolina, USA provided strong evidence of the role of macroinvertebrates (Cufney et al. 1990). Folowing a drastic reduction (>1,00,00 organisms/wk) of macroinvertebrates, up to a 74% reduction in breakdown occurred in the treated sections with no efect on bacterial density or microbial respiration. Whole-stream experiments such as this are usualy not feasible, so most of the evidence directly linking macroinvertebrates and breakdown has ben done on smaler spatial scales. More comonly, researchers have manipulated macroinvertebrate abundance by excluding them from liter in mesh bags of diferent mesh sizes. Generaly, a coarse mesh (~ 5.0m) is used as a control to alow for macroinvertebrate colonization, whereas a finer mesh (~1.0m) serves as the exclusion treatment (Boulton and Boon, 191). For example, Stewart (192) used mesh exclusions to assess the efect of macroinvertebrates on decomposition rates for several leaf species bordering woodland streams of southern Africa. In that study, macroinvertebrate efects on breakdown varied with shreder density, which varied among sites. At the site with the highest shreder density, Stewart (192) observed breakdown rates of al leaf species examined were significantly faster in coarse (control) than fine (exclusion) mesh. Macroinvertebrate efects on breakdown were not present at sites with lower shreder densities (Stewart, 1992). Benfield and Webster (1985) found a similar result in an Appalachian stream where species-specific leaf breakdown rates varied with shreder abundance. More recently, the presence of shreders has ! 18 been implicated as a critical factor influencing leaf processing in streams (Sponseler and Benfield, 201). Studies have examined the contribution and comunity composition of stream icrobes and macroinvertebrates associated with leaf liter, although to date no studies have ben designed to quantify the efects of macroinvertebrates on the microbial comunities associated with liter in streams. Disturbance plays a major role in stream comunity organization, whose efects can be both direct and indirect (Resh et al., 198; Maloney and Weler, 201). Disturbance is generaly defined as ?any relatively discrete event in time that is characterized by a frequency, intensity, and severity outside of predictable range, and that disrupts ecosystem, comunity, or population structure and changes resources or the physical environment? (Resh et al., 198). In stream ecosystems, disturbance can be natural or anthropogenic. Major examples of natural disturbance in streams are typicaly related to hydrologic regime, such as flods and droughts (Lake, 200). Alternatively, anthropogenic disturbance is defined as ?any human-mediated event or activity that is virtualy unknown in natural systems in terms of type, frequency, intensity, duration, spatial extent, or predictability over the last century? (Naiman et al., 205). Anthropogenic disturbance can result from any land use activities including, but not limited to, acid mining, agricultural practices, timber harvesting, and urbanization (Allan, 204). According to the US Census Bureau, the US population is expected to increase by ~ 21% over the next 2 decades (U.S. Census, 208). Increased land use is often associated with increased human population growth (Allan, Erickson ! 19 and Fay, 197). Thus, increases in human population size of this magnitude are likely to increase occurrence of human-mediated disturbance in many stream ecosystems. Human-mediated changes to the landscape within watersheds can alter instream hydrology and geomorphology, causing altered flow regime, chanel shape, and increased sediment erosion and deposition within the chanel (reviewed by Allan, 204). Increased sedimentation can alter or degrade many instream variables, including stream habitat and the associated benthic food web (Henley et al., 2000; Allan, 204; Downes et al., 2006;). By altering the stream bed, sedimentation may lead to decreased benthic habitat heterogeneity by infiling of interstitial spaces (Lake, 200; Allan, 204). Habitat heterogeneity (complexity) has ben investigated in several studies for its role in structuring the benthic comunity. Habitat alteration can limit those organisms capable of colonizing a given location (Pof, 197). Sedimentation has ben shown to significantly influence species richness and diversity in aquatic insect assemblages (Lemly, 1982), and the overal structure of a benthic habitat can significantly afect species diversity and abundance (Downes et al., 198). As such, disturbance can lead to a homogenized stream reach colonized primarily by relatively tolerant biota that are more suited to altered habitat than less tolerant species (Helms et al., 209). Overal, sediment disturbance has a high potential to yield an altered, more homogenous, and less diverse biotic comunity (Harrison, 207). ! 20 Sediment disturbance can afect aquatic invertebrate population size and comunity structure in multiple ways. Aquatic invertebrate density and diversity are often directly related to substrate diversity (Gore, 1985). The efects of sediment on habitat heterogeneity can reduce habitat availability for some invertebrates and also increase their susceptibilty to predation (Newcombe and Macdonald, 191). Beyond its influence on habitat availability, sediment also can affect invertebrate functional feding group composition and abundance. For example, because of its efects on primary producer biomass and composition, sediment can alter abundance of secondary consumers, such as invertebrate grazers (Newcombe and Macdonald, 191). Sedimentation can also afect invertebrate colector-filters, by clogging feding structures, or even gils of non- filter feders, leading to reduction in feding eficiency and growth and increased mortality (Hynes, 1970; Lemly, 1982). In sandy coastal plains streams anthropogenic disturbance also can lead to impaired stream etabolism through reduced respiration and primary production (Houser et al., 205). These reductions separately or in combination can afect leaf breakdown rates by altering the rates at which leaf conditioning, fragmentation, and/or macroinvertebrate consumption occur. Instream environmental measures, such as those used to assess water quality, are greatly afected by sedimentation. Streamwater temperature, turbidity, and dissolved oxygen available in stream water are among the main notable factors afected (reviewed by Ryan, 191). Increased turbidity, along with an overal increase in suspended sediment load, is a comon result of increased ! 21 sedimentation (Lemly, 1982). Elevated water temperature often is associated with high sedimentation, where radiation-stored sediment erodes into a stream chanel increasing the water temperature (Hagans et al., 1986). As water temperature increases, there is also an associated decrease in the concentration of dissolved oxygen. Much research has ben done to elucidate the response of macroinvertebrates to instream sedimentation, but comparatively litle research has ben conducted on the efects of sedimentation on the microbial comunity associated with leaf liter. Given that microbial conditioning must occur before macroinvertebrates generaly colonize and consume liter, it is important to examine the degree to which sedimentation afects microbial comunities associated with instream litter. The efects of sedimentation on leaf breakdown have ben equivocal. Benfield et al. (201) found increased sedimentation imediately folowing the start of loging decreased leaf breakdown. Other studies concluded that decreased breakdown from sedimentation occurred because of burial of liter, which created an anoxic environment preventing fungal and macroinvertebrate colonization (Webster and Waide, 1982). In other situations, increased breakdown rates after loging, may cause increased stream flow and erosion of finer sediment, leaving behind coarser sediment that increased liter abrasion and fragmentation (Benfield et al. 201). In a related study, Sponseler and Benfield (201) found leaf breakdown rate was positively correlated with substrate particle size. In this context, depending on substrate particle size, sediment can either bury liter and reduce breakdown, or increase abrasion and ! 22 accelerate breakdown. Aside from burial impacts, sedimentation and its subsequent efects on instream habitat, as mentioned above, can reduce secondary consumer composition, which also can in turn reduce breakdown rates (Sponseler and Benfield, 201). These interactive efects betwen physical factors (e.g. sedimentation) and stream organisms can lead to drastic alterations in leaf breakdown and energy release into stream fod webs. However, litle is known regarding their efects on leaf liter-associated microbial comunity composition. Knowing how sediment disturbance afects microbial comunity composition of liter could provide further insight into the mechanism by which sedimentation alters leaf breakdown. B. SUMARY Allochthonous leaf liter provides a fundamentaly important source of nutrients and energy inputs for stream ecosystems. Therefore, litter breakdown and concomitant release of these inputs from terrestrial sources is vital to fueling stream ecosystems. Typicaly, leaf breakdown studies involve incubation of leaf liter in situ and monitoring changes over time both in amount and quality of remaining leaf litter. Leaf breakdown, whether viewed as a stepwise process or overlaping phases, involves 3 processes: 1) leaching, 2) microbial conditioning, and 3) fragmentation. Of these, microbial conditioning is particularly important because of the dual role of microbes in nutrient transformation and in increasing litter quality and preference for macroinvertebrates. My dissertation focuses on investigating several aspects of leaf breakdown, in particular processes afecting microbial conditioning processes. ! 23 Because of data indicating a larger contribution by fungi than bacteria during breakdown, several studies have examined the specific fungal taxa involved in leaf breakdown. Less work has ben done studying bacterial taxa present during breakdown. Recent advances in methodology alow for the examination of microbial comunities at varying levels of resolution, ranging from very broad levels (e.g., fingerprinting techniques) to finer levels of resolution (e.g., next-generation sequencing). My dissertation decribes a pluralistic aproach using traditional and modern molecular methods to characterize the stream icrobial comunity associated with leaf liter breakdown, particularly in reference to how comunities vary temporaly and in response to diferences in macroinvertebrate shreder abundances and leaf chemistry. Many of these techniques employ the use of Polymerase Chain Reaction (PCR) making them susceptible to inhibition from co-extracted enzymatic inhibitors. Several purification methods exist, but vary in eficiency based on sample source, cost, processing time, and purified genomic DNA yield. I developed and tested a low- cost, rapid method to further purify extracted genomic DNA that causes minimal concomitant loss of genomic DNA concentration. Although biotic factors such as leaf liter chemistry and macroinvertebrates have ben shown to afect leaf breakdown, few studies have examined the degree to which these consumers afect the associated microbial comunity. Both natural and anthropogenic (human-mediated) disturbance can drasticaly afect stream ecosystems, and with projected increases in the human population, anthropogenic disturbance is likely to greatly increase in the near future. One ! 24 form of anthropogenic disturbance of particular importance in stream ecosystems is land-derived sedimentation, which can afect both instream habitat and associated stream organisms. Sedimentation can both increase and decrease rate of breakdown, likely because of the varied efects of substrate particle size, and can also alter secondary consumer composition. Yet, surprisingly litle is known about how sedimentation afects leaf liter-associated microbial comunity composition. My dissertation research is separated into 4 main chapters, with the first data chapter (Chapter I) describing a method for purifying genomic DNA extracted from environmental sources for later use in PCR. Chapter II focuses on the efects of stream acroinvertebrates on leaf liter microbial comunities in situ by reducing macroinvertebrate shreder abundance within leaf liter during breakdown. Chapter IV examines the efects of diferent leaf species on liter breakdown and how this efect alters microbial succession on litter. The last chapter (Chapter V) describes an additional leaf breakdown study designed to examine the efect of sediment disturbance on leaf liter microbial comunities. ! 25 C. LITERATURE CITED Abelho, M. 201. From literfal to breakdown in streams: a review. The Scientific World Journal, 1, 656-680. Allan, J.D. 1995. Stream ecology: structure and function of runing waters. Chapman and Hal. London, UK. Allan, J.D. 204. Landscapes and riverscapes: the influence of land use on stream ecosystems. Annual Review of Ecology, Evolution, and Systematics, 35, 257-284. Allan, D., Erickson, D. and Fay, J. 197. The influence of catchment land use on stream integrity acros multiple spatial scales. Freshwater Biology, 37, 149-161. Aman, R.I., Ludwig, W. and Schleifer, K.H. 195. Phylogenetic identification and in situ detection of individual microbial cels without cultivation. Microbiology and Molecular Biology Reviews, 59, 143-169. Ardon, M. and Pringle, C.M. 208. Do secondary compounds inhibit microbial- and insect-mediated leaf breakdown in a tropical rainforest stream, Costa Rica? Oecologia, 155, 311-323. Arsufi, T.L. and Suberkrop, K. 1989. Selective feding by shreders on leaf- colonizing stream fungi: comparison of macroinvertebrate taxa. Oecologia, 79, 30-37. Baldy, V., Gessner, M.O. and Chauvet, E. 195. Bacteria, fungi and the breakdown of leaf liter in a large river. Oikos, 74, 93-102. ! 26 Barnes, J.R., McArthur, J.V. and Cushing, C.E. 1986. Effect of excluding shreders on leaf liter decomposition in two streams. Western North American Naturalist, 46, 204-207. Benfield, E.F. and Webster, J.R. 1985. Shreder abundance and leaf breakdown in an Appalachian Mountain stream. Freshwater Biology, 15, 113-120. Boulton, A.J. and Boon, P.I. 191. A review of methodology used to measure leaf liter decomposition in lotic environments: time to turn over an old leaf? Marine and Freshwater Research, 42, 1-43. Bournaud, M., Tachet, H. and Roux, A.L. 1987. The efects of seasonal and hydrological influences on the macroinvertebrates of the Rhone River, France. I. Ecological aspects. Archiv f?r Hydrobiologie, Suplement, 76, 25-51. Canhoto, C. and Graca, M.A.S. 195. Fod value of introduced eucalypt leaves for a Mediterranean stream detritivore: Tipula lateralis. Freshwater Biology, 34, 209-214. Canhoto, C. and Gra?a, M.A.S. 196. Decomposition of Eucalyptus globulus leaves and three native leaf species Alnus glutinosa, Castanea sativa and Quercus faginea in a Portuguese low order stream. Hydrobiologia, 333, 79-85. Canhoto, C. and Gra?a, M.A.S. 199. Leaf barriers to fungal colonization and shreders Tipula lateralis. consumption of decomposing Eucalyptus globulus. Microbial Ecology, 37, 163-172. ! 27 Cargil, A.S., Cumins, K.W., Hanson, B.J. and Lowry, R.. 1985. The role of lipids, fungi, and temperature in the nutrition of a shreder cadisfly, Clistoronia magnifica. Freshwater Invertebrate Biology, 4, 64-78. Coleman, M.J. and Hynes, H.B.N. 1970. The vertical distribution of the invertebrate fauna in the bed of a stream. Limnology and Oceanography, 15, 31-40. Cufney, T.F., Walace, J.B. and Lugthart, G. 190. Experimental evidence quantifying the role of benthic invertebrates in organic mater dynamics of headwater streams. Freshwater Biology, 23, 281-299. Cumins, K.W. 1973. Trophic relations of aquatic insects. Annual Review of Entomology, 18, 183-206. Cumins, K.W. 1974. Structure and function of stream ecosystems. BioScience, 24, 631-641. Cumins, K.W. and Klug, M.J. 1979. Feding ecology of stream invertebrates. Annual Review of Ecology and Systematics, 10, 147-172. Cumins, K.W., Spengler, G.L., Ward, G.M., Speaker, R.M., Ovink, R.W., Mahan, D.C. and Matingly, R.L. 1980. Processing of confined and naturaly entrained leaf liter in a wodland stream ecosystem. Limnology and Oceanography, 25, 952-957. Das, M., Royer, T.V. and Leff, L.G. 206. Fungal, bacterial and actinomycete diversity on leaves decomposing in a stream. Applied and Environmental Microbiology, 73, 756-767. ! 28 Downes, B.J., Lake, P.S., Glaister, A. and Bond, N.R. 206. Effects of sand sedimentation on the macroinvertebrate fauna of lowland streams: are the efects consistent? Freshwater Biology, 51, 144-160. Downes, B.J., Lake, P.S., Schreiber, E.S.G. and Glaister, A. 198. Habitat structure and regulation of local species diversity in a stony, upland stream. Ecological Monographs, 68, 237-257. Duong, L.M., Jeewon, R., Lumyong, S. and Hyde, K.D. 206. DGE coupled with ribosomal DNA gene phylogenies reveal uncharacterized fungal phylotypes. Fungal Diversity, 23, 121-138. Elwod, J.W., Newbold, J.D., Trimble, A.F. and Stark, R.W. 1981. The limiting role of phosphorus in a wodland stream ecosystem: efects of P enrichment on leaf decomposition and primary producers. Ecology, 62, 146-158. Enriquez, S., Duarte, C.M. and Sand-Jensen, K. 193. Paterns in decomposition rates among photosynthetic organisms: the importance of detritus C: N: P content. Oecologia, 94, 457-471. Findlay, S. 2010. Stream icrobial ecology. Journal of the North American Benthological Society, 29, 170-181. Fisher, S.G. 1971. Annual energy budget of a smal forest stream ecosystem: Bear Brook, West Thornton, New Hampshire, Ph.D. Thesis. Dartmouth Colege, Hanover, New Hampshire. 97 p. ! 29 Fisher, S.G. and Likens, G.E. 1973. Energy flow in Bear Brook, New Hampshire: an integrative aproach to stream ecosystem etabolism. Ecological Monographs, 43, 421-439. Gesner, M. 205. Proximate lignin and celulose. In: Methods to study liter decomposition: a practical guide. (Eds. F. Barlocher, Graca, M.A.S., Gesner, M.O.), p. 15-120. Kluwer Academic Pub, Netherlands. Gessner, M.O. 197. Fungal biomass, production and sporulation associated with particulate organic mater in streams. Limnetica, 13, 33-44. Gessner, M.O. and Chauvet, E. 194. Importance of stream icrofungi in controling breakdown rates of leaf liter. Ecology, 75, 1807-1817. Gessner, M.O., Chauvet, E. and Dobson, M. 199. A perspective on leaf liter breakdown in streams. Oikos, 85, 377-384. Gore, J.A. 1985. Mechanisms of colonization and habitat enhancement for benthic macroinvertebrates in restored river chanels. The Restoration of Rivers and Streams: Theories and Experience Ed. J.A. Gore, p. 81- 101. Buterworth Publishers, Boston, Massachusets. Graca, M. and Zimer, M. 205. Leaf toughnes. In: Methods to Study Liter Decomposition: a practical guide. (Eds. F. Barlocher, Graca, M.A.S., Gesner, M.O.), p. 121-125. Kluwer Academic Pub, Netherlands. Graca, M.A.S. 201. The role of invertebrates on leaf liter decomposition in streams: a review. International Review of Hydrobiology, 86, 383-393. ! 30 Gulis, V. and Suberkrop, K. 203. Interactions betwen stream fungi and bacteria associated with decomposing leaf liter at diferent levels of nutrient availability. Aquatic Microbial Ecology, 30, 149-157. Hagans, D.K., Weaver, W.E. and Madej, M.A. 1986. Long-term on-site and of- site efects of loging and erosion in the Redwod Creek Basin, northern California. p. 38-65 in Papers presented at the American Geophysical Union meting on cumulative efects. Technical buletin 490. National Council for Air and Stream Improvement, New York. Harrison, E.T. 207. The impact of fine sediment accumulation on benthic macroinvertebrates: implications for river management. In: Proceedings of the 5th Australian Stream Management Conference. Helms, B. S., Schonover J. E., and Feminela J. W. 209. Seasonal variability of landuse impacts on stream acroinvertebrate assemblages in streams of western Georgia, USA. Journal of the North American Benthological Society, 28, 991-1006. Henley, W.F., Paterson, M.A., Neves, R.J. and Lemly, A.D. 200. Effects of sedimentation and turbidity on lotic fod webs: a concise review for natural resource managers. Reviews in Fisheries Science, 8, 125-139. Hieber, M. and Gessner, M.O. 202. Contribution of stream detrivores, fungi, and bacteria to leaf breakdown based on biomass estimates. Ecology, 83, 1026-1038. Houser, J.N., Mulholand, P.J. and Maloney, K.O. 205. Catchment disturbance and stream etabolism: paterns in ecosystem respiration and gross ! 31 primary production along a gradient of upland soil and vegetation disturbance. Journal of the North American Benthological Society, 24, 538-552. Howarth, R.W. and Fisher, S.G. 1976. Carbon, nitrogen, and phosphorus dynamics during leaf decay in nutrient enriched stream microecosystems. Freshwater Biology, 6, 221-228. Hular, M.A.J., Kaplan, L.A. and Stahl, D.A. 206. Recurring seasonal dynamics of microbial comunities in stream habitats. Applied and Environmental Microbiology, 72, 713. Huse, S.M., Huber, J.A., Morrison, H.G., Sogin, M.L. and Welch, D.M. 2007. Accuracy and quality of massively paralel DNA pyrosequencing. Genome Biology, 8, R143. Irons, J.G., Oswod, M.W., Stout, R. and Pringle, C.M. 194. Latitudinal paterns in leaf liter breakdown: is temperature realy important? Freshwater Biology, 32, 401-411. Jansen, P.H., Yates, P.S., Grinton, B.E., Taylor, P.M. and Sait, M. 202. Improved Culturability of Soil Bacteria and Isolation in Pure Culture of Novel Members of the Divisions Acidobacteria, Actinobacteria, Proteobacteria, and Verrucomicrobia. Appl. Environ. Microbiol., 68, 2391- 2396. Kaushik, N.K. and Hynes, H.B.N. 1971. The fate of the dead leaves that fal into streams. Archiv f?r Hydrobiologie, 68, 465-515. ! 32 Lake, P.S. 200. Disturbance, patchiness, and diversity in streams. Journal of the North American Benthological Society, 19, 573-592. Larsson, U. and Hagstrom, A. 1979. Phytoplankton exudate release as an energy source for the growth of pelagic bacteria. Marine Biology, 52, 199-206. Lear, G., Dong, Y. and Lewis, G. 2010. Comparison of methods for the extraction of DNA from stream epilithic biofilms. Antonie van Leuwenhoek, 98, 567-571. Lemly, A.D. 1982. Modification of benthic insect comunities in poluted streams: combined efects of sedimentation and nutrient enrichment. Hydrobiologia, 87, 229-245. Maier, R.M., Peper, I.L. and Gerba, C.P. 209. Environmental microbiology, Academic Press. Maloney, D.C. and Lamberti, G.A. 195. Rapid decomposition of sumer-input leaves in a northern Michigan stream. American Midland Naturalist, 133, 184-195. Maloney, K.O. and Weler, D.E. 201. Anthropogenic disturbance and streams: land use and land-use change afect stream ecosystems via multiple pathways. Freshwater Biology, 56, 611-626. Marmonier, P., Piscart, C., Sarriquet, P.E., Azam, D. and Chauvet, E. 2010. Relevance of large liter bag burial for the study of leaf breakdown in the hyporheic zone. Hydrobiologia, 641, 203-214. ! 33 McArthur, J.V. and Barnes, J.R. 198. Comunity dynamics of leaf liter breakdown in a Utah alpine stream. Journal of the North American Benthological Society, 7, 37-43. Melilo, J.M., Naiman, R.J., Aber, J.D. and Eshleman, K.N. 1983. The influence of substrate quality and stream size on wod decomposition dynamics. Oecologia, 58, 281-285. Mille-Lindblom, C. and Tranvik, L.J. 203. Antagonism betwen bacteria and fungi on decomposing aquatic plant liter. Microbial Ecology, 45, 173- 182. Minshal, G.W. 1967. Role of alochthonous detritus in the trophic structure of a wodland springbrook comunity. Ecology, 48, 139-149. Moriarty, D.J.W. and Pulin, R.S.V. (1987) Detritus and microbial ecology in aquaculture, International Center for Living Aquatic Resources Management, Manilla, Philippines. Naiman, R.J., Decamps, H. and Mcclain, M.E. 205. Riparia: ecology, conservation, and management of streamside comunities, Academic Press, Burlington, MA. Newcombe, C.P. and Macdonald, D.. 191. Effects of suspended sediments on aquatic ecosystems. North American Journal of Fisheries Management, 11, 72 - 82. Nikolcheva, L.G. and Barlocher, F. 204. Taxon-specific fungal primers reveal unexpectedly high diversity during leaf decomposition in a stream. Mycological Progress, 3, 41-49. ! 34 Nykvist, N. 1961. Leaching and decomposition of liter. II. Experiments on leaf liter of Betula verrucosa. Oikos, 12, 249-263. Nykvist, N. 1963. Leaching and decomposition of water-soluble organic substances from diferent types of leaf and nedle liter. Studio Forestalia Suecica, 3, 1-29. Palmer, M.A. and Pof, N.L. 197. The influence of environmental heterogeneity on paterns and processes in streams. Journal of the North American Benthological Society, 16, 169-173. Pereira, A.P., Graca, M.A.S., and Moles, M. 198. Leaf liter decomposition in relation to liter physico-chemical properties, fungal biomass, arthropod colonization, and geographical origin of plant species. Pedobiologia, 42, 316-327. Petersen, R.C. and Cumins, K.W. 1974. Leaf processing in a wodland stream. Freshwater Biology, 4, 343-368. Porteous, L.A., Seidler, R.J. and Watrud, L.S. 197. An improved method for purifying DNA from soil for polymerase chain reaction amplification and molecular ecology aplications. Molecular Ecology, 6, 787-791. Pusch, M., Fiebig, D., Bretar, I., Eisenman, H., Ellis, B.K., Kaplan, L.A., Lock, M.A., Naegeli, M.W. and Traunspurger, W. 198. The role of micro- organisms in the ecological conectivity of runing waters. Freshwater Biology, 40, 453-495. Res, G.N., Watson, G.O., Baldwin, D.S. and Mitchel, A.M. 206. Variability in sediment microbial comunities in a semipermanent stream: impact of ! 35 drought. Journal of the North American Benthological Society. 25, 370- 378. Resh, V.H., Brown, A.V., Covich, A.P., Gurtz, M.E., Li, H.W., Minshal, G.W., Reice, S.R., Sheldon, A.L., Walace, J.B. and Wissmar, R.C. 1988. The role of disturbance in stream ecology. Journal of the North American Benthological Society, 7, 433-455. Ripinen, M.P., Davy Bowker, J. and Dobson, M. 209. Comparison of structural and functional stream assessment methods to detect changes in riparian vegetation and water pH. Freshwater Biology, 54, 2127-2138. Romani, A.M., Fischer, H., Mile-Lindblom, C. and Tranvik, L.J. 206. Interactions of bacteria and fungi on decomposing liter: diferential extracelular enzyme activities. Ecology, 87, 2559-2569. Rose-Amsaleg, C.L., Garnier-Silam, E. and Harry, M. 201. Extraction and purification of microbial DNA from soil and sediment samples. Applied Soil Ecology, 18, 47-60. Ryan, P.A. 191. Environmental efects of sediment on New Zealand streams: a review. New Zealand Journal of Marine and Freshwater Research, 25, 207-221. Simon, K.S. 200. Organic Mater Dynamics and Trophic Structure in Karst Groundwater. Ph.D. Thesis, Virginia Polytechnic Institute and State University, Blacksburg, VA. ! 36 Sinsabaugh, R.L., Antibus, R.K. and Linkins, A.E. 191. An enzymic aproach to the analysis of microbial activity during plant liter decomposition. Agriculture, Ecosystems and Environment, 34, 43-54. Sinsabaugh, R.L. and Morhead, D.L. 194. Resource alocation to extracelular enzyme production: a model for nitrogen and phosphorus control of liter decomposition. Soil Biology and Biochemistry, 26, 1305-1311. Sparrow, F.K. 1968. Ecology of freshwater fungi. In: The fungal population, the fungi, an advanced treatise. Eds G.C. Ainsworth and A.S. Sussman., p. 41-93. Academic Press, New York. Sponseler, R.A. and Benfield, E.F. 201. Influences of land use on leaf breakdown in southern Appalachian headwater streams: a multiple-scale analysis. Journal of the North American Benthological Society, 20, 44- 59. Sponseler, R.A., Benfield, E.F. and Valet, H.M. 201. Relationships betwen land use, spatial scale and stream acroinvertebrate comunities. Freshwater Biology, 46, 1409-1424. Stewart, B.A. 192. The efect of invertebrates on leaf decomposition rates in two smal wodland streams in southern Africa. Archiv f?r Hydrobiologie. Stutgart, 124, 19-33. Stout, R.J. 1989. Effects of condensed tanins of leaf processing in mid-latitude and tropical streams: a theoretical aproach. Canadian Journal of Fisheries and Aquatic Sciences, 46, 1097-1106. ! 37 Suberkrop, K., Arsufi, T.L. and Anderson, J.P. 1983. Comparison of degradative ability, enzymatic activity, and palatability of aquatic hyphomycetes grown on leaf liter. Applied and Environmental Microbiology, 46, 237-244. Suberkrop, K. and Chauvet, E. 195. Regulation of leaf breakdown by fungi in streams: influences of water chemistry. Ecology, 76, 1433-1445. Suberkrop, K., Godshalk, G.L. and Klug, M.J. 1976. Changes in the chemical composition of leaves during processing in a wodland stream. Ecology, 57, 720-727. Suberkrop, K. and Klug, M.J. 1976. Fungi and bacteria associated with leaves during processing in a wodland stream. Ecology, 57, 707-719. Suberkrop, K. and Klug, M.J. 1980. The maceration of deciduous leaf liter by aquatic hyphomycetes. Canadian Journal of Botany, 58, 1025-1031. Tate, C.M. and Gurtz, M.E. 1986. Comparison of mass loss, nutrients, and invertebrates associated with elm leaf liter decomposition in perenial and intermitent reaches of talgrass prairie streams. The Southwestern Naturalist, 31, 511-520. Taylor, B.R. and Barlocher, F. 196. Variable efects of air-drying on leaching losses from tree leaf liter. Hydrobiologia, 325, 173-182. Tebe, C.C. and Vahjen, W. 193. Interference of humic acids and DNA extracted directly from soil in detection and transformation of recombinant DNA from bacteria and a yeast. Applied and Environmental Microbiology, 59, 2657. ! 38 Triska, F.J. 1970. Seasonal distribution of aquatic hyphomycetes in relation to the disapearnace of leaf liter form a wodland stream. Ph.D. Thesis, University of Pitsburgh. U. S. Census Bureau. 208. National Population Projects. Retrieved June 8, 201, from htp:/ww.census.gov/population/ww/projections/208projections.html. Walace, J.B. and Webster, J.R. 196. The role of macroinvertebrates in stream ecosystem function. Annual Review of Entomology, 41, 115-139. Walace, J.B., Webster, J.R. and Cufney, T.F. 1982. Stream detritus dynamics: regulation by invertebrate consumers. Oecologia, 53, 197-200. Webster, J.R. and Benfield, E.F. 1986. Vascular plant breakdown in freshwater ecosystems. Annual Review of Ecology and Systematics, 17, 567-594. Webster, J.R. and Waide, J.B. 1982. Effects of forest clearcuting on leaf breakdown in a southern Appalachian stream. Freshwater biology, 12, 331-344. Weyers H. S. and Suberkrop, K. 196. Fungal and bacterial production during the breakdown of yelow poplar leaves in 2 streams. Journal of the North American Benthological Society 15:408-420. Widmer, F., Seidler, R.J., Donegan, K.K. and Red, G.L. 197. Quantification of transgenic plant marker gene persistence in the field. Molecular Ecology, 6, 1-7. ! 39 Widmer, F., Seidler, R.J. and Watrud, L.S. 196. Sensitive detection of transgenic plant marker gene persistence in soil microcosms. Molecular Ecology, 5, 603-613. Witkamp, M. 1963. Microbial populations of leaf liter in relation to environmental conditions and decomposition. Ecology, 44, 370-377. Womack, K.E., Bhavsar, J. and Ravel, J. 208. Metagenomics: read length maters. Applied and Environmental Microbiology, 74, 1453. Wods, P.V. and Raison, R.J. 1982. An apraisal of techniques for the study of liter decomposition in eucalypt forests. Australian Journal of Ecology, 7, 215-225. Wright, M.S. and Covich, A.P. 205. The efect of macroinvertebrate exclusion on leaf breakdown rates in a tropical headwater stream. Biotropica, 37, 403-408. Young, R.G., Mathaei, C.D. and Townsend, C.R. 208. Organic mater breakdown and ecosystem etabolism: functional indicators for assessing river ecosystem health. Journal of the North American Benthological Society, 27, 605-625. ! 40 Figure 1.1. Stepwise process of instream leaf breakdown as proposed by Petersen and Cumins, 1974 (after Gessner, et al., 1999). ! 41 Figure 1.2. Contemporary (overlaping) aproach to studying view of the processes driving leaf liter breakdown in streams, accounting for (a) factors acting on degrading leaf liter, and (b) primary and secondary products produced during breakdown (Gessner et al., 1999). DOM, dissolved organic mater; FPOM, fine particulate organic mater. ! 42 CHAPTER I PURIFICATION OF GENOMIC DNA EXTRACTED FROM ENVIRONMENTAL SOURCES FOR USE IN A POLYMERASE CHAIN REACTION A. ABSTRACT The ability to amplify genomic DNA in a polymerase chain reaction (PCR) is dependent upon the purity of the DNA template. Environmental genomic DNA often contains contaminants (e.g., polyphenols, humic acids, polysaccharides) that reduce template purity and can be dificult to remove, thereby inhibiting PCR amplification. There is thus a ned for a method to purify extracted genomic DNA without reducing DNA concentration. In this protocol, extracted genomic DNA is embeded in agarose plugs and incubated in a formamide and salt (NaCl) solution to remove contaminants. The NaCl works to deproteinize and stabilize the DNA. The formamide serves to denature the DNA (which wil subsequently be renatured within the agarose plug) and any contaminants that may be bound to the DNA. The purified DNA is extracted from the agarose plug using a standard comercial agarose extraction method, and the DNA may then be used as a template for PCR. Genomic DNA purified using this method has ben shown to serve as an eficient template for PCR, without significant loss of DNA yield. An aditional advantage of the method is that it alows the simultaneous processing of large numbers of samples at once. ! 43 B. INTRODUCTION The increase in PCR eficiency using genomic DNA purified by this method is shown in Figure 2.1. A quantitative PCR was performed using a genomic DNA template purified by this method and the results were compared to those from templates prepared using a comercial kit and other treatment combinations. Liles et al. (208) have described a method for purifying high- molecular-weight DNA for use in the construction of metagenomic libraries (see also Isolation and Cloning of High-Molecular-Weight Metagenomic DNA from Soil Microrganisms [Liles et al. 209]). However, the protocol reported here is more rapid and is specificaly designed for the purification of genomic DNA to be used as a template for PCR amplification. C. RELATED INFORMATION The increase in PCR eficiency using genomic DNA template purified using this protocol is shown in Figure 1. A quantitative PCR was performed using a genomic DNA template purified by this method and the results were compared to those from templates prepared using a comercial kit and other treatment combinations. Liles, et al. (208) have described a method for purifying high-molecular-weight DNA for use in the construction of metagenomic libraries (see also Isolation and Cloning of High-Molecualr-Weight Metagenomic DNA from Soil Microorganisms [Liles et al., 209]). However, the protocol reported here is more rapid and is specificaly designed for the purification of ! 44 genomic DNA to be used as a template for PCR amplification. D. PROTOCOL 1. Materials 1.1. Reagents Agarose, molecular biology grade (Fisher) in 1x TAE Conical tubes, 15 mL and 50 mL (sterile) Formamide solution, stored at 4?C [R] Glass beaker for making soft agarose (sterile) Microcentrifuge tubes, 2 mL TAE bufer stock, 50X [R] 1.2. Equipment: Gel extraction kit (such as QIAquick Gel Extraction Kit or Wizard SV Gel and PCR Clean-Up System) Graduated cylinder Incubator, 15?C water bath Laboratory microwave Pipetes, p20 and p100 Pipete tips, 20 !l and 1000 !l Weigh boats 2. Method 2.1. Generation of agarose plugs and overnight incubation: 1. Extract genomic DNA from environmental sample. Genomic DNA may be directly extracted from an environmental sample using a bead beat lysis method (or comercial kit). Folow the published method for DNA extraction in terms of the amount of environmental saple to be processed. Typicaly, comercial kits wil cal for <50 mg of ! 45 environmental sample for extraction. Even with comercial kits that advertise ?PCR-ready? DNA wil be obtained, it is not unusual to have samples that may not be readily used as a PCR template. If after serial dilution of the DNA template no PCR product is obtained, and the positive controls have ben successful, this protocol may be useful as a purification method to provide truly PCR-ready DNA. 2. Prepare an apropriate volume of 2% agarose solution in 1X TAE, heat in microwave, and let cool to 45?C. Mix briefly to produce a homogenous agarose plug. The purpose of embeding in agarose is to alow rapid DNA purification and subsequent washing of agarose plugs, without the ned for DNA precipitation. Generaly, 10 ml of 2% agarose is neded per sample if this is the volume of extracted DNA from the environmental sample. 3. Mix equal volumes of 2% agarose solution and an extracted DNA solution by pipeting briefly with a 1 mL pipet tip, in a 2 mL microcentrifuge tube. Here, 10 ?L of 2% agarose was mixed by pipeting with 10 ?L of extracted DNA solution. Embeding DNA into agarose alows for a matrix to contain DNA once it is denatured by the formamide/NaCl solution and alow contaminants to move into the surrounding liquid. 4. Allow agarose plug to solidify at room temperature in the microcentrifuge tubes directly. Less than 10 min should be required. 5. Add 5X volume of an 80% formamide/1.3M NaCl solution. Mix by slowly inverting in rack. Formamide serves as a denaturant to denature the DNA, and NaCl alows for stabilization of the DNA during incubation. If the total volume of the agarose plug is 20 ml (10 ml of 2% agarose and 10 ml of extracted DNA solution) then ad a volue of 1 ml of formamide/salt solution. 6. Incubate samples for 1 hour at 15?C. 2.2. Washing of agarose plugs and extraction of DNA from agarose 7. Remove formamide and salt solution. 8. Wash 5X using 1mL 1X TAE. Add 1X TAE, invert in rack to mix, pipete of solution, and repeat. ! 46 9. Extract DNA from agarose. Effective kits for DNA removal include Qiagen?s QIAquick Gel Extraction Kit or Promega?s Wizard SV Gel and PCR Clean-Up System. 10. Gel-extracted DNA may now be used for subsequent procedures. DNA may be stored at -20?C or -85?C until further use. E. DISCUSSION Several methods exist for the purification of environmental DNA, including those that use phenolchloroform, hexadecyltrimethylamonium bromide (CTAB), polyvinylpolypyrrolidone (PVPP), cesium chloride density centrifugation, and hydroxyapatite column chromatographic purification (for reviews, see Rose- Amsaleg et al., 2001; Robe et al., 203; see also Ogram et al., 1987; Holben et al., 198; Knaebel & Crawford 195). Stefan et al. (198) have demonstrated that many of these methods (such as those using PVPP, cesium chloride, and hydroxyapatite) lower DNA yield. Often, a combination of two or more purification methods (such as a phenol-chloroform extraction folowed by use of CTAB) is required to atain adequate purification of environmental DNA. However, using numerous purification steps not only can decrease DNA yield, but also increases sample-processing time. The later is particularly disadvantageous when working with large numbers of samples. The protocol described here alows multiple samples to be processed at once. The purification and amplification of environmental DNA can often be dificult because of low yields and co-isolation of contaminants. Incubation of genomic DNA in agarose plugs during formamide and salt treatment alows ! 47 removal of contaminants without significant loss of DNA. The protocol described here has ben shown to be efective in purifying DNA from various environmental sources, such as soils, leaf liter, and marine corals, which have never successfuly provided templates for PCR amplification. Comercial kits for genomic DNA extraction typicaly shear the DNA, resulting in fragment sizes <20 kb. The DNA yield obtained using this procedure wil vary greatly depending on the initial DNA concentration. After folowing this procedure, each of the purified DNAs yielded abundant amplicons using ~20 ng of purified DNA as a template for PCR. F. TROUBLESHOTING Problem: The agarose plug remains stuck in the botom of the microcentrifuge tube. [Step 4] Solution: It may be necessary to release the agarose plug from the botom of the microcentrifuge tube by pressing the pipete tip down the side of the agarose plug. This step wil insure total suspension of the agarose plug in the formamide and NaCl solution. Problem: Insuficient purity of DNA. [Step 5] Solution: Depending upon the degree of contamination, the incubation time may be extended to overnight incubation at 15?C. Removal of the formamide and NaCl solution and replacing with fresh solution would be advisable with highly contaminated samples (i.e., having a change in color from phenolic or humic ! 48 compounds). Problem: Insuficient yield of DNA. [Step 8] Solution: Reducing the volume of the elution bufer (to aproximately 30 ml), as wel as passing the elution bufer over the column two or more times, can help to maximize the yield of DNA recovered from the column. This is the only significant loss of DNA during the protocol, folow manufacturer?s recomendations in DNA recovery from the agarose gel. ! 49 G. LITERATURE CITED Holben, W.E., Jansson, J.K., Chelm, B.K., and Tiedje, J.M. 198. DNA probe method for the detection of specific microorganisms in the soil bacterial comunity. Appl Envirnon Microbiol 54:703-711. Knaebel, D.B. and R.L. Crawford. 195. Extraction and purification of microbial DNA from petroleum-contaminated soils and detection of low numbers of toluene, octane and pesticide degraders by multiple polymerase chain reaction and southern analysis. Mol Ecol 4:579-591. Liles, M.R., Wiliamson, L.L., Rodbumrer, J., Torsvik, V., Godman, R.M., and Handelsman, J. 208. Recovery, purification, and cloning of high molecular weight DNA from soil microorganisms. Appl Environ Microbiol 74: 3302-3305. Ogram, A., Sayler, G.S., Barkay, T. 1987. The extraction and purification of microbial DNA from sediments. J Microbiol Methods 7:57-66. Stefan, R.J., Goksoyr, J., Bej, A.K., Atlas, R.M. 198. Recovery of DNA from soils and sediments. Appl Environ Microbiol 54: 2908-2915. ! 50 Figure 2.1. Bar graph representing amplified environmental genomic DNA isolated from red maple (Acer rubrum) leaf liter. Genomic DNA was purified using several treatment combinations and gel quantified prior to amplification. A standardized amount of DNA (20ng) was then used in PCR reactions containing a fluorescent probe (SYBR Green, Bio-Rad) that binds to double-stranded DNA, and the fluorescence (R n ) was measured over time. Final DNA concentration was then calculated by comparing fluorescence of a PCR standard of known DNA concentration to the fluorescence of each sample. All points represent standardized DNA quantity ? standard error. Leters indicate Tukey?s multiple comparison groupings. ! 51 CHAPTER II EFFECTS OF BENTHIC MACROINVERTEBRATES ON MICROBIAL COMMUNITIES ASSOCIATED WITH LEAF BREAKDOWN IN SMAL COASTAL PLAINS STREAMS A. ABSTRACT Processing of alochthonous leaf liter within smal, forested streams represents a valuable source of energy for stream fod webs. Benthic macroinvertebrates and microorganisms play major separate but conected roles in leaf breakdown, with microorganisms colonizing and conditioning leaf liter, which alows for subsequent colonization and consumption by macroinvertebrates. The efect of macroinvertebrates on the leaf liter-associated microbial comunity was examined by conducting a 128-d in situ incubation using red maple and water oak leaves confined within coarse and fine mesh bags to control abundance of shreding macroinvertebrates. Samples were colected at various time points over 128 d to quantify leaf breakdown and characterize macroinvertebrate assemblages and microbial comunities. Phospholipid faty acid (PLFA) analysis was used to examine diferences in fungal and bacterial biomass, and ribosomal intergenic spacer analysis (RISA) was used to assess changes in overal bacterial assemblage composition. Maple ! 52 breakdown significantly decreased in fine (vs. coarse) mesh treatments. Identification of macroinvertebrates within leaf packs revealed inefective reduction of shreders due to a high abundance of smaler shreders including Polypedilum and Leuctra species. No significant diference occurred betwen coarse and fine mesh for microbial biomass estimates or RISA profiles. The observed diference in maple breakdown betwen coarse and fine mesh leaf packs may have occurred because of mesh-specific efects on the leaf pack internal environment. Use of 1-m mesh to reduce macroinvertebrate shreder abundance may not be apropriate for al studies, depending on average macroinvertebrate shreder size. Future studies should gauge the particular shreders in the specific study streams and be sure to control for al sizes and habits of shreders. In adition, studies should correct for mesh-specific leaf mass loss by conducting mesh control studies under environmental conditions similar to those experienced during the leaf breakdown study. B. INTRODUCTION Allochthonous inputs, particularly terrestrial leaf liter, represent a significant portion of the available energy fueling heterotrophic, temperate-deciduous forested streams (Fisher and Likens, 1973), but apears to play an insignificant role in more autotrophic stream ecosystems (Minshal, 1978; Schade and Fisher, 1997). Instream breakdown of liter and its subsequent energy release to the aquatic food web, integrates both physical (fragmentation) and biological (nutrient mineralization) transformations, which involve the combined eforts of ! 53 microorganisms and macroinvertebrates (Webster and Benfield, 1986; Hieber and Gessner, 202). Microorganisms play a major role in leaf breakdown, contributing up to 28% of overal mass loss (Hieber and Gessner, 202). Fungi tend to dominate during the initial phases of breakdown and decrease in biomass over time, contributing ~15% to overal mass loss (Kaushik and Hynes, 1971; Suberkrop and Klug, 1976; Hieber and Gessner, 202). Although typicaly lower than fungi, bacterial biomass often increases as fungal biomass decreases and may contribute ~13% to overal leaf mass loss (Suberkrop and Klug, 1976; Weyers and Suberkrop 196; Hieber and Gessner, 202). Given their greater contribution during initial breakdown, studies have historicaly explored fungal colonization and succession on liter (Suberkrop and Klug, 1976; Nikolcheva and Barlocher, 204); comparatively less work has ben done to quantify the bacterial assemblage present during breakdown. Traditionaly, studies have focused on culturable bacterial taxa present during breakdown (Suberkrop and Klug, 1976); however, more recent methodological advances alow for examination of bacterial assemblage composition without the previous culturing bias (Findlay, 2010). By conditioning liter during breakdown, microorganisms also increase leaf palatability and quality for macroinvertebrates, which, in turn, can accelerate breakdown by fragmentation and consumption (Petersen and Cumins, 1974; Graca et al., 193; Wright and Covich, 205). Fragmentation by macroinvertebrates can contribute up to 64% of leaf mass loss (Hieber and Gessner, 202). Nutritional subsidies from icroorganisms to macroinvertebrates ! 54 during breakdown have ben recognized as being considerably higher than liter alone (Cumins, 1974; Graca, 201). Macroinvertebrates preferentialy consume and derive greater nutrition from, microbialy conditioned leaf liter, but the efects macroinvertebrates have on microbial biomass and comunity composition on decomposing leaf liter have not ben explored. The purpose of this study was to assess the influence of macroinvertebrate presence on the litter-associated microbial comunity during breakdown, both as overal microbial biomass and bacterial assemblage composition. Specificaly, this study was designed to examine 1) if bacterial and fungal biomass increased when abundances of large shreding macroinvertebrates were reduced, and 2) if bacterial assemblage composition difered betwen leaf packs where larger shreders were alowed to colonize and those where shreders were reduced. C. METHODS 1. Study site This study was conducted in Kings Mil Crek (UTM 0720701E 360036N), a second-order, low-gradient stream at the Fort Bening Military Instalation (FBMI) in west-central Georgia, USA. FBMI occurs south of the Fal Line in the Sand Hils subecoregion of the Southeastern Plains ecoregion (Grifith et al., 201). Kings Mil Creek is a clear (TSS = 5.45 mg/L), low-nutrient (NO3-N = 5.13 ?g N/L; SRP = 6.04 ?g/L), and acidic (pH = 4.3) stream with sandy substrate (Maloney et al., 205), and an intact deciduous riparian canopy consisting mostly of red maple (Acer rubrum), flowering dogwood (Cornus florida), yelow poplar ! 55 (Liriodendron tulipifera), swetgum (Liquidambar styraciflua), swetbay magnolia (Magnolia virginiana), black gum (Nyssa sylcatica), and water oak (Quercus nigra) (Cavalcanti, 204). The Kings Mil Creek watershed was largely forested (85.6% cover, Maloney et al., 205) with a high abundance of shreder macroinvertebrates (K.O. Maloney and R. M. Mitchel, unpubl. data), implying the importance of liter to the stream?s trophic economy. 2. Experimental design An in situ liter decomposition experiment was conducted using artificial leaf packs of contrasting mesh sizes to control the abundance of large shreder macroinvertebrates colonizing and consuming leaf liter. Acer rubrum (red maple) and Quercus nigra (water oak) were used as leaf species for leaf pack construction. These species span a moderate range of breakdown rates, with red maple having a medium breakdown rate (k=0.05-0.010) and water oak a relatively slow rate (k<0.05) (Webster and Benfield, 1986). Both species were comon in riparian zones at the study site and at FBMI in general (Cavalcanti, 204; Lockaby et al., 205). There were 9 colection dates (days 0, 1, 2, 4, 8, 16, 32, 64, and 128) from January 4, 2007 to May 12, 207, which included both initial microbial colonization and later periods of microbial succession as liter breakdown proceeded. Nine experimental blocks were established consisting of one run habitat per block containing 4 replicates of each leaf species x mesh size treatment, with each block sampled on 1 of the 9 dates; blocks were chosen randomly during the 128-d incubation. ! 56 Artificial leaf packs were held within nylon and fiberglass mesh bags (0.1524 m x 0.3048 m) placed in situ, a comon method for studying leaf breakdown in streams (Boulton and Boon, 191). Mesh bags of control leaf packs had coarse (6.35 m) mesh on one side to alow colonization of macroinvertebrates and a smaler (3.175-m) mesh on the oposite side to reduce loss of liter particles from inside the bag during incubation. In contrast, mesh bags of treatment leaf packs had a much finer mesh size (1-m) on both sides composed of fiberglass window screening to exclude large macroinvertebrates. Control leaf packs were sewn closed with nylon, whereas the fine mesh leaf packs were closed using cable ties. Leaves were colected wekly from a single tree of each species during fal 206 (December-January) using tarps strung below trees to accumulate abscissed leaves. Leaves were then air-dried in a sterile Class I biosafety cabinet to a constant mass, weighed into 4-g packs (as dry mass), and then placed into sterilized mesh bags until deployed in the stream. A separate set of packs was sampled for each leaf species x mesh size treatment on day 0 by briefly diping packs into stream water and then removing and returning them to the laboratory to quantify handling loss (Petersen and Cumins, 1974). Day 0 packs were considered to represent the initial phylosphere (terrestrial) microbial comunity for each leaf species. On the specified colection date, each leaf pack was removed from the block, placed in a Ziploc bag, and returned on ice to the laboratory. A 2-leaf subsample was then colected from each leaf pack for microbial processing, and the remaining leaves were used for determining ! 57 breakdown rate (below). The leaf subsample was ground in liquid N 2 and stored at ?80?C until processed for microbial comunity characterization (below). To determine breakdown rate, the remaining leaf sample was dried to a constant mass at 60 o C, weighed, and then combusted the sample in a mufle furnace at 550 o C for 2 h. Once ashed, the residue was reweighed and subtracted from the pre-combusted dry mass for determination of ash-free dry mass (AFDM). Breakdown rates were calculated using an exponential decay model (Petersen and Cumins, 1974) as the slope of the regression line of ln(% AFDM remaining) vs time. Sampled macroinvertebrates were stored in 70% ethanol and then sorted, measured for length (nearest m), and identified to genus when possible, except for Oligochaeta and Acari (identified to subclass). Given their major role in liter processing, shreding macroinvertebrates were identified according to Merrit and Cumins (196). Mean macroinvertebrate and shreder density (as number/g AFDM remaining) were estimated for each leaf species x mesh size treatment on each date. Liter pack macroinvertebrate assemblage composition was examined by estimating the percentage of taxa in the family Chironomidae and the orders Ephemeroptera, Plecoptera, and Trichoptera (EPT) as wel as by estimating total taxon richness and Shanon diversity (H?). Total biomass of macroinvertebrates was calculated by converting body length into AFDM using published body-length dry-mass relationships for each leaf species x mesh size treatment (Benke et al., 1999). ! 58 To characterize variation in physicochemical conditions known to afect breakdown (Webster and Benfield 1986; Dangles et al., 2004) streamwater temperature, depth, and current velocity also were quantified. Water temperature was measured hourly with HOBO Temp data logers secured to instream tree roots or submerged coarse wody debris. Current velocity diferences lead to diferential liter breakdown (Webster and Benfield, 1986) and also have ben shown to afect microbial comunity establishment (e.g. initial biofilm formation) (Batin et al., 203). Streamwater depth also can alter leaf breakdown through its efects on water temperature, dissolved oxygen, and macroinvertebrate habitat (Webster and Benfield, 1986), and these efects on the instream environment have the potential to alter liter microbial comunities. Because of these potential efects, diferences in depth and current velocity were quantified initially and imediately before removal for each leaf pack to assess variation in leaf breakdown and microbial comunities on a given sampling date atributable to depth or velocity diferences. These values were compared betwen leaf species treatments to ensure that any observed diferences betwen leaf species were not due to diferences in depth or current velocity regime. Depth and current velocity measures were also used as covariates when comparing AFDM remaining and macroinvertebrate metrics betwen leaf species. Leaf pack depth was measured using a meter stick placed at the top center of each leaf pack, and a Marsh-McBirney Flowmate current meter was used to measure flow ithin each leaf pack. Mesh size diferences could potentialy alter current velocity, which, in turn, has the potential to alter the microbial comunity within a leaf ! 59 pack, so actual velocity conditions inside each leaf pack were simulated by positioning an empty ?dumy? bag over the probe placed imediately upstream of each leaf pack prior to measurement. 3. Microbial lipids and comunity characterization 3.1 Microbial lipids. Phospholipid faty acid (PLFA) analysis was used to quantify relative abundance of diferent lipids associated with bacteria and fungi (Zeles, 199) on liter over the experiment using the folowing procedure. PLFA was adapted from Sasser (190) for saponification, formation of faty acid methyl esters (FAMEs), extraction, and a base wash. First, ~680-mg sample of the liquid N 2 -ground liter was placed in a 20-mL test tube. Samples were saponified to liberate faty acids from lipids of lysed cels with 1.0 mL saponification reagent (45 g NaOH, 150 mL methanol, 150 mL deionized water), vortexed for 10 s, heated to 10?C for 5 min in a boiling water bath, and then vortexed and reheated to 10?C for 25 min. FAMEs were formed through methylation by adding 2 mL of methylating reagent (325 mL 6.0 N HCl, 275 mL methanol), and vortexing and heating them to 80?C for 10 min. FAMEs were extracted from the aqueous phase into an organic phase using 1.25 mL extraction reagent (10 mL hexane, 100 mL methyl-tert butyl ether) tumbled for 10 min. Last, the aqueous phase was removed with a Pasteur pipete, washed samples with 3 mL of base wash (10.8 g NaOH, 90 mL distiled water) and tumbled for 5 min. Prior to chromatographic analysis, the organic phase containing FAMEs was transferred to glass vials, and FAMEs were analyzed using the Microbial Identification System (MIDI, Inc., Newark, DE, USA). The output provided known faty acid ! 60 (FA) peak responses for fungi and bacteria from each sample, which were then used to estimate relative abundance of bacterial and fungal lipids. Based on FA data, relative fungal and bacterial lipid abundance was estimated and used to determine microbial lipid diferences betwen species over the study. Relative abundance of bacteria was estimated using branched-chain saturated (e.g. iso and anteiso), hydroxyl (OH), monounsaturated, and cyclopropyl FAs (Zeles, 199). Fungal relative abundance was estimated using 3 lipid markers (18:2"6,9c, 18:1"9c, and 18:3"6c) (Guckert et al., 1985; Vestal and White, 1989). 3.2 DNA extraction for molecular analyses. Leaf subsamples were used for molecular analyses of bacterial comunities via ribosomal intergenic spacer analysis (RISA). For this procedure, genomic DNA was isolated from 0.10-g leaf liter using a Qiagen genomic DNA extraction kit (Qiagen, Valencia, CA, USA). DNA was purified using a cetyltrimethylamonium bromide (CTAB) extraction procedure (Ausubel, 194). In some cases, particularly for leaves at day 0, extracted DNA was not suficiently pure to serve as template for PCR. For those samples, an aditional round of genomic DNA purification was conducted using a combination 80% formamide and 1M NaCl treatment to provide PCR-ready genomic DNA template (see Chapter I). This formamide purification step has ben tested with DNA extracted from any diferent environments and has not ben observed to result in any loss of DNA or corresponding loss of diversity as assessed by denaturing gradient gel electrophoresis (DGE). If this method was demed necessary by the lack of PCR amplification using DNA templates ! 61 derived from comercial kit extraction, then the formamide purification method was aplied consistently to al samples from that date. 3.3 RISA of ITS regions. Ribosomal Intergenic Spacer Analysis (RISA) was conducted by PCR amplification of bacterial internal transcribed spacer (ITS) regions and separation of polymorphic ITS amplicons within a polyacrylamide gel matrix. PCR was conducted within a volume of 10 ?L containing GoGreen Master Mix (Promega; Madison, WI), 1x Bovine Serum Albumin (BSA), nuclease free water, primers, and aprox. 1-5 ng genomic DNA template, quantified spectrophotometricaly with a NanoDrop ND-100 (Thermo Fisher Scientific, Wilmington, DE, USA). Primers used for these reactions were the universal bacterial primers IRDYE 80-labeled ITSF (5'-GTCGTAACAAGGTAGCGTA-3') (Cardinale et al., 204) and ITSReub (5'-GCAAGGCATCACC-3') (Cardinale et al., 204) at a final concentration of 0.20 ?M. This primer set has ben shown to be less susceptible as other primers to known PCR biases such as those from substrate reanealing (Suzuki and Giovanoni, 196) and preferential amplification of shorter DNA templates (Cardinale et al., 204). Amplification was done according to the method of Fisher and Triplet (199), as folows: reaction mixtures were held at 94?C for 2 min, followed by 30 cycles of amplification at 94?C for 15 s, 5?C for 15 s, and 72?C for 45 s, and a final extension of 72?C for 2 min. PCR products were verified on a 1% agarose gel stained with ethidium bromide. Folowing verification of product yield and size, amplicons were separated in a 5.5% polyacrylamide gel matrix and images were recorded using a Li-Cor 430 (Li-Cor Inc., Lincoln, NE, USA). Bands were defined relative to the ! 62 highest band density for that patern; al bands with a density >10% of the highest band density were used to create a presence-absence matrix for further analysis. 4. Mesh control study To determine whether or not observed liter breakdown diferences were due to mesh-specific efects on breakdown rather than actual reduced macroinvertebrate abundance, leaf breakdown was quantified in both mesh sizes in the absence of macroinvertebrates. In May 2011 a mesh-control study was conducted in an indor artificial stream (84 in. x 14 in. x 10.75 in.). Five gram maple and oak leaf packs were constructed and contained in both fine and coarse mesh bags, with 4 replicates of each leaf x mesh treatment arrayed in a Latin square design. Packs were incubated for 32 d exposed to a continuous flow of water from a nearby pond. Current velocity (mean = 0.09 m/s, for both species) within the chanels was controled to miic conditions experienced in the original study. After 32 d, leaf packs were removed, and AFDM remaining was calculated as previously mentioned. An aditional set of packs from each treatment was used to estimate handling loss, processed as above. 5. Hypotheses and analyses Our overal hypothesis was that a reduction in macroinvertebrate abundance, particularly shreders, would lead to a significant shift in liter-associated microbial comunity composition. Given the preferential feding of macroinvertebrates on microbialy conditioned leaves, we also hypothesized that bacterial biomass would be lower in leaf packs with macroinvertebrates present than in leaf packs without macroinvertebrates. With regard to leaf breakdown, we ! 63 hypothesized that both leaf species would have a slower breakdown rate under reduced macroinvertebrate abundance. A general linear model, including days in stream, mesh treatment, leaf species and al possible interactions, using depth as a covariate, was used to test for diferences in relative abundance of bacterial and fungal biomass and depth based on mesh and leaf species treatments. Tukey?s multiple comparison tests (Zar, 199) were used to compare bacterial and fungal biomass among leaf species x mesh treatments for a given time point. Relative abundance of bacterial and fungal biomass, as wel as depth, was square root transformed to achieve normality. Current velocity did not folow a normal distribution and was compared betwen mesh treatments and among days in stream using a Kruskal- Walis test. AFDM remaining in our initial and mesh control studies was compared betwen mesh treatments for each leaf species using a one-way ANOVA. AFDM remaining values in our initial study were arcsine transformed to obtain a normal distribution prior to analysis. Due to their non-normal distribution, a nonparametric Kruskal-Walis test was used to compare macroinvertebrate metrics betwen coarse and fine mesh for each leaf species (Kruskal and Walis, 1952). Bray-Curtis dissimilarity matrices (Bray and Curtis, 1957) were calculated and used to compare microbial comunity (FAME profiles) and bacterial assemblage (RISA profiles) composition betwen leaf species and mesh treatments. In order to examine natural groupings among samples, cluster analysis was conducted to help visualize diferences in bacterial assemblage composition from RISA ! 64 profiles. A hierarchical, average linkage method was used because of its ability to be less influenced by extreme dissimilarity values, and Bray-Curtis was used because it is not influenced by joint absences of species in two samples (Bray and Curtis, 1957). An Analysis of Similarity (ANOSIM) (Clarke, 193) was used to test whether observed separation among treatment groups based on FAME and RISA profiles was significant. ANOSIM is equivalent to a 1-way ANOVA based on multispecies data and produces a global R value that ranges from -1 to +1 with values greater than 0 indicating greater dissimilarity betwen treatments than among samples (Chapman and Underwod, 199). Global R values near 0 indicate no significant diference betwen treatments. For al statistical tests #=0.05. D. RESULTS Water temperature during the 128-d incubation (January to May) ranged from 3.7 to 30.3 o C, with a mean of 13.4 o C. Current velocity imediately upstream of leaf packs did not difer significantly over time (p=0.236) or betwen mesh treatments (p=0.584) (Table 3.1). Depth of leaf packs varied significantly over the study (p<0.01); however, mean depth overal did not difer betwen mesh treatments (p=0.949) (Table 3.1) and was not a significant covariate (p=0.415) in explaining AFDM remaining throughout this study. Overal, in coarse-mesh packs, maple leaves (k=0.017) degraded ~2.8x faster than oak (k=0.06). Whereas, in fine-mesh packs, diferences betwen oak and maple were less pronounced (Table 3.1). Regarding mesh treatment, AFDM ! 65 remaining for maple leaf packs was significantly lower (breakdown was faster) for coarse (vs. fine) mesh (p<0.01; Fig. 3.1). Although breakdown of oak was slightly less in fine- (vs. coarse) mesh packs (k=0.05 and 0.006, respectively), there was no significant diference in AFDM remaining (p=0.159) (Table 3.1). The mesh-size treatments were inefective in producing diferences in macroinvertebrate assemblages, as no significant diferences existed betwen coarse and fine mesh packs of oak or maple for any macroinvertebrate metrics (Table 3.2). Chironomidae larvae dominated leaf packs of both species and mesh treatments (60-75% of total abundance, Table 3.2). Mean macroinvertebrate density ranged from ~24 to 51 individuals per g AFDM. Maple leaf packs had a higher diversity of macroinvertebrates than oak (p=0.038) and contained a significantly higher density of shreder macroinvertebrates than oak (p=0.05). The shreder functional feding group present within maple leaf packs was dominated by Polypedilum sp. (Chironomidae) (~63%), Leuctra sp. (Leuctridae) (~34%), and to a lesser extent, oribatid mites (Acari: Oribatida) (~3%). Oak leaf packs contained similar relative abundances of shreders, being dominated by Polypedilum sp. (~61%), Leuctra sp. (~37%), and oribatid mites (~2%). Specificaly loking at shreder biomass, the dominant contributor to shreder biomass in maple packs was Talaperla sp. (~56%), folowed by Tipula sp. (~19%), Polypedilum sp. (~14%), and Leuctra sp. (~8%). Shreder biomass in oak leaf packs consisted largely of Polypedilum sp. (~46%) and Leuctra sp. (~33%), and to a lesser extent Tipula sp. (~7%) and Talaperla sp. (~6%). With regard to overal macroinvertebrate biomass, biomass consisted of larvae of the ! 66 family Chironomidae (~18%) and members of the families Odontoceridae (~18%) and Elmidae (~14%). Overal comparison of FAME profiles to examine diferences in microbial comunity structure indicated no significant diference in lipid profile composition betwen coarse and fine mesh, either for maple (Global R = 0.02, p=0.236) or oak (Global R=-0.016, p=0.563). However, there was a significant diference in lipid profile composition betwen leaf species (Global R=0.47, p=0.01). Comparison of bacterial and fungal biomass estimates using the selected bacterial- and fungal-associated lipid markers revealed no diference betwen coarse and fine mesh packs (p=0.80 and 0.675, respectively) (Figs. 3.2 and 3.3). However, there was a significant diference betwen maple and oak in both bacterial and fungal biomass with oak having more fungal biomass and maple more bacterial biomass (p<0.01) (Figs. 3.2 and 3.3). Both bacterial and fungal biomass varied significantly with incubation time (p=0.07 and <0.01, respectively). The most comon bacterial lipids were 15:1w6c, 15:0 ISO, cy19:0, and cy17:0, and the most comon fungal lipid was 18:2w6,9c. RISA results indicated no diference in the bacterial assemblage composition betwen coarse and fine mesh overal (Global R=0.06, p=0.250). However, bacterial assemblage composition difered betwen leaf species (Global R=0.038, p=0.03). When separated by leaf species, no significant diference was observed betwen coarse and fine mesh for maple (Global R=0.026, p=0.108) or oak (Global R=0.023, p=0.128) separately (Figs. 3.4 and 3.5). Bacterial ! 67 assemblage composition also varied significantly over time (Global R=0.464, p=0.01). Mean current velocity was 0.09 m/s for both maple and oak leaf species during the mesh-control study conducted in artificial streams, and did not difer betwen coarse and fine mesh for maple or oak (p=0.83 and 0.206, respectively). After 32 d, maple in coarse mesh bags had 26.58% (? 1.01% SE) AFDM remaining vs. 30.08% (? 1.80% SE) for fine mesh bags. Oak in coarse mesh contained 71.98% (? 0.72% SE) AFDM remaining compared to 74.75% (? 0.83% SE) AFDM remaining in fine mesh leaf packs. Although fine mesh leaf packs contained a larger amount of AFDM remaining than coarse mesh leaf packs overal, the diference betwen mesh types was only slightly significant for oak (F=6.34, p=0.045) and not maple (F=2.3, p=0.178). E. DISCUSSION Based on abundance, richness, and biomass data, the intended reduction of macroinvertebrates in fine- vs. coarse-mesh bags, especialy for large shreders, was not achieved. This result was surprising given the widespread and efective use of fine (1m) mesh in macroinvertebrate exclusion studies of leaf breakdown (O?Conor et al., 200; Wright and Covich, 205). Many studies using this mesh size, however, have been atempting to exclude larger shreders (e.g. freshwater shrimps) (Hein and Crowl, 2010). The most comonly found shreders observed in our study were Polypedilum larvae and nymphs of the stonefly Leuctra, many of which were ~1mm in length and aparently were capable of passing through ! 68 the fine-mesh bags. The exclusion (or reduction) of macroinvertebrates in this study was not efective. So, we conclude that there was no diference in the microbial comunity, as biomass or bacterial assemblage composition betwen coarse and fine mesh treatments, likely because there was no significant reduction in macroinvertebrates. Given that the exclusion was not efective, a possible efect of macroinvertebrates on the microbial comunity canot be ruled out. In order to circumvent this uncertainty, an aditional study, possibly one using a mesh size <1m, would ned to be conducted that ensured an efective exclusion of macroinvertebrate shreders and other groups. The observed mesh efect on maple leaf breakdown rate could have potentially been due to an altered internal environment of fine mesh leaf packs given the slight increase in AFDM remaining in fine mesh packs during our mesh control study. AFDM remaining was not significantly diferent betwen coarse and fine mesh in our control study, although this was likely due to variability in AFDM remaining for maple leaves during the warmer season in which the control study was conducted (May vs. January). If this mesh control study were repeated during the same time period (and temperature regime) that our initial study was conducted, we may have ben more likely to see a greater efect of mesh size on AFDM remaining for maple leaves. Future studies atempting to exclude macroinvertebrate shreders of a length similar to that observed in this study and using 1-m mesh would ned to quantify mesh size-specific efects during the same season that their study is conducted and correct for this when calculating AFDM remaining. ! 69 Aside from the inability to efectively test macroinvertebrate-feding efects on the leaf litter-associated microbial comunity in this study, we did observe significant leaf species diferences in the microbial comunity, both in overal biomass as wel as bacterial assemblage composition. Others have noted efects of leaf species on overal microbial activity and biomass (Gulis and Suberkrop, 203). Blair et al. (190) atributed diferences in N fluxes in mixed- and single- species leaf packs to diferences in the decomposer comunity, including microbial density, again sugesting the ability of leaf species to alter microbial comunity biomass and/or composition. In this study, estimates of microbial biomass and bacterial assemblage composition varied betwen maple and oak leaves. These leaf species also showed significant diferences in macroinvertebrate diversity and shreder density, as wel as rate of breakdown, with maple leaves having higher macroinvertebrate diversity, shreder density, and a faster breakdown rate. It is possible that the observed leaf species diferences in microbial comunity structure are due to combined diferences in the leaf liter macroinvertebrate assemblage (e.g. diversity, density), as wel as diferences in leaf chemistry (e.g. N or lignin content). The results of our study indicate that use of 1-m mesh to exclude macroinvertebrate shreders is not an efective practice in streams where many macroinvertebrate shreder taxa are fairly short in length (<5m) and/or slender bodied, such as in our study. In adition, our results highlight the potential of leaf species diferences to alter macroinvertebrate assemblage composition, as wel as microbial comunity biomass and composition during leaf breakdown in ! 70 streams. In order to tease apart whether the diferences in microbial comunity composition observed in this study were due to leaf chemistry diferences or macroinvertebrate assemblage composition, an efective macroinvertebrate exclusion would ned to be accomplished and would likely require use of a much finer (<1m) mesh. ! 71 F. LITERATURE CITED Ausubel, F.M. 194. Preparation of plant DNA using CTAB. Current Protocols in Molecular Biology, Supl 27, 2.3-2.3.7. Batin, T.J., Kaplan, L.A., Newbold, J.D., Cheng, X. and Hansen, C. 203. Efects of Current Velocity on the Nascent Architecture of Stream Microbial Biofilms. Applied and Environmental Microbiology, 69, 5443-5452. Benke, A.C., Huryn, A.D., Smock, L.A. and Walace, J.B. 199. Length-mass relationships for freshwater macroinvertebrates in North America with particular reference to the southeastern United States. Journal of the North American Benthological Society, 308-343. Blair, J.M., Parmele, R.W. and Beare, M.H. 190. Decay rates, nitrogen fluxes, and decomposer comunities of single-and mixed-species foliar liter. Ecology, 71, 1976-1985. Boulton, A.J. and Boon, P.I. 191. A review of methodology used to measure leaf liter decomposition in lotic environments: time to turn over an old leaf? Marine and Freshwater Research, 42, 1-43. Bray, J.R. and Curtis, J.T. 1957. An ordination of the upland forest comunities of southern Wisconsin. Ecological monographs, 27, 325-349. Cardinale, M., Bruseti, L., Quatrini, P., Borin, S., Puglia, A.M., Rizzi, A., Zanardini, E., Sorlini, C., Corseli, C. and Dafonchio, D. 204. Comparison of Diferent Primer Sets for Use in Automated Ribosomal Intergenic Spacer Analysis of Complex Bacterial Comunities. Applied and Environmental Microbiology, 70, 6147-6156. ! 72 Cavalcanti, G.. 204. Effects of sediment deposition in aboveground net primary productivity, vegetation composition, structure, and fine root dynamics in riparian forests. MS Auburn University, Auburn, AL. Chapman, M.G. and Underwod, A.J. 199. Ecological paterns in multivariate asemblages: information and interpretation of negative values in ANOSIM tests. Marine ecology. Progress series, 180, 257-265. Clarke, K. R. 193. Non-parametric multivariate analysis of changes in comunity structure. Australian Journal of Ecology 18, 117-143. Cumins, K.W. 1974. Structure and function of stream ecosystems. BioScience, 24, 631-641. Dangles, O., Gessner, M.O., Guerold, F. and Chauvet, E. 204. Impacts of stream acidification on liter breakdown: implications for assessing ecosystem functioning. Journal of Applied Ecology, 41, 365-378. Findlay, S. 2010. Stream icrobial ecology. Journal of the North American Benthological Society, 29, 170-181. Fisher, M.. and Triplet, E.W. 199. Automated aproach for ribosomal intergenic spacer analysis of microbial diversity and its aplication to freshwater bacterial comunities. Applied and Environmental Microbiology, 65, 4630-4636. Fisher, S.G. and Likens, G.E. 1973. Energy flow in Bear Brook, New Hampshire: an integrative aproach to stream ecosystem etabolism. Ecological Monographs, 43, 421-439. ! 73 Graca, M., Maltby, L. and Calow, P. 193. Importance of fungi in the diet of Gamarus pulex and Aselus aquaticus. Oecologia, 96, 304-309. Graca, M.A.S. 201. The role of invertebrates on leaf liter decomposition in streams?a review. International Review of Hydrobiology, 86, 383-393. Grifith, G.E., Omernik, J. M., Comstock, J. A., Lawrence, S., Martin, G., Godard, A., Hulcher, V. J., Foster, T. 201. Ecoregions of Alabama and Georgia color poster with map, descriptive text, sumary tables, and photographs.. US Geological Survey, Reston, Virginia. Guckert, J.B., Antworth, C.P., Nichols, P.D. and White, D.C. 1985. Phospholipid, ester-linked faty acid profiles as reproducible assays for changes in prokaryotic comunity structure of estuarine sediments. FEMS Microbiology Leters, 31, 147-158. Gulis, V. and Suberkrop, K. 203. Interactions betwen stream fungi and bacteria associated with decomposing leaf liter at diferent levels of nutrient availability. Aquatic Microbial Ecology, 30, 149-157. Hein, C.L. and Crowl, T.A. 2010. Runing the predator gauntlet: do freshwater shrimp (Atya lanipes) migrate above waterfals to avoid fish predation? Journal of the North American Benthological Society, 29, 431-443. Hieber, M. and Gessner, M.O. 202. Contribution of stream detrivores, fungi, and bacteria to leaf breakdown based on biomass estimates. Ecology, 83, 1026-1038. Houser, J.N., Mulholand, P.J. and Maloney, K.O. 205. Catchment disturbance and stream etabolism: paterns in ecosystem respiration and gross ! 74 primary production along a gradient of upland soil and vegetation disturbance. Journal of the North American Benthological Society, 24, 538-552. Kaushik, N.K. and Hynes, H.B.N. 1971. The fate of the dead leaves that fal into streams. Archiv f?r Hydrobiologie, 68, 465-515. Kruskal, W.H. and Walis, W.A. 1952. Use of ranks in one-criterion variance analysis. Journal of the American statistical Association, 47, 583-621. Lockaby, B.G., Schiling, R., Cavalcanti, E. and Hartsfield, G. 205. Effects of sedimentation on soil nutrient dynamics in riparian forests. Journal of Environmental Quality, 34, 390. Maloney, K., Mulholand, P. and Feminela, J. 205. Influence of Catchment- Scale Military Land Use on Stream Physical and Organic Mater Variables in Smal Southeastern Plains Catchments USA.. Environmental Management, 35, 677-691. Maloney, K.O. and Feminela, J.W. 206. Evaluation of single-and multi-metric benthic macroinvertebrate indicators of catchment disturbance over time at the Fort Bening Military Instalation, Georgia, USA. Ecological Indicators, 6, 469-484. Merrit, R.W. and Cumins, K.W. 196. (Editors) An introduction to the aquatic insects of North America, 2 nd edition. Kendal Hunt. Minshal, G.W. 1978. Autotrophy in stream ecosystems. BioScience, 28, 767- 771. ! 75 Nikolcheva, L.G. and Barlocher, F. 204. Taxon-specific fungal primers reveal unexpectedly high diversity during leaf decomposition in a stream. Mycological Progress, 3, 41-49. O'conor, P.J., Covich, A.P., Scatena, F.N. and Loope, L.L. 2000. Non- indigenous bambo along headwater streams of the Luquilo Mountains, Puerto Rico: leaf fal, aquatic leaf decay and paterns of invasion. Journal of Tropical Ecology, 16, 499-516. Petersen, R.C. and Cumins, K.W. 1974. Leaf processing in a wodland stream. Freshwater Biology, 4, 343-368. Sasser, M. 190. Identification of bacteria by gas chromatography of celular faty acids. In: Technical Note #101. Microbial ID, Newark, DE. Schade, J.D. and Fisher, S.G. 197. Leaf liter in a Sonoran Desert stream ecosystem. Journal of the North American Benthological Society, 16, 612- 626. Suberkrop, K. and Klug, M.J. 1976. Fungi and bacteria associated with leaves during processing in a wodland stream. Ecology, 57, 707-719. Suzuki, M.T. and Giovanoni, S.J. 196. Bias caused by template anealing in the amplification of mixtures of 16S rRNA genes by PCR. Applied and Environmental Microbiology, 62, 625. Vestal, J.R. and White, D.C. 1989. Lipid analysis in microbial ecology. BioScience, 39, 535-541. Webster, J.R. and Benfield, E.F. 1986. Vascular plant breakdown in freshwater ecosystems. Annual Review of Ecology and Systematics, 17, 567-594. ! 76 Weyers H. S. and Suberkrop, K. 196. Fungal and bacterial production during the breakdown of yelow poplar leaves in 2 streams. Journal of the North American Benthological Society 15:408-420. Wright, M.S. and Covich, A.P. 205. The efect of macroinvertebrate exclusion on leaf breakdown rates in a tropical headwater stream. Biotropica, 37, 403-408. Zar, J.H. 199. Biostatistical analysis, Prentice hal Uper Sadle River, NJ. Zeles, L. 199. Faty acid paterns of phospholipids and lipopolysaccharides in the characterization of microbial comunities in soil: a review. Biology and Fertility of Soils, 29, 111-129. ! 77 Table 3.1. Mean (? 1SE) physicochemical variables and leaf liter breakdown rates for each mesh and leaf species treatment. Coarse = 6.38m; Fine = 1mm. Maple Oak Variable Coarse Fine Coarse Fine Depth (m) 0.17? 0.01 0.16? 0.01 0.17 ? 0.01 0.18 ? 0.01 Current velocity (m/s) 0.06 ? 0.01 0.06 ? 0.01 0.08 ? 0.01 0.08 ? 0.01 Breakdown rate (k) 0.017 0.008 0.006 0.005 ! ! 78 ! ! Figure 3.1. Ash-free dry mass (AFDM) remaining of red maple and water oak leaf packs confined within fine- and coarse-mesh treatments over a 128-d incubation in Kings Mil Creek, GA, USA. Ploted points are means (? 1SE). ! 79 ! p 0.224 0.314 0.399 0.736 0.979 0.259 0.315 0.428 0.925 Fine 102.37 ? 31.86 49.37 ? 18.83 7.88 ? 1.22 1.24 ? 0.12 14.81 ? 7.64 35.37 ? 11.17 16.82 ? 6.29 68.76 ? 5.95 17.07 ? 4.84 Oak C o a rse 63.94 ? 22.48 20.94 ? 7.39 6 .69 ? 1.10 1.14 ? 0.15 8.37 ? 2.40 23.93 ? 8.80 7.85 ? 2.89 59.80 ? 6.39 13.87 ? 4.43 p 0.052 0.093 0.106 0.301 0.485 0.153 0.179 0.756 0.554 Fine 114.8 ? 35.21 50.19 ? 17.64 9.22 ? 1.12 1.54 ? 0.09 15.72 ? 5.16 51.17 ? 15.99 22.34 ? 7.81 74.86 ? 4.50 8.16 ? 1.99 Map l e C o a rse 53.52 ? 19.19 17.90 ? 7.33 7.52 ? 1.17 1.37 ? 0.13 11.08 ? 3.92 34.77 ? 13.45 12.0 ? 5.11 69.47 ? 5.56 9.31 ? 2.52 Tab l e 3. 2. Be n t h i c m a cro i n v e rt e b ra t e m e t ri cs (m e a n ? 1 SE) f o r e a ch l e a f sp e ci e s a n d m e sh t re a t m e n t co m b i n a t i o n d u ri n g a 128 - d i n cu b a t i o n i n Ki n g s M i l C re e k, G A, U SA . C o a rse = 6 . 3 8 ; F i n e = 1 . M e tr i c M a cro i n v e rt e b ra t e a b u n d a n ce Sh re d e r a b u n d a n ce R i ch n e ss Sh a n o n d i v e rsi t y ( H ?) Bi o m a ss Ma cro i n v e rt e b ra t e d e n si t y (i n d . / g AF D M ) Sh re d e r d e n si t y (i n d . / g AF D M ) % C h i ro n o m i d a e % EPT ! 80 ! ! ! Figure 3.2. Relative abundance (mean % ? 1SE) of select bacterial lipid markers of red maple and water oak leaf packs over the 128-d incubation in Kings Mil Creek, GA, USA. ! 81 Figure 3.3. Relative abundance (mean % ? 1SE) of select fungal lipid markers of red maple and water oak leaf packs over the 128-d incubation in Kings Mil Creek, GA, USA. ! ! 82 Figure 3.4. Average linkage dendrogram based on Bray-Curtis dissimilarity measures betwen red maple leaf pack bacterial assemblage RISA profiles of coarse and fine mesh treatments. ! 83 Figure 3.5. Average linkage dendrogram based on Bray-Curtis dissimilarity measures betwen water oak leaf pack bacterial assemblage RISA profiles of coarse and fine mesh treatments. ! 84 CHAPTER IV INFLUENCE OF LEAF SPECIES ON LITER BREAKDOWN AND MICROBIAL SUCESSION IN A SMAL FORESTED STREAM A. ABSTRACT Microbial succession during leaf breakdown was investigated using several culture-independent techniques. Red maple and water oak leaves were incubated in a smal, forested stream in west-central Georgia, USA, for 128 days, and leaf packs were sampled throughout the incubation period to quantify leaf breakdown rates and microbial comunity composition. In situ breakdown rates were higher for red maple than water oak. Phospholipid faty acid analysis (PLFA) revealed a significant efect of time on the microbial lipid profiles of both maple and oak. Maple leaf microbial comunities contained higher bacterial biomass than fungal biomass, and bacterial biomass increased over the study for both leaf species. Bacterial assemblage structure was examined using complementary molecular methods, including ribosomal intergenic spacer analysis (RISA) and denaturing gradient gel electrophoresis (DGE). RISA results showed that time in stream was the most important factor structuring bacterial assemblages, with diferences between leaf species more distinct at earlier time points. DGE profiles revealed higher variability in bacterial assemblages of red maple compared to water oak, and sequencing of DGE- ! 85 resolved amplicons indicated the persistent association of Collimonas spp. in red maple microbial assemblages, taxa that are frequently known to express chitinase activity. Water oak was dominated over time by Citrobacter spp., known for tanic acid degradation. Our results sugest incubation time is the most significant factor influencing leaf liter microbial comunity composition with diferences in leaf species chemistry afecting earlier stages of microbial colonization and these leaf species-specific assemblages dissipating over time. B. INTRODUCTION Allochthonous inputs are the major source of energy and nutrients within food webs of many smal, forested streams (Webster and Benfield, 1986), with this input primarily entering streams as leaf liter from surrounding riparian vegetation (Abelho, 201; Fisher and Likens, 1973; Vanote et al., 1980; Webster et al., 195). Liter breakdown and associated nutrient release mediated by stream icroorganisms has ben, and continues to be, recognized as a critical process for system etabolism (Abelho, 201; Minshal, 1967; Young et al., 208). Allochthonous liter also provides a vital structural habitat for many stream benthic macroinvertebrates (Mackay and Kalf, 1969). Thus, the structural and energetic importance of leaf liter to forested streams makes it an integral part of overal ecosystem integrity and function (Gessner and Chauvet, 2002; Webster and Benfield, 1986; Young et al., 208). Leaf breakdown within streams consists of 3 primary phases. First, liter undergoes chemical leaching, which occurs within the first 24 to 48h after ! 86 submersion (Petersen and Cumins, 1974). Second, colonization and conditioning by fungi and bacteria makes liter softer and begins to facilitate further decomposition within a few days after submersion (Cumins, 1974; Suberkrop and Klug, 1974). Last, liter is subsequently fragmented by physical abrasion and processing by macroinvertebrate consumers, particularly the shreder functional feding group (Cumins, 1974; Graca, 201; Walace and Webster, 196), which greatly increase liter breakdown (Walace et al., 1982). Leaf chemistry traits, such as initial N concentration, C:N ratio, and lignin, vary among leaf species (Ostrofsky, 197); such leaf chemistry variation can alter breakdown rates (Petersen and Cumins, 1974). Ostrofsky (197) showed that the best individual leaf chemistry predictors of breakdown rates were %N, C:N ratio, condensed tanins, and %lignin:%N ratio. Coulson and Buterfield (1978) showed high N, and to a lesser extent P, concentrations were positively correlated to increased microbial densities and breakdown within a bog. High C:N ratios have ben shown to be associated with decreased microbial activity and are often found in leaf liter that is high in celulose and lignin with slower breakdown rates (Witkamp, 196). In adition, several studies have shown concentrations of structural compounds (e.g., lignin, hemicelulose and celulose) within liter demonstrate an inverse relationship with liter breakdown, by inhibition of fungal and bacterial colonization of liter (Ardon, 208; Mentemeyer, 1978; Berg and Staf, 1980). Fungal and bacterial species are critical to liter breakdown, and their relative contributions to the leaf conditioning process have ben assessed, ! 87 indicating a greater contribution by fungi than bacteria with bacterial contribution increasing over time and in the presence of polution (Hieber and Gessner, 202; Pascoal and C?ssio, 204). Several studies have characterized fungal and bacterial assemblages present at diferent stages of leaf breakdown using a combination of cultivation, microscopy, and assaying for reproductive structures and metabolic products (Barlocher and Kendrick, 1974; Suberkrop and Klug, 1976; McArthur et al., 194; Kreutzweiser and Capel, 203; Gulis and Suberkrop, 203). The advent of molecular techniques, including DNA sequencing and fingerprinting, provides an oportunity to characterize microbial comunity dynamics during the conditioning process with greater resolution at the microbial identity level, as wel as the ability to measure comunity similarity/dissimilarity (Findlay, 2010). For example, fingerprinting techniques have ben used to quantify fungal preferences of leaves during colonization (Nikolcheva et al., 205), and recent work on bacterial and fungal comunities demonstrated the eficacy of denaturing gradient gel electrophoresis (DGE) in revealing temporal shifts in microbial comunities during leaf conditioning (Das et al., 2007; Lyautey et al., 2005; Rees et al., 206). However, the later study provided only limited phylogenetic resolution (Das et al., 2007). Phylogenetic resolution of bacterial taxa during the leaf breakdown process would thus contribute greatly to our knowledge of bacterial population dynamics, as wel as the role of leaf chemistry as a potential modulator of bacterial assemblage structure during breakdown. Molecular techniques also have known biases, such as diferential amplification ! 88 of bacterial taxa by varying primer sets, co-migration of ribotypes, or reproducibility of the denaturing gradient. This study sought to reduce the impact of these biases by employing multiple culture-independent techniques. Our study involved the use of several complementary methods (i.e., DGE, RISA, PLFA) to quantify dynamics of stream icrobial comunities during the leaf liter conditioning process. Specificaly, sequencing of DGE ribotype bands was used to increase phylogenetic resolution during liter breakdown by 1) characterizing bacterial succession on leaf liter in a smal, forested coastal plains stream, and 2) comparing diferences in leaf bacterial comunities betwen leaf species (i.e., water oak and red maple) with strongly contrasting breakdown rates. We hypothesized that fungal lipid abundance would be higher than bacterial lipid abundance for both leaf species, but that water oak would have higher fungal lipid abundance than red maple. In adition, because of diferences in leaf chemistry and its influence on breakdown rate, we hypothesized that chemical diferences betwen oak and maple species would be reflected in disparate bacterial assemblages. C. METHODS 1. Study site The study was conducted at Kings Mil Creek (UTM 0720701E 360036N), a second-order, low-gradient stream at the Fort Bening Military Instalation (FBMI) in west-central Georgia, USA. FBMI occurs south of the Fal Line in the Sand Hils subecoregion of the Southeastern Plains ecoregion (Grifith ! 89 et al., 201). Kings Mil Creek is a low-nutrient, acidic (pH = 4.3) stream with sandy substrate (Maloney et al., 205), and an intact deciduous riparian canopy (Houser et al., 205; Maloney and Feminela, 206) consisting mostly of red maple (Acer rubrum), dogwod (Cornus spp.), yelow poplar (Liriodendron tulipifera), swetgum (Liquidambar styraciflua), swetbay magnolia (Magnolia virginiana), black gum (Nyssa sylcatica), and water oak (Quercus nigra) (Cavalcanti, 204). The Kings Mil Creek watershed was largely forested (85.6% cover, Maloney et al. 205) with a high abundance of shreder macroinvertebrates (K.O. Maloney, unpubl. data), implying the importance of liter to the stream?s trophic economy. 2. Experimental design An in situ liter decomposition experiment was conducted using Acer rubrum (red maple) and Quercus nigra (water oak) as the two leaf species. These species span a moderate range of breakdown rates, with red maple having a medium breakdown rate (k=0.05-0.010) and water oak a relatively low breakdown rate (k<0.05) (Webster and Benfield, 1986). In adition, maple species show a contrasting leaf chemistry compared to oak species, with maple (Acer spp.) having higher N content (low C:N) and oak (Quercus spp.) having a higher C:N and lignin content (Ostrofsky, 197). Both species were comon in riparian zones at the study site and at FBMI in general (Lockaby et al., 205). We used 9 colection dates (days 0, 1, 2, 4, 8, 16, 32, 64, and 128) from January to May 207, which included both early microbial colonization and those changes occurring during microbial succession as leaf liter breakdown proceeds. ! 90 Leaf packs of both species were placed in mesh bags with mesh size large enough (6.35-m) to alow macroinvertebrate colonization. Each block consisted of one run habitat containing 4 replicates of each leaf species with each block sampled on 1 of the 9 dates; blocks were chosen randomly during the 128-d study. Artificial leaf packs were held within mesh bags (0.1524 m x 0.3048 m) and placed in situ, a comon method for studying leaf breakdown in streams (Boulton and Boon, 191). Leaves were colected from a single tree of each species during fal 206 (December-January) using tarps strung below trees to accumulate abscissed leaves. Leaves were air-dried in a sterile Class I biosafety cabinet to a constant mass, weighed into 4-g aliquots, and then placed into sterilized mesh bags until deployed. Mesh bags of leaf packs had coarse (6.35 m) mesh on one side to alow colonization of macroinvertebrates and a smaler (3.175-m) mesh on the other side to reduce loss of liter particles from inside the bag during incubation. Once filed, mesh bags were sewn closed with nylon and then anchored in the stream with rebar. Leaf species were sampled on day 0 by briefly diping packs into the stream water and then removing and returning them to the laboratory to quantify handling loss (Petersen and Cumins, 1974). Day 0 packs were selected to represent the initial phylosphere microbial comunity for each leaf species, and these packs were treated similarly to all others for further processing. On the specified colection date, each leaf pack was removed from the block, placed in a Ziploc bag, and returned on ice to the laboratory. A 2-leaf subsample was removed from each leaf pack for microbial processing and the ! 91 remaining leaves were used to determine the breakdown rate (below). The leaf subsample was ground in liquid N 2 and stored at ?80?C until processed for microbial comunity characterization (below). To determine breakdown rate, the remaining leaf liter was rinsed and dried to a constant mass at 60 o C, weighed, and then combusted in a mufle furnace at 50 o C for 2 h. The ashed residue was weighed and this weight was subtracted from the pre-combusted dry mass for determination of ash-free dry mass (AFDM). Breakdown rates were calculated using an exponential decay model (Petersen and Cumins, 1974) as the slope of the regression line of ln(% AFDM remaining) vs time. Leaf carbon and nitrogen content were measured, covering a range of colection dates including pre-imersion, 1, 16, and 64 days incubation, by thermal combustion using a Perkin-Elmer 240 CHN Analyzer. Pre-immersion estimates of celulose and lignin were determined sequentialy according to procedures of Van Soest et al. (191). To characterize variation in physicochemical conditions known to afect breakdown (Webster and Benfield 1986; Dangles et al., 204) streamwater temperature, depth, and current velocity were quantified. Temperature was recorded hourly with HOBO Temp data logers. Diferences in depth and current velocity were quantified to assess variation in leaf breakdown and microbial comunities atributable to depth or velocity diferences. Leaf pack depth was measured using a meter stick placed at the top center of each leaf pack and a Marsh-McBirney Flowmate current meter was used to measure flow ithin each ! 92 leaf pack. Flow inside each leaf pack was measured by positioning an empty ?dumy? bag over the probe placed imediately upstream of each leaf pack. 3. Microbial lipids and comunity characterization 3.1 Microbial lipids. We used phospholipid faty acid (PLFA) analysis to quantify relative abundance of diferent lipids associated with bacteria and fungi on liter over the experiment (Zeles, 199). The PLFA analysis method was adapted from Sasser (190) for saponification, formation of faty acid methyl esters (FAMEs), extraction, and a base wash. First, we placed an aproximately 680- mg sample of the liquid N 2 -ground liter in a 20 mL test tube. Samples were saponified to liberate faty acids from lipids of lysed cels with 1.0 mL saponification reagent (45 g sodium hydroxide, 150 mL methanol, 150 mL deionized water), vortexed for 10 s, heated to 10?C for 5 min in a boiling water bath, and then vortexed and reheated to 10?C for 25 min. FAMEs were formed through methylation by ading 2 mL of methylating reagent (325 mL 6.0 N HCl, 275 mL methanol), and vortexing and heating them to 80?C for 10 min. FAMEs were extracted from the aqueous phase into an organic phase using 1.25 mL extraction reagent (10 mL hexane, 10 mL methyl-tert butyl ether) tumbled for 10 min. Last, the aqueous phase was removed with a Pasteur pipete, washed with 3 mL of base wash (10.8 g sodium hydroxide, 90 mL distiled water) and tumbled for 5 min. Prior to chromatographic analysis, the organic phase containing FAMEs was transferred to glass vials, and FAMEs were analyzed using the Microbial Identification System (MIDI, Inc., Newark, DE USA). The output provided faty acid (FA) peak responses for fungi and bacteria from each ! 93 sample, which were then used to estimate relative abundance of bacterial and fungal lipids. Based on FA data, we estimated relative fungal and bacterial lipids and used these measures to determine microbial lipid diferences betwen species throughout the study. Relative abundance of bacteria was estimated using branched-chain saturated (e.g. iso and anteiso), hydroxyl (OH), monounsaturated, and cyclopropyl FAs (Zeles, 199). Fungal relative abundance was estimated using only three lipid markers (18:2"6, 18:1"9c, and 18:3"6c) (Guckert et al., 1985; Vestal and White, 1989). 3.2 DNA extraction for molecular analyses. Leaf subsamples were used for two separate molecular analyses of bacterial comunities: 1) ribosomal intergenic spacer analysis (RISA) and 2) DGE. For these procedures, genomic DNA was extracted from 0.10-g leaf liter using a Qiagen genomic DNA extraction kit (Qiagen, Valencia, CA, USA). DNA was purified using a cetyltrimethylamonium bromide (CTAB) extraction procedure (Ausubel, 194). In some cases, particularly for leaves at day 0, extracted DNA was not suficiently pure to serve as template for PCR. For those samples, we conducted an aditional round of genomic DNA purification using a combination of 80% formamide and 1M NaCl treatment to provide PCR-ready genomic DNA template (see Chapter I). This formamide purification step has ben tested with DNA extracted from any diferent environments and has not ben observed to result in any loss of DNA or corresponding loss of diversity as assessed by DGE. If this method was demed necessary by the lack of PCR amplification using DNA ! 94 templates derived from comercial kit extraction, then the formamide purification method was consistently aplied to al samples from that sampling date. 3.3 RISA analysis of ITS regions. RISA analysis was conducted by PCR amplification of bacterial internal transcribed spacer (ITS) regions and separating polymorphic ITS amplicons within a polyacrylamide gel matrix. PCR was conducted with a reaction volume of 10 ?L containing GoGreen Master Mix (Promega; Madison, WI), 1x Bovine Serum Albumin (BSA), nuclease free water, primers, and aprox. 1-5 ng genomic DNA template, quantified spectrophotometricaly with a NanoDrop ND-100 (Thermo Fisher Scientific, Wilmington, DE, USA). Primers used for these reactions were the universal bacterial primers IRDYE 80-labeled ITSF (5'-GTCGTAACAAGGTAGCGTA-3') (Cardinale et al., 204) and ITSReub (5'-GCAAGGCATCACC-3') (Cardinale et al., 204) at a final concentration of 0.20 ?M. This primer set has ben shown to not be as susceptible as other primers to known PCR biases such as those due to substrate reanealing (Suzuki and Giovanoni, 196) and preferential amplification of shorter DNA templates (Cardinale et al., 2004). Amplification was done according to the method of Fisher and Triplet (199), as folows: reaction mixtures were held at 94?C for 2 min, followed by 30 cycles of amplification at 94?C for 15 s, 5?C for 15 s, and 72?C for 45 s and a final extension of 72?C for 2 min. We verified PCR products on a 1% agarose gel stained with ethidium bromide. Folowing verification of product yield and size, we separated amplicons in a 5.5% polyacrylamide gel matrix and images were recorded using a Li-Cor 430 (Li-Cor Inc., Lincoln, NE, USA). ! 95 3.4 DGE analysis of 16S rRNA amplicons. For DGE, replicates from each sampling date were examined using a 2-step process. Genomic DNA extracted from leaves was used as a template in a PCR. Fifty ?L reactions were conducted using a GoGreen Master Mix (Promega; Madison, WI) that included Taq polymerase, dNTPs, and Mg +2 -containing bufer (at 1x concentration). In addition, the PCR reactions included 5 ?L of 1:50 diluted DNA template, 1x BSA and 0.20 ?M each of the universal bacterial primer set 518R (5$- ATTACCGCGCTGCTG-3$) (Lane, 191) and 38F-GC (5?- CGCCGCGCGCCGCGCCGTCCGCGCCCGCCTCTACGGA GCAGCAG-3?) (Muyzer et al., 193). This technique is known to preferentialy amplify the most abundant bacterial ribotypes (Muyzer et al., 193), and this specific primer set was chosen to amplify the V3 region of the 16S rDNA, which has ben shown to resolve bacterial taxa and produce comparable results to ful- length (V1-V9) 16S rDNA gene sequence (Huse et al., 208). PCR conditions included 2-min of denaturation at 95?C folowed by 30 cycles of 95?C for 1 min, 1 min of annealing at 55?C, and then 2 min of extension at 72?C (Mulis et al., 194). An 8% polyacrylamide gel was poured that contained a vertical gradient of formamide and urea at a final gradient concentration range of 45 to 5%. PCR products were loaded in the gel with 20 ?L (aprox. 198-240 ng) per lane and electrophoresed for 15 h at 10 V and 60?C. After electrophoresis, the gels were stained with ethidium bromide for 10 min and then rinsed in deionized water for 15 min, after which bands were visualized using an AlphaImager HP gel documentation system (Alpha Inotech, San Leandro, CA). Individual bands ! 96 were considered distinct ribotypes (Muyzer and Smala, 198). Abundant rRNA amplicons for a given sampling time were visualy identified, excised, and used as template in a subsequent PCR. All subsequent reactions were in a total volume of 25 ?L containing GoGreen Master Mix (Promega; Madison, WI), primers 518R and 38F (without the GC clamp), and 2 ?L of excised PCR product. All PCR reactions generated product, without requiring further resolution of bands, and were sequenced using 518R (5 ?M) and BigDye sequencing chemistry by the Lucigen Corporation (Midleton, WI). Unaligned sequences were compared to the GenBank nr/nt database using the BLASTn search algorithm at the National Center for Biotechnology Information (NCBI) to obtain the nearest neighbor (%95% similarity) based on 16S rDNA gene sequence identity. Some studies have found that a portion of 16S rDNA isolated DGE amplicons, created using universal 16S-rDNA bacterial primers, corresponded to plant 16S rDNA sequences (Kim et al., 2010; Saito et al., 207). However, no mitochondrial or plastid sequences were obtained from our excised DGE amplicons. 4. Data analyses Diferences were compared betwen species and over time in environmental variables and liter breakdown, using 2-way analysis of variance (ANOVA) (Zar, 199). AFDM data were arcsine transformed prior to analysis to satisfy the assumptions of normality and equality of variance. Current velocity and depth measures were square root transformed prior to analysis to satisfy normality and equality of variance. Overal lipid profiles and bacterial assemblage ! 97 composition were compared for each leaf species over time using Analysis of Similarity (ANOSIM) based on a Bray-Curtis dissimilarity matrix. Fungal and bacterial faty acid abundance estimates were square root transformed and compared for each date using a 2-way ANOVA based on days in stream and leaf species, as wel as Tukey?s multiple comparisons within each time point. RISA gel images were analyzed using BioNumerics Software v5.0 (Applied Maths, Kortrijk, Belgium) to quantify bacterial assemblage similarity. Bands were defined relative to the highest band density on that patern, where al bands with a density >10% of the highest band density were used to create a presence- absence matrix for further analysis. Similarities betwen band presence-absence fingerprints were calculated using Jaccard?s similarity coeficient. Cluster analysis using Ward?s method was then used to create dendrograms for visualization of bacterial assemblage similarity (Saitou and Nei, 1987). A variety of clustering algorithms were compared, and Ward?s method of hierarchical clustering yielded the most satisfactory result. An alpha level of 0.05 was used for al statistical analyses. D. RESULTS 1. Physicochemical conditions Water temperature during the 128-d study (January to May) ranged from 3.7 to 30.3 o C, with a mean temperature of 13.4 o C over the entire incubation period. Mean water depth at individual leaf packs was 0.17 m, which did not difer betwen species (F=0.43, p=0.515). Mean current velocity imediately upstream ! 98 of leaf packs was 0.07 m/s, which also did not difer betwen species (F=3.73, p=0.059). Mean water depth at leaf packs decreased significantly (p=<0.01) from 0.31 m on day 1 to 0.09 m on day 128. Mean current velocity also varied significantly over the study (p=<0.01) with the highest velocity experienced on day 4 (0.1 m/s) and the lowest on day 16 (0.02 m/s). The degree to which pH and oxygen varied over the course of this study was not measured. However, streamwater pH and dissolved oxygen measurements on day 1 indicated the stream was acidic (pH = 4.3), but wel oxygenated (8.65 mg/L, 8% saturation). 2. Litter breakdown Mean leaf liter breakdown rates for red maple (hereafter maple) and water oak (hereafter oak) over the study were k = 0.075 d -1 for maple and k = 0.026 d -1 for oak (Fig. 4.1). Maple breakdown was significantly faster than oak (p<0.01). The exponential decay model explained 83.6 and 63.1% of the variation in maple and oak leaf breakdown, respectively (Fig. 4.1). Maple AFDM decreased rapidly from 10% AFDM remaining at day 0 to 81.2% after 1 day in situ; in contrast, oak showed litle AFDM change during the same 1-d interval (~2% loss). After 128 d, maple had 46.6% AFDM remaining, compared to 73.9% remaining for oak. Oak leaves also had a higher nitrogen content than maple (1.276% vs. 0.43%) and a lower C:N ratio (35.43) than maple (106.83). Nitrogen content of imersed oak leaves decreased over the study from 0.52% folowing 1 day instream incubation to 0.4% after 64 days; whereas, maple leaf nitrogen content increased from 0.20% on day 1 to 0.32% on day 64. Non-imersed oak leaves ! 99 contained higher celulose (21.31%) and lignin (14.63%) compared to maple leaves (10.87% and 6.68%, respectively). 3. Microbial comunity characterization Overal, lipid profiles varied significantly over time for both maple (p=0.01) and oak (p=0.05). Bacterial and fungal lipid relative abundance, estimated by FAME analysis, significantly difered betwen maple and oak on al dates except day 128 (Fig. 4.2 and Fig. 4.3). Overal, bacterial lipid abundance on maple was higher than oak (p<0.01) (Fig. 4.2), whereas oak showed higher abundance of fungal lipids than maple (p<0.01) (Fig. 4.3). Fungal lipid relative abundance on oak leaves tended to decrease over the 128-d incubation, as bacterial lipids steadily increased (Fig. 4.3). Analysis of bacterial assemblages in leaf packs using RISA demonstrated a dependence of comunity structure upon time of incubation (p<0.01) (Fig. 4.4). The degree to which leaf species played a role in structuring bacterial assemblages, although significant overal (p=0.03), tended to vary with time. Cluster analysis indicated assemblage structure aportioned into 3 temporal groupings, pre-imersion, early breakdown, and later breakdown assemblages (Fig. 4.4). However, within each grouping, leaf species apeared to play a greater role structuring bacterial assemblages during pre-imersion (day 0) (p=0.04) than during later breakdown (day 128) (p=0.103). Overal bacterial ribotype eveness, as revealed by DGE, was higher for maple than oak, with 21 distinct ribotypes on maple (Fig. 4.5) and ! 100 18 on oak (Fig. 4.6). The highest ribotype eveness for maple (14) was on day 32, whereas eveness on oak was highest on days 0 and 1 (10 and 16 ribotypes, respectively). Overal, seven diferent bacterial ribotypes were considered abundant or consistent members of maple leaf packs and were selected for excision and sequencing. Bacterial members of maple liter included the genera Ralstonia (!-Proteobacteria; day 0), Sphingopyxis ("- Proteobacteria; days 1 and 32), Comamonas (!-Proteobacteria; day 4), Herbaspirilum (!-Proteobacteria; day 4), Mesorhizobium ("-Proteobacteria; day 4), Nitrosospira (!-Proteobacteria; day 8), and Colimonas spp. (!- Proteobacteria; days 16, 64, and 128) with percent sequence identities to GenBank matches ranging from 95 to 10%. In contrast, the oak liter bacterial assemblage was less variable than maple, as indicated by less ribotype variation over the study. The genus Citrobacter (#-Proteobacteria) occurred on oak leaves on al dates. Genera from five other ribotypes also were dominant including Mesorhizobium ("-Proteobacteria; day 0), Sphingomonas ("-Proteobacteria; day 1), Aquabacterium (!-Proteobacteria; day 8), Sphingopyxis ("-Proteobacteria; days 0), and Thiobacilus (!- Proteobacteria; day 128) with percent sequence identities to GenBank matches ranging from 95 to 10%. E. DISCUSSION Leaf breakdown rates strongly difered betwen species (i.e., fast for maple and slower for oak), a result that was consistent with previous work (Webster and Benfield, 1986). Such diferences reflect contrasting leaf chemistry, ! 101 with higher lignin content likely contributing to slower breakdown and resulting in diferences in microbial comunities betwen these leaf species (Webster and Benfield, 1986; Gessner and Chauvet, 194; Ostrofsky, 197). Similar to previous studies (Das et al., 207), lipid profiles of both leaf species varied significantly with incubation time. Maple leaf packs showed higher abundance of bacterial lipids compared to oak. This diference could be a function of oak having lower surface area available for colonization by bacteria than maple as wel as having higher lignin content making it harder for bacteria to colonize (Das et al., 207). In adition, the faster breakdown of maple could increase nutrient availability on leaf surfaces and thus stimulate bacterial growth. Alternatively, oak showed higher fungal marker abundance than maple on most sampling dates, possibly because fungi are more capable of colonizing lignin-rich oak leaves than bacteria and tend to dominate for a longer period; in contrast, on maple leaves, with less lignin, bacteria are capable of earlier colonization folowing fungal conditioning and potential fungal degradation, thus potentialy reducing fungal biomass (Gessner and Chauvet, 194; Gulis and Suberkrop, 203). Fungal lipid marker abundance on oak did decrease over the 128-d incubation (Fig. 3.3), so increased bacterial lipid abundance on oak leaves may have resulted from increased nutrient availability for bacteria folowing fungal colonization and conditioning (Gulis and Suberkrop, 203). RISA results indicated that bacterial assemblage structure was more influenced by incubation time than by leaf species. These results, as wel as our microbial lipid profiles, agree with a study by Das et al. (207) that found that ! 102 time was a key factor in structuring fungal, bacterial, and actinomycete assemblages compared to the influence of leaf species. The greater impact of time on assemblage structure could be, in part, due to observed temporal variation in streamwater environmental conditions (e.g., temperature, water depth, and current velocity). Slight variations in streamwater conditions could favor some bacterial taxa over others potentialy altering bacterial assemblage structure as breakdown proceeded. The lower degree of separation betwen maple and oak bacterial assemblages during later breakdown (days 128) likely is atributable to plant compounds (e.g. tanins, phenolics, etc.) present in higher quantities earlier in the breakdown process that leached out and/or diminished in quantity. Canhoto and Graca (199) demonstrated the inhibitory efects of secondary compounds, such as tanic acid, on microbial biomass. In their study, decreased growth of four diferent aquatic hyphomycetes was observed folowing the adition of increasing concentrations of tanic acid and eucalyptus oils. In our study, chemical diferences betwen maple and oak would likely be at their highest during the initial days of incubation prior to significant leaching; thus, our observations of the most extreme separation betwen maple and oak bacterial assemblages occurring during these earlier stages of breakdown are consistent with this mechanism. This increased similarity betwen maple and oak bacterial assemblages over time could also be a reflection of increased bacterial colonization on oak leaves folowing fungal conditioning, permiting colonization of leaf litter-associated bacteria that were previously incapable of colonizing the oak leaf surface due to increases in exposed surface area. The ability of fungal ! 103 activity to increase bacterial colonization has ben sugested (Suberkrop and Klug, 1976), and prior studies have also shown increased bacterial colonization in response to increased surface area and available organic mater (Yamamoto and Lopez, 1985). DGE results of dominant taxa revealed a more diverse and variable bacterial assemblage associated with maple compared to oak. Increasing similarity over time betwen maple and oak bacterial assemblages, as indicated by the RISA results, indicates the presence of similar bacterial taxa on red maple and water oak leaf liter. However, when comparing the abundant bacterial taxa identified via DGE, a trend towards increasing Collimonas spp. was observed with maple assemblages, whereas a more consistent high relative abundance of Citrobacter spp. was observed in oak assemblages. Although these two specific taxa were diferent betwen oak and maple leaves, other abundant and sequenced taxa were shared betwen leaf species (e.g. Mesorhizobium, Sphingopyxis). Das et al. (207) also found a band (ribotype) that was only found on sugar maple leaves and one that was only found on white oak leaves. Although each leaf species does seem to have its own leaf species-specific taxa, increased similarity over time betwen leaf species is likely due to increased colonization by bacterial taxa of lower abundance. Abundance of bacterial taxa within the genus Collimonas associated with maple leaf liter sugests the presence of chitinolytic bacteria, as many Colimonas spp. are known to express chitinase activity (Fritsche et al., 208). Colimonas spp. presence occurred concurrently with a decrease in fungal lipid ! 104 marker abundance, potentialy resulting from increased fungal chitin degradation (Fig. 4.5, ribotype G). In contrast, a less variable bacterial assemblage associated with oak had a high relative abundance of Citrobacter spp. Many species within the genus Citrobacter, such as Citrobacter freundi, have the ability to degrade tanic acid (Fig. 4.6, ribotype J) (Murugan et al., 208). This is significant given that oak (Quercus spp.) leaves tend to have relatively high concentrations of hydrolysable tanins (Zimer et al., 202), which in turn form complexes with other macromolecules to create tanic acid. In adition, oak leaves have ben shown to have slow leaching rates of phenolics, likely due to their thick cuticle (Kuiters and Sarink, 1986), and would potentialy explain why Citrobacter spp. were found in association with oak leaves on al sampling dates. It is also important to note that PCR amplification may under represent ribotype richness (Acinas et al., 205), so bacterial taxa observed in this analysis of maple and oak leaves are likely only the more abundant ribotypes at their respective dates of incubation; therefore, it is expected that a much greater bacterial taxa richness at lower relative abundance exists within these bacterial assemblages. Our results showed a more variable bacterial assemblage in fast breakdown leaf species (i.e., red maple), whereas slower degrading species such as water oak were dominated by higher fungal lipid marker abundance and a less variable bacterial assemblage. However, as leaf liter leachate rates decrease over time, colonization by aditional bacterial species on oak may become less constrained by leaf chemistry diferences and become more ! 105 phylogeneticaly diverse, exhibiting properties more similar to the bacterial assemblages of fast species such as red maple, whose leachate is more easily dispersed and are more readily colonized. Overal, our results sugest leaf chemistry diferences play a role early in the breakdown process prior to leaching and colonization by lesser abundant bacterial taxa, yet the main factor controling microbial comunity structure apears to be incubation time. Future studies should examine microbial succession within leaf microbial comunities in relation to changes in various environmental factors, particularly examining the efect of anthropogenic disturbances. Disturbance frequently leads to alteration of instream environmental conditions and leaf breakdown, so it is likely that such alterations also afect associated microbial comunities given the extreme sensitivity of microbes to environmental change (Young et al., 2008; Ager et al., 2010). Enhanced understanding of the structural dynamics and functional roles of microorganisms in streams also wil enable a beter understanding of the impacts of disparate environmental influences on ecosystem-level processes. ! 106 F. LITERATURE CITED Abelho, M. 201. From literfal to breakdown in streams: a review. Scientific World Journal, 1, 656-680. Acinas, S.G., Sarma-Rupavtarm, R., Klepac-Ceraj, V. and Polz, M.F. 205. PCR- induced sequence artifacts and bias: insights from comparison of two 16S rRNA clone libraries constructed from the same sample. Applied Environmental Microbiology, 71, 8966-8969. Ager, D., Evans, S., Li, H., Liley, A.K. and van der Gast, C. 2010. Anthropogenic disturbance afects the structure of bacterial comunities. Environmental Microbiology, 12, 670-678. Ausubel, F.M. 194. Preparation of plant DNA using CTAB. Current Protocols in Molecular Biology, Supl 27, 2.3-2.3.7. B?rlocher, F., Seena, S., Wilson, K.P. and Wiliams, D.. 208. Raised water temperature lowers diversity of hyporheic aquatic hyphomycetes. Freshwater Biology, 53, 368-379. Berg, B.R. 200. Liter decomposition and organic mater turnover in northern forest soils. Forest Ecology and Management, 133, 13-22. Berg, B.R. and Tam, C.O. 191. Decomposition and nutrient dynamics of liter in long-term optimum nutrition experiments. Scandinavian Journal of Forest Research, 6, 305-321. Boulton, A.J. and Boon, P.I. 191. A review of methodology used to measure leaf liter decomposition in lotic environments: time to turn over an old leaf? Marine and Freshwater Research, 42, 1-43. ! 107 Canhoto, C. and Gra?a, M.A.S. 199. Leaf barriers to fungal colonization and shreders Tipula lateralis. consumption of decomposing Eucalyptus globules. Microbial Ecology, 37, 163-172. Cardinale, M., Bruseti, L., Quatrini, P., Borin, S., Puglia, A.M., Rizzi, A., Zanardini, E., Sorlini, C., Corseli, C. and Daffonchio, D. 204. Comparison of diferent primer sets for use in automated ribosomal intergenic spacer analysis of complex bacterial comunities. Applied and Environmental Microbiology, 70, 6147-6156. Cavalcanti, G. 204. Effects of sediment deposition on aboveground net primary productivity, vegetation composition, structure, and fine root dynamics in riparian forests. M.S. Thesis, Auburn University, Auburn, AL. Dang, H. and Lovel, C.R. 200. Bacterial primary colonization and early succession on surfaces in marine waters as determined by amplified rRNA gene restriction analysis and sequence analysis of 16S rRNA genes. Applied and Environmental Microbiology, 66, 467-475. Dangles, O., Gessner, M.O., Guerold, F. and Chauvet, E. 204. Impacts of stream acidification on liter breakdown: implications for assessing ecosystem functioning. Ecology, 41, 365-378. Das, M., Royer, T.V. and Lef, L.G. 207. Diversity of fungi, bacteria, and actinomycetes on leaves decomposing in a stream. Applied and Environmental Microbiology, 73, 756-767. DeSantis, T. Z., Hugenholtz, P., Larsen, N., Rojas, M., Brodie, E.L., Keler, K., Huber, T., Dalevi, D., Hu, P., and Andersen, G.L. 206a. Greengenes, a ! 108 Chimera-Checked 16S rRNA Gene Database and Workbench Compatible with ARB. Applied and Environmental Microbiology 72, 5069-72. DeSantis, T. Z., Hugenholtz, P., Keler, K., Brodie, E.L., Larsen, N., Piceno, Y.M., Phan, R., and Andersen, G.L. 206b. NAST: a multiple sequence alignment server for comparative analysis of 16S rRNA genes. Nucleic Acids Res 34, W394-399. Fisher, S.G. and Likens, G.E. 1973. Energy flow in Bear Brook, New Hampshire: an integrative aproach to stream ecosystem etabolism. Ecological monographs, 43, 421-439. Fisher, M.. and Triplet, E.W. 199. Automated aproach for ribosomal intergenic spacer analysis of microbial diversity and its aplication to freshwater bacterial comunities. Applied and Environmental Microbiology, 65, 4630-4636. Fritsche, K., De Boer, W., Gerards, S., Van Den Berg, M., Van Veen, J.A. and Leveau, J.H.J. 208. Identification and characterization of genes underlying chitinolysis in Colimonas fungivorans Ter331. FEMS Microbiology Ecology, 66, 123-135. Gartner, T.B. and Cardon, Z.G. 204. Decomposition dynamics in mixed-species leaf liter. Oikos, 104. 230-246. Gessner, M.O. and Chauvet, E. 194. Importance of stream icrofungi in controling breakdown rates of leaf liter. Ecology, 75, 1807-1817. Gessner, M.O. and Chauvet, E. 202. A case for using liter breakdown to assess functional stream integrity. Ecological Applications, 12, 498-510. ! 109 Gessner, M.O., Chauvet, E. and Dobson, M. 199. A perspective on leaf liter breakdown in streams. Oikos, 85, 377. Gleave, A. P., R. K. Taylor, B. A. M. Morris, and D. R. Greenwod. 1995. Cloning and sequencing of a gene encoding the 69-kDa extracelular chitinase of Janthinobacterium lividum. FEMS Microbiol. Let. 131, 279?288. Graca, M.A.S. 201. I. Leaf liter processing and invertebrates: The role of invertebrates on leaf liter decomposition in streams ? a review. International Review of Hydrobiology, 86, 383-393. Greathouse, E.A., March, J.G. and Pringle, C.M. 205. Recovery of a tropical stream after a harvest-related chlorine poisoning event. Freshwater Biology, 50, 603-615. Grifith, G.E., Omernik, J.M., Comstock, J.A., Lawrence, S., Martin, G., Godard, A., Hulcher, V.J., and Foster, T. 201. Ecoregions of Alabama and Georgia, color poster with map, descriptive text, sumary tables, and photographs: Reston, Virginia, U.S. Geological Survey map scale 1:1,700,000. Gulis, V. and Suberkrop, K. 203. Interactions betwen stream fungi and bacteria associated with decomposing leaf liter at diferent levels of nutrient availability. Aquatic Microbial Ecology, 30, 149-157. H?tenschwiler, S., Aeschliman, B., Co?teaux, M.-M., Roy, J. and Bonal, D. 208. High variation in foliage and leaf liter chemistry among 45 tree species of a neotropical rainforest comunity. New Phytologist, 179, 165- 175. ! 110 Hieber, M. and Gessner, M.O. 202. Contribution of stream detritivores, fungi, and bacteria to leaf breakdown based on biomass estimates. Ecology, 83, 1026-1038. Hoisington, D. Khairalah, M. and Gonzalez-de-Leon, D. 194. Laboratory Protocols: CIMYT Applied Biotechnology Center. Second Edition, Mexico, D.F.: CIMYT. Houser, J.N., Mulholand, P.J. and Maloney, K.O. 205. Catchment disturbance and stream etabolism: paterns in ecosystem respiration and gross primary production along a gradient of upland soil and vegetation disturbance. Journal of the North American Benthological Society, 24, 538-552. Huse, S.M., Dethlefsen, L., Huber, J.A., Mark Welch, D., Relman, D.A. and Sogin, M.L. 208. Exploring microbial diversity and taxonomy using SSU rRNA hypervariable tag sequencing. PLoS Genetics, 4(11): e1000255. Jenkins, C.. and Suberkrop, K. 195. The influence of water chemistry on the enzymatic degradation of leaves in streams. Freshwater Biology, 33, 245- 253. Kabagale, A.C., Cornu, B., van Vliet, F., Meyer, C-L., Mergeay, M., Simbi, J-B.L., Droogmans, L., Wauven, C.V. and Verbrugen, N. 2010. Diversity of endophytic bacteria from the cuprophytes Haumaniastrum katangense and Crepidorphopalon tenuis. Plant Soil, 334, 461-474. Kim, S-Y., Yoo, K-S., Kim, J.E., Kim, J-S., Jung, J.Y., Jin, Q., Eom, H-J., and Han, N.S. 2010. Diversity analysis of lactic acid bacteria in Korean rice ! 111 wines by culture-independent method using PCR-denaturing gradient gel electrophoresis. Fod Science and Biotechnology, 19, 749-755. Kinkel, L.L. 197. Microbial population dynamics on leaves. Annual Review of Phytopathology, 35, 327-347. Klapenbach, J.A., Saxman, P.R., Cole, J.R. and Schmidt, T.M. 201. rrnDB: the Ribosomal RNA Operon Copy Number Database. Nucleic Acids Research, 29, 181-184. Kreutzweiser, D.P. and Capel, S.S. 203. Benthic microbial utilization of diferential dissolved organic mater sources in a forest headwater stream. Canadian Journal of Forest Research, 33, 1444-1451. Kuiters, A.T. and Sarink, H.M. 1986. Leaching of phenolic compounds from leaf and nedle liter of several deciduous and coniferous trees. Soil Biology and Biochemistry, 18, 475-480. Lane, D.J. 191. 16S/23S rRNA sequencing. In: Nucleic Acid Techniques in Bacterial Systematics. Eds. E. Stackebrandy and M. Goodfellow, pp. 115- 175. John Wiley and Sons, New York. Lockaby, B.G., Governo, R., Schiling, E., Cavalcanti, G. and Hartsfield, C. 205. Effects of sedimentation on soil nutrient dynamics in riparian forests. Journal of Environmental Quality, 34, 390-396. Lyautey, E., Jackson, C.R., Cayrou, J., Rols, J.L. and Garabetian, F. 2005. Bacterial comunity succession in natural river biofilm assemblages. Microbial Ecology, 50, 589-601. Mackay, R.J. and Kalf, J. 1969. Seasonal variation in standing crop and species ! 112 diversity of insect comunities in a smal Quebec stream. Ecology, 50, 101-109. Maidak, B.L., Cole, J.R., Lilburn, T.G., Parker, C.T., Jr., Saxman, P.R., Farris, R.J., Garrity, G.M., Olsen, G.J., Schmidt, T.M. and Tiedje, J.M. 201. The RDP-I Ribosomal Database Project. Nucleic Acid Research, 29, 173-174. Maloney, K.O. and Feminela, J.W. 206. Evaluation of single- and multi-metric benthic macroinvertebrate indicators of catchment disturbance over time at the Fort Bening Military Instalation, Georgia, USA. Ecological Indicators, 6, 469-484. Maloney, K.O., Mulholand, P.J. and Feminela, J.W. 205. Influence of catchment-scale military land use on stream physical and organic mater variables in smal Southeastern Plains catchments USA. Environmental Management, 35, 677-691. McArthur, J.V., Aho, J.M., Rader, R.B. and Mils, G.L. 194. Interspecific leaf interaction during decomposition in aquatic and flodplain ecosystems. Journal of the North American Benthological Society, 13, 57-67. Mulis, K.B., Ferr?, F., Gibs, R.A. and Watson, J.D. 194. The polymerase chain reaction. Birkhauser, Boston. Murugan, S., Devi, P.U., Mahesh, P., Suja, S. and Mani, K.R. 208. Production of tanase by Citrobacter freundi under solid state fermentation and its aplication in fruit juices debitering. Biosciences Biotechnology Research Asia, 5, 301-306. ! 113 Muyzer, G. and Smala, K. 198. Application of denaturing gradient gel electrophoresis DGE. and temperature gradient gel electrophoresis TGE. in microbial ecology. Antonie van Leuwenhoek, 73, 127-141. Muyzer, G., De Wal, E.C. and Uiterlinden, A.G. 193. Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Applied and Environmental Microbiology, 59, 695-700. Nikolcheva, L.G., Bourque, T. and B?rlocher, F. 205. Fungal diversity during initial stages of leaf decomposition in a stream. Mycological Research, 109, 246-253. Nol, M., Mathies, D., Frenzel, P., Derakshani, M. and Liesack, W. 205. Succession of bacterial comunity structure and diversity in a pady soil oxygen gradient. Environmental Microbiology, 7, 382-395. Ostrofsky, M.L. 197. Relationship betwen chemical characteristics of autumn- shed leaves and aquatic processing rates. Journal of the North American Benthological Society, 16, 750-759. Petersen, R.C. and Cumins, K.W. 1974. Leaf processing in a wodland stream. Freshwater Biology, 4, 343-368. Preston, C.M., Trofymow, J.A., Niu, J. and Fyfe, C.A. 202. Harvesting and climate efects on organic mater characteristics in British Columbia coastal forests. Journal of Environmental Quality, 31, 402-413. Res, G.N., Watson, G.O., Baldwin, D.S. and Mitchel, A.M. 206. Variability in sediment microbial comunities in a semipermanent stream: impact of ! 114 drought. Journal of the North American Benthological Society, 25, 370- 378. Saito, A., Ikeda, S., Ezura, H. and Minamisawa, K. 207. Microbial comunity analysis of the phytosphere using culture-independent methodologies. Microbes and Environments, 22, 93-105. Saitou, N. and Nei, M. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Molecular Biology and Evolution, 4, 406-425. Sasser, M. 190. Identification of bacteria by gas chromatography of celular faty acids. In: Technical Note #101. Microbial ID, Newark, DE. Schloss, P.D. and Handelsman, J. 205. Introducing DOTUR, a computer program for defining operational taxonomic units and estimating species richness. Applied and Environmental Microbiology, 71, 1501-1506. Shrestha, P.M., Nol, M. and Liesack, W. 207. Phylogenetic identity, growth- response time and rRNA operon copy number of soil bacteria indicate diferent stages of comunity succession. Environmental Microbiology, 9, 2464-2474. Suberkrop, K. and Klug, M.J. 1976. Fungi and bacteria associated with leaves during processing in a wodland stream. Ecology, 57, 707-719. Trofymow, J.A., More, T.R., Titus, B., Prescot, C., Morrison, I., Siltanen, M., Smith, S., Fyles, J., Wein, R. and Camir?, C. 202. Rates of liter decomposition over 6 years in Canadian forests: influence of liter quality and climate. Canadian Journal of Forest Research, 32, 789-804. ! 115 Walace, J.B., Webster, J.R. and Cufney, T.F. 1982. Stream detritus dynamics: regulation by invertebrate consumers. Oecologia, 53, 197-200. Walace, J.B., Eggert, S.L., Meyer, J.L. and Webster, J.R. 197. Multiple trophic levels of a forest stream linked to terrestrial liter inputs. Science, 277, 102-104. Webster, J.R. and Benfield, E.F. 1986. Vascular plant breakdown in freshwater ecosystems. Annual Review of Ecology and Systematics, 17, 567-594. Wright, M.S. and Covich, A.P. 205. The efect of macroinvertebrate exclusion on leaf breakdown rates in a tropical headwater stream. Biotropica, 37, 403-408. Yamamoto, N. and Lopez, G. 1985. Bacterial abundance in relation to surface area and organic content of marine sediments. Journal of Experimental Marine Biology and Ecology, 90, 209-220. Young, I.M., Crawford, J.W., Nunan, N., Oten, W., Spiers, A. and Donald, L.S. 208. Chapter 4 Microbial Distribution in Soils: Physics and Scaling. In: Advances in Agronomy. pp. 81-121. Academic Press. Zar, J.H. 199. Biostatistical analysis. 4 th ed. Prentice-Hal, Uper Sadle River, NJ. Zeles, L. 199. Faty acid paterns of phospholipids and lipopolysaccharides in the characterisation of microbial comunities in soil: a review. Biology and Fertility of Soils, 29, 111-129. Zimer, M., Penings, S.C., Buck, T.L. and Carefot, T.H. 202. Species specific paterns of liter procesing by terrestrial isopods (Isopoda: ! 116 Oniscidea) in high intertidal salt marshes and coastal forests. Functional Ecology, 16, 596-607. ! 117 Figure 4.1. Ash free dry mass (AFDM) remaining over time during breakdown of red maple (o) and water oak (!) leaf packs incubated for 128 days in Kings Mil Creek, GA, USA. Ploted points are means (? 1SE). ! 118 Figure 4.2. Relative abundance (%) of bacterial lipid markers of red maple and water oak leaf packs over the 128-d incubation in Kings Mil Creek, GA, USA. Points on graph represent mean relative abundance (%) ? 1SE. (* = p<0.01, ns = not significantly diferent) ! 119 Figure 4.3. Relative abundance (%) of fungal lipid markers of red maple and water oak leaf packs over the 128-d incubation In Kings Mil Creek, GA, USA. Points on graph represent mean relative abundance (%) ? 1SE. (* = p<0.01, ns = not significantly diferent) ! 120 Figure 4.4. Dendrogram of RISA electropherograms displaying bacterial assemblage similarities calculated using Ward?s method based on Jaccard?s similarity coeficient. Day Leaf species Jaccard?s coefficient ! 121 Figure 4.5. Denaturing gradient gel electrophoresis (DGE) analysis of red maple leaf pack bacterial assemblages. Uper case leters indicate sequenced ribotypes. (Ribotype key: A = Comamonas, B = Sphingopyxis, C = Herbaspirilum, D = Nitrosospira, E = Mesorhizobium, F = Ralstonia, G = Colimonas) ! 122 Figure 4.6. Denaturing gradient gel electrophoresis (DGE) analysis of water oak leaf pack bacterial assemblages. Uper case leters indicate sequenced ribotypes. (Ribotype key: B = Sphingopyxis, E = Mesorhizobium, H = Sphingomonas, I = Aquabacterium, J = Citrobacter, K = Thiobacilus) ! 123 CHAPTER V EFECTS OF SEDIMENT DISTURBANCE ON BACTERIAL ASSEMBLAGE ASSOCIATED WITH LEAF BREAKDOWN IN SMAL COASTAL PLAINS STREAMS A. ABSTRACT This study investigated the efects of sediment disturbance via upland land use on the bacterial assemblage of decomposing leaf liter. A 64-day in situ breakdown study was conducted using red maple and water oak leaf packs incubated within six streams (3 low- and 3 high-sediment disturbance) in west- central Georgia, USA. Leaf packs were sampled throughout the incubation and used to calculate breakdown rates as wel as to characterize the leaf liter- associated macroinvertebrate and bacterial assemblages. In situ breakdown rates of maple were slower under high-sediment disturbance. High-sediment disturbance leaf packs of both leaf species had significantly less macroinvertebrate abundance, density, taxa richness, Shanon?s diversity, and biomass. The bacterial assemblage structure of red maple leaf packs was examined using both ribosomal intergenic spacer analysis (RISA) and 454 bar- coded pyrosequencing. Both RISA and pyrosequencing results showed significant separation betwen low- and high-sediment disturbance leaf pack ! 124 bacterial assemblage composition folowing 32 and 64 days instream incubation. Pyrosequencing results revealed an increased relative abundance of taxa within the phylum Acidobacteria in low-sediment disturbance leaf packs. High-sediment disturbance leaf packs had an increased abundance of taxa within the phyla Firmicutes (Clostridiaceae) and delta-Proteobacteria (Geobacteraceae). Both low- and high-sediment disturbance leaf packs decreased in the relative abundance of gamma-Proteobacteria (Pseudomonadaceae) betwen 32 and 64 days. Thirty-one percent of the variation in bacterial assemblage composition was significantly explained by streamwater pH, which also correlated with disturbance intensity. Our results sugest a shift in leaf pack bacterial assemblage composition under increased sedimentation and catchment disturbance intensity toward a bacterial assemblage dominated by taxa capable of withstanding harsh environmental conditions (e.g. increased streamwater pH and temperature). B. INTRODUCTION Allochthonous leaf liter inputs represent a valuable source of energy for stream ecosystems, especialy in smal, forested streams. In their classic study, Fisher and Likens (1973) showed aproximately 9% of the energy available within their study stream at Hubard Brook Experimental Forest (West Thornton, NH) consisted of alochthonous inputs. The breakdown of this alochthonous leaf liter, as defined by Hieber and Gessner (202), is ?the combined result of physical and biological mineralization and transformation processes?. Leaf ! 125 breakdown consists of several phases including leaching, microbial conditioning, and fragmentation (by both physical abrasion and macroinvertebrate shreding activity) (Abelho, 201; Boulton and Boon, 191; Petersen and Cumins, 1974; Webster and Benfield, 1986). A variety of instream factors (both abiotic and biotic) have ben shown to influence the rate of leaf breakdown, including leaf chemistry, macroinvertebrate abundance, dissolved nutrients, pH, dissolved oxygen, water temperature, and sedimentation (Sponseler and Benfield, 201; Webster and Benfield, 1986). Disturbance, both natural and anthropogenic, plays a major role in stream ecology and has ben shown to directly and indirectly afect stream ecosystems (Resh et al., 1988; Maloney and Weller, 201). Anthropogenic disturbance is defined as ?any human-mediated event or activity that is virtualy unknown in natural systems in terms of type, frequency, intensity, duration, spatial extent, or predictability over the last century? (Naiman et al., 205). Anthropogenic disturbance can result from any land use activities, including but not limited to acid mining, agricultural practices, timber harvesting, and urbanization. According to the U.S. Census Bureau, the U.S. population is expected to increase by aproximately 21% in the next twenty years (U.S. Census, 208). Increases in human population size of this magnitude wil greatly increase the occurrence of human-mediated disturbance in our natural ecosystems. Human-mediated changes to the landscape can alter geomorphic processes and destabilize existing chanel shapes and are often associated with above normal sedimentation rates via increased erosion and deposition (Maloney ! 126 et al., 205). This increased sedimentation can alter or degrade many instream variables, such as stream habitat stability, as wel as afect the structure of stream fod webs (Allan, 204; Downes et al., 2006; Henley et al., 200). Instream environmental measures, similar to those that afect leaf breakdown, are greatly afected by sedimentation. Stream water temperature, turbidity and dissolved oxygen available in stream water are a few of the most notable factors afected (for review see Ryan, 191; Lemly, 1982). Specificaly with regard to aquatic invertebrates, sedimentation can afect population size and assemblage structure in multiple ways. Aquatic invertebrate density and diversity are directly related to substrate diversity (Reice, 1974; Gore, 1985). Sedimentation can reduce habitat availability for some invertebrates, making them ore susceptible to predation (Newcombe and Macdonald, 191). Aside from its efects on habitat availability, sediment can also afect certain invertebrate functional feding groups, altering population size through its efects on primary production as wel as by cloging feding structures (Newcombe and Macdonald, 191; Lemly, 1982). With regard to the overal process of leaf breakdown, sedimentation has ben shown to be capable of either increasing or decreasing the rate of breakdown, likely due to physical abrasion or burial, respectively (Benfield et al., 201; Webster and Waide, 1982). Sponseler and Benfield (201) found leaf breakdown rate to be positively correlated with substrate particle size. In this sense, smaler sediment likely buries leaf liter and reduces leaf breakdown, whereas larger sediment particles increase abrasion thus speding up leaf ! 127 breakdown. Aside from burial, sedimentation and its subsequent efects on instream habitat can alter secondary producer composition, which can also lead to decreased breakdown rates (Sponseler and Benfield, 201). These interactive efects betwen physical factors (i.e. sedimentation) and stream organisms can lead to drastic alterations in leaf breakdown and energy release. Previous studies have shown decomposing leaf liter to be dominated by Gram-negative bacteria, including Proteobacteria (particularly the classes &- and '-), Actinobacteria, as wel as members of the Cytophaga-Flavobacterium- Bacteroidetes (CFB) group (Das et al., 206; Suberkrop and Klug, 1976). Changes in many environmental factors (e.g. nutrient levels, temperature, pH) have been shown to afect bacterial assemblages (Hil et al., 200; Witkamp, 1963 and 196). These are also factors comonly altered via anthropogenic disturbance. In a broad sense, many of the bacteria previously found to be associated with leaf liter are also comonly found in aquatic sediments and terrestrial soils. Bacterial taxa found to be sediment-associated in the substrata of streams include many Proteobacteria (#-, &-, and '-) and members of the CFB group (Santmire and Lef, 207). In terrestrial environments, comon soil bacteria include members of the phyla Acidobacteria, Verrucomicrobia, Firmicutes, Gemmatimonadetes, Actinobacteria, and Proteobacteria (specificaly #-, &-, and '-) (Joseph et al., 203). Although studies have explored comon bacterial taxa in both aquatic and terrestrial environments, litle is known regarding the efects of sedimentation on leaf liter-associated microbial comunity composition. ! 128 Given the role microbes play in the critical process of leaf breakdown, and the known efects of sedimentation on instream environmental conditions, as wel as their potential efects on microbial comunities, it is important to investigate the efects of sedimentation on leaf liter-associated microbial comunities. The purpose of this study was specificaly to examine the efects of sediment disturbance on bacterial assemblages in leaf packs during leaf breakdown. We hypothesized that increased amounts of sedimentation would both decrease the rate of leaf breakdown and decrease leaf liter-associated bacterial assemblage similarity. In adition, we hypothesized that this efect of sediment disturbance on leaf pack bacterial assemblage similarity would be the same regardless of leaf species. C. METHODS 1. Study sites This study was conducted in west-central Georgia at Fort Bening Military Instalation (FBMI). FBMI occurs within the Sand Hils subecoregion of the Southeastern Plains ecoregion, south of the Fal Line. At FBMI, a combination of military training and forest management within military compartments has led to disturbance of upland terrestrial vegetation, underlying soil and, in turn, alterations in stream physicochemistry, organic mater abundance, and aquatic biota (Houser et al., 2006; Maloney, Mulholland and Feminella, 2005; Maloney and Feminella, 2006). Within FBMI, I conducted a disturbance study in 6 streams (Kings Mil Creek tributary, Lois Creek, Bonham tributary, and 3 Saly Branch ! 129 tributaries, Table 5.1). All streams were second order except two (a first order Saly Branch tributary in compartment F1 and a third order Saly Branch Tributary in compartment F3). The study streams were low gradient with sandy substrate (mean particle size 0.56-0.89 m) (Maloney et al., 2005) and intact riparian canopy (Houser et al., 2005; Maloney and Feminella, 2006). To quantify the efects of stream disturbance on leaf breakdown, this study was conducted in 6 streams, 3 of which occur in highly disturbed watersheds (BC, SB2, SB3), with stream chanels that show high sediment disturbance and low biotic integrity, and 3 streams in relatively less-disturbed watersheds (KM, LC, SB1) with correspondingly lower sediment disturbance and higher biotic integrity. Study streams were selected based on their disturbance intensity level, measured as the percentage of the watershed occurring as bare ground and road cover (Maloney et al., 2005). 2. Experimental design An in situ liter decomposition experiment was conducted using leaf species that were comon riparian species at FBMI (Lockaby et al., 2005). The leaf species included Acer rubrum (red maple) and Quercus nigra (water oak). These leaf species were chosen because they span a range of breakdown rates, with red maple having a medium breakdown rate (k=0.05-0.010) and water oak a relatively low breakdown rate (k<0.05) (Webster and Benfield, 1986). There were 5 colection dates (days 0, 8, 16, 32, and 64) ranging from January to March 207. The 3 streams from each disturbance treatment (low- vs. high-disturbance) were used as replicates and sampled for leaf liter bags on ! 130 each date. Each block consisted of one run habitat containing 4 replicates of each leaf species with each block sampled on 1 of the 5 dates; blocks were chosen randomly during the 64-d study. Artificial leaf packs held within mesh bags (0.1524 m x 0.3048 m) and placed in situ were used, as this is a comon method for studying leaf breakdown in streams (Boulton and Boon, 191). Leaves were colected from a single tree of each species during fal 206 (December-January) using tarps strung below trees to accumulate abscissed leaves. Leaves were air-dried in a sterile Class I biosafety cabinet to a constant mass, weighed into 4-g aliquots, and then placed into sterilized mesh bags until deployed. Mesh bags of leaf packs had coarse (6.35 m) mesh on one side to alow colonization of macroinvertebrates and a smaler (3.175-m) mesh on the other side to reduce loss of liter particles from inside the bag during incubation. Once filed, mesh bags were sewn closed with nylon and then anchored in the stream with rebar. For each leaf species day 0 samples were taken by briefly diping packs into the stream water and then removing and returning them to the laboratory to quantify handling loss (Petersen and Cumins, 1974). On the specified colection date, we removed each leaf pack from the block, placed it in a Ziploc bag, and returned it on ice to the laboratory. A 2-leaf subsample was taken from each leaf pack for microbial processing, and the remaining leaves were used for determination of breakdown rate (below). The leaf subsample was ground in liquid N 2 and stored it at ?80?C until processed for microbial comunity characterization (below). ! 131 The remaining leaf liter was rinsed to remove any benthic macroinvertebrates and sediment. All macroinvertebrates were stored in 70% ethanol and then sorted, measured (to the nearest m), and identified to genus when possible (except for Oligochaeta and Acari, which were identified to subclass). Given their major role in leaf liter fragmentation, shreding macroinvertebrates were identified according to Merrit and Cumins (196). Mean abundance of macroinvertebrates and shreders, as wel as total macroinvertebrate and shreder density (ind./g AFDM remaining) was calculated for each leaf species/disturbance combination at each time point. The percentages of taxa in the family Chironomidae and the orders Ephemeroptera, Plecoptera, and Trichoptera (EPT), were estimated due to both their general sedimentation tolerance (Shaw and Richardson, 201) and disturbance sensitivity (Barbour et al., 199), respectively. In adition, we estimated total taxa richness, Shanon diversity (H?), and total biomass (Benke, 199). To determine breakdown rates, the remaining leaf liter was then dried to a constant mass at 60 o C, weighed, and then combusted in a mufle furnace at 550 o C for 2 h. The ashed residue was reweighed and subtracted from the pre- combusted dry mass for determination of ash-free dry mass (AFDM). Breakdown rates were calculated using an exponential decay model (Petersen and Cumins, 1974) as the slope of the regression line of ln(% AFDM remaining) vs time. The amount of sediment from each pack was quantified from al leaf packs, excluding those from KM, which were from a separate study where no sediment was recorded. All sediment rinsed from leaf liter and remaining after sorting of ! 132 macroinvertebrates was dried to a constant mass at 60?C, weighed, and combusted in a mufle furnace at 50?C for 2 h to move al organic material. Folowing this, the sediment was weighed and recorded as the amount of sediment in each leaf pack. To characterize variation in physicochemical conditions known to afect breakdown (Webster and Benfield 1986; Dangles et al., 204), we quantified streamwater temperature, pH, depth, and current velocity. Water temperature was measured hourly with HOBO Temp data logers. Streamwater pH was measured using a Thermo Orion Model 420. Diferences in depth and current velocity were quantified to assess variation in leaf breakdown and microbial comunities atributable to depth or velocity diferences. Leaf pack depth was measured using a meter stick placed at the top center of each leaf pack, and a Marsh-McBirney Flowmate current meter was used to measure current velocity within each leaf pack. Current velocity inside each leaf pack was measured by positioning an empty ?dumy? bag over the probe placed imediately upstream of each leaf pack. 3. Bacterial asemblage characterization 3.1 DNA extraction for molecular analyses. We used leaf subsamples for 2 separate molecular analyses of bacterial comunities: 1) ribosomal intergenic spacer analysis (RISA) and 2) 454 pyrosequencing. For these procedures, we isolated genomic DNA from 0.10-g leaf liter using a Qiagen genomic DNA extraction kit (Qiagen, Valencia, CA, USA). DNA was purified using a cetyltrimethylamonium bromide (CTAB) extraction procedure (Ausubel, 194). ! 133 In some cases, extracted DNA was not suficiently pure to serve as template for PCR. For those samples, we conducted an aditional round of genomic DNA purification using a combination of 80% formamide and 1M NaCl treatment to provide PCR-ready genomic DNA template (see Chapter I). This formamide purification step has ben tested with DNA extracted from any diferent environments and has not ben observed to result in any loss of DNA or corresponding loss of diversity as assessed by DGE. If this method was demed necessary by the lack of PCR amplification using DNA templates derived from comercial kit extraction, then the formamide purification method was consistently aplied to al samples from that sampling date. 3.2 RISA analysis of ITS regions. RISA analysis was accomplished by PCR amplification of bacterial internal transcribed spacer (ITS) regions and separating polymorphic ITS amplicons within a polyacrylamide gel matrix. PCR was conducted within a volume of 10 ?L containing GoGreen Master Mix (Promega; Madison, WI), 1x bovine serum albumin (BSA), nuclease free water, primers, and aprox. 1-5 ng genomic DNA template, quantified spectrophotometricaly with a NanoDrop ND-100 (Thermo Fisher Scientific, Wilmington, DE, USA). Primers used for these reactions were the universal bacterial primers IRDYE 80-labeled ITSF (5'-GTCGTAACAAGGTAGCGTA-3') (Cardinale et al., 204) and ITSReub (5'-GCAAGGCATCACC-3') (Cardinale et al., 204) at a final concentration of 0.20 ?M. This primer set has ben shown to not be as susceptible as other primers to known PCR biases such as those due to substrate reanealing (Suzuki and Giovanoni, 196) and preferential amplification of shorter DNA ! 134 templates (Cardinale et al., 204). Some studies have found that a smal percentage of 16S rDNA amplicons, derived from using universal 16S-rDNA bacterial primers with DNA template from leaves, corresponded to plant 16S rDNA sequences (Kim et al., 2010; Saito et al., 207). However, the ITSF/ITSReub primers used in this study for ITS region amplification have not shown this problem (Cubaka Kabagale et al., 2010). Amplification was done according to the method of Fisher and Triplet (199), as folows: reaction mixtures were held at 94?C for 2 min, followed by 30 cycles of amplification at 94?C for 15 s, 5?C for 15 s, and 72?C for 45 s and a final extension of 72?C for 2 min. PCR products were verified on a 1% agarose gel stained with ethidium bromide. Folowing verification of product yield and size, we separated amplicons in a 5.5% polyacrylamide gel matrix and images were recorded using a Li-Cor 430 (Li-Cor Inc., Lincoln, NE, USA). 3.3 Bar-coded pyrosequencing of bacterial assemblages. Twelve red maple samples were selected for further pyrosequencing of their bacterial assemblages. These 12 samples represented the later time points (32 and 64 days), and for each time point we used two replicates from the highest (SB3) and lowest (LC) disturbance sites and one replicate from the second highest (SB2) and lowest (KM) disturbance sites. A 457-bp region of the 16S rRNA gene was amplified that included the hypervariable regions V3 and V4, using a fusion primer set found to be suitable for classification of 16S rRNA genes from complex microbiomes (Nossa et al., 208). The forward primer (5?- CGTATCGCTCCTCGCGCATCAG-NNNNNNNNNN- ! 135 GAGGCAGCAGTRGAAT-3?) contained the Roche adaptor A, a unique 10- bp MID-barcode used to tag each PCR product (designated by NNNNN; see Table 5.2), and the bacterial primer 347F (Nossa et al., 208). The reverse primer (5??CTATGCGCTGCAGCCGCTCAG- CTACCRGGTATCTAATC-3?) contained the Roche adaptor B, and the bacterial primer 803R (Nossa et al., 2008). PCRs consisted of 2 ?l of each 12.5 ?M forward and reverse fusion primer, 2 ?l of BSA (10X), 5 ?l of template DNA, and 25 ?l PfuUltra Hotstart PCR Master Mix (2X) (Agilent Technologies; Wilmington, DE, USA) and were adjusted to a final volume of 50 ?l using nuclease free water. Samples were initialy denatured at 95?C for 1 min, followed by 30 cycles of 95?C for 1 min, 5?C for 1 min, and 72?C for 1 min. A final extension of 10 min at 72?C was aded to ensure complete amplicons extension. Folowing verification of PCR products on a 1% agarose gel, PCR products were purified using an ethanol precipitation, and DNA concentration was quantified using a NanoDrop ND-100 (Thermo Fisher Scientific, Wilmington, DE, USA). Purified PCR products were then diluted to equimolar concentrations (16ng/?l), poled, and sequenced using 454 GS FLX Titanium chemistry (Engencore; Columbia, SC, USA). 4. Data analyses Physicochemical conditions were compared betwen low- and high- sediment disturbance sites using 1-way ANOVAs (Zar, 199). When necessary, physicochemical values (particularly sediment) were log transformed in order to achieve a normal distribution. Breakdown rates were compared for each leaf ! 136 species betwen low- and high-sediment disturbance sites using 1-way ANOVAs (Zar, 1999). Macroinvertebrate metrics were not normaly distributed and were compared betwen low- and high-sediment disturbance sites for each leaf species using a Kruskal-Walis test (Kruskal and Walis, 1952). Correlations betwen macroinvertebrate metrics and the amount of sediment found in leaf packs were tested using Spearman?s rank correlation for al macroinvertebrate metrics that showed a significant diference betwen high- and low-sediment disturbance sites for multiple time points. RISA gel images were analyzed using BioNumerics Software v5.0 (Applied Maths, Sint-Martens-Latem, Belgium). Bands were defined relative to the highest band density on that patern, where al bands, with a density >10% of the highest band density, were selected and used to create a presence-absence matrix for further analysis. Nonmetric multi- dimensional scaling (NMDS) based on a Bray-Curtis dissimilarity index (Bray and Curtis, 1957) and three dimensions was used to visualize diferences in RISA profiles of bacterial assemblages, and an Analysis of Similarity (ANOSIM) (Clarke, 193) was used to determine if low- and high-sediment disturbance profiles for each leaf species were significantly diferent. Pyrosequencing data was processed using the software pipeline Quantitative Insights Into Microbial Ecology (QIME) as described (Caporaso et al., 2010). Prior to analysis in QIME, sequences were trimed using the CLC Genomics Workbench (CLC Bio), and only sequences with a quality score of at least 25, containing no ambiguous nucleotides or mismatches in the primer sequence, and a minimum length of 20bp were used for further analyses. ! 137 Sequences were imported into QIME, sorted based on their respective bar codes and denoised using Denoiser (version 0.91) (Reder and Knight, 2010). Sequences were grouped into Operational Taxonomic Units (OTUs) using the ?uclust? method and a similarity threshold of 97%. These individual OTUs were then classified using the RDP classifier (>80% confidence) (Wang et al., 207) and aligned using PyNAST (Caporaso et al., 209). To estimate diversity, these sequences were rarefied and analyzed at the same level of surveying efort (185 sequences per sample). Alpha diversity was estimated within each sample and included measuring observed species (OTUs), Shanon-Wiener diversity (H?), and estimated species richness (Chao1). Presence and absence of individual OTUs among samples (beta diversity) was compared using a pairwise, unweighted UniFrac distance matrix and visualized in two dimensions using NMDS based on a Bray-Curtis dissimilarity index (Bray and Curtis, 1957). An ANOSIM (Clarke, 193) was used to determine if low- and high-sediment disturbance bacterial assemblage compositions were significantly diferent. Diferences betwen the relative abundance of specific bacterial taxonomic groups in low- vs. high-sediment disturbance sites at individual times were tested for using 1-way ANOVAs (Zar, 199). Redundancy analysis (RDA) was used, folowing a Helinger transformation, to see if the physicochemical variables measured explained a significant portion of the variation observed in pyrosequenced bacterial assemblage composition (Legendre and Galagher, 2001). Forward stepwise regression was then used to determine which physicochemical variables specificaly explained the most variation. Correlations ! 138 were also tested for among environmental variables and betwen environmental variables and relative abundance of bacterial taxa using Pearson?s correlation. For al statistical tests, unless otherwise noted, a significance level of #=0.05 was used to determine significance. All statistical computations were completed using SigmaPlot for Windows (version 12), MINITAB (version 15), and R Statistical Software (including the packfor and vegan packages) (version 2.13.1). D. RESULTS 1. Physicochemical conditions Over the 64-d study (January to March) average water temperature was coldest in LC (10.03?C) and warmest in SB2 (10.54?C) (Table 5.3). Overal, average water temperature in low-sediment disturbance streams (10.08?C) was significantly cooler than high-sediment disturbance streams (10.36?C) (p=0.04). Streamwater pH ranged from a low of 3.92 to a high of 5.31 (mean=4.49) (Table 3). The average depth of leaf packs betwen low- and high-sediment disturbance sites was not significantly diferent (p=0.249) (Table 5.3). Average current velocity measurements betwen low- and high-sediment disturbance sites were also not significantly diferent (p=0.416) (Table 5.3). Sediment in leaf packs varied significantly by site (p<0.01) and steadily increased over the 64-day study period. Leaf packs from SB1, although considered a low-sediment disturbance site based on calculated disturbance intensity, had the highest mean amount of sediment (153.8 g). We believe the increased sedimentation in this stream is likely due to a legacy efect from past ! 139 land use practices at Fort Bening. For instance, site data from 194 indicates the percent bare ground and road cover to be aprox. 25.7% versus its recent measure of 8.4% (K.O. Maloney, unpubl. data). In adition, BC, which was considered a high-sediment disturbance site based on its calculated disturbance intensity had very litle sediment within leaf packs (mean=7.70g). This stream has ben noted to behave diferently than predicted for its catchment disturbance level in previous studies of these sites (Mulholand et al., 205). It is believed that the presence of a broad forested flodplain bordering the catchment of this stream alows this stream to remain fairly undisturbed, despite having a disturbed catchment (Mulholand et al., 205). Leaf packs from KM were from a separate study, and at the time, no sediment data were recorded. However, from personal observation the amount of sediment in these packs was minimal and comparable to the amount found in leaf packs from site LC. Leaf packs from SB3 (mean=95.50 g) and SB2 (mean=91.60 g) both contained large amounts of sediment, and LC leaf packs had the lowest mean amount of sediment (2.49 g). 2. Litter breakdown Rates of leaf breakdown for red maple (mean k=0.021) were significantly faster than water oak (mean k=0.05) (p<0.01). Overal, the rate of breakdown for red maple (hereafter maple) in low-sediment disturbance sites was significantly faster than in high-sediment disturbance sites (p=0.05) with k=0.026 and k=0.016 in low- and high-sediment disturbance sites, respectively (Figure 5.1). Water oak (hereafter oak) breakdown rates were not significantly ! 140 diferent betwen oak leaves incubated in low- and high-sediment disturbance streams (p=0.879) with k=0.04 in both treatments (Figure 5.1). 3. Macroinvertebrates For both leaf species, mean taxa richness steadily increased throughout the study and was significantly higher in low-disturbance leaf packs than high- disturbance for both leaf species (maple p=0.014 and oak p=0.04) (see Table 5.4). Shanon?s diversity and macroinvertebrate abundance, density, and biomass were significantly higher in low-disturbance leaf packs for both leaf species (see Table 5.4). Shreder abundance and density were not significantly diferent betwen low- and high-disturbance streams for maple leaf packs. However, oak leaf packs in low-disturbance streams contained significantly higher shreder abundance and density. In general, both leaf species showed higher shreder abundance and density in low-disturbance streams than high- disturbance streams, but this diference was only significant for oak leaf packs (p=0.012). Taxa richness and Shanon?s diversity estimates were significantly correlated to the amount of sediment within a leaf pack for both maple and oak on day 64 (see Figures 5.2 and 5.3). The percentage of macroinvertebrates from the family Chironomidae and the orders Ephemeroptera, Plecoptera, and Trichoptera was not significantly diferent between high- and low-disturbance stream leaf packs for either leaf species (see Table 5.4). 4. Bacterial asemblage characterization NMDS of RISA profiles using the Bray-Curtis dissimilarity index and 3 dimensions revealed separation of high- and low-sediment disturbance leaf pack ! 141 RISA profiles folowing an aprox. 1-month instream incubation (days 32 and 64) for both leaf species (Figures 5.4 and 5.5; see Figures 5.4 and 5.5 for plots of first 2 dimensions and stress levels). Analysis of similarity (ANOSIM) between low- and high-sediment disturbance leaf pack RISA profiles revealed the separation observed in the NMDS plots at days 32 and 64 to be significant for both maple (day 32, global R=0.23, p=0.016; day 64, global R=0.31, p=0.03) and oak (day 32, global R=0.74, p=0.01; day 64, global R=0.23, p=0.03). There was no significant separation of RISA profiles betwen low- and high- sediment disturbance leaf packs for maple or oak on day 8 (ANOSIM, p= 0.076 and p=0.052, respectively) or day 16 (ANOSIM, p=0.47 and p=0.241, respectively). Given the significant diference in RISA profiles betwen low- and high- sediment disturbance leaf packs at days 32 and 64, these time points were further selected for pyrosequencing and analysis of their bacterial assemblages. Since a significant diference in breakdown rate was only observed for maple leaves in low- vs. high-sediment disturbance sites, a subset of twelve maple samples from days 32 and 64 were used for this analysis. These samples included 3 low- and 3 high-sediment disturbance samples from each time point (days 32 and 64). From these analyses, we were able to obtain suficient high- quality sequences from 10 samples, 6 from day 32 (3 low and 3 high) and 4 from day 64 (2 low and 2 high). A total of 5694 sequences were obtained for classification with a mean of aprox. 570 classifiable sequences per sample (range 185-178). In general, after 32 days instream, low-sediment disturbance ! 142 leaf pack bacterial assemblages had more OTUs and higher diversity than high- sediment disturbance leaf packs (Table 5.5), and the reverse of this trend was observed after 64 days insteam (Table 5.5). Overal, these diferences were not statisticaly significant for either time point. Pairwise, unweighted UniFrac distances were visualized using NMDS (k=2; stress=0.081) (see Figure 5.6) and overal showed significant separation betwen low- and high-sediment disturbance samples (ANOSIM; global R=0.52, p=0.019). Sumaries of the bacterial assemblage composition for each day x disturbance treatment are depicted in Figures 5.7 and 5.8. Low-sediment disturbance leaf packs on day 32 contained many bacterial phyla including taxa that affiliated with the phyla Proteobacteria (77.1%), Acidobacteria (12.6%), CFB group (4.5%), Actinobacteria (1.5%), and Firmicutes (0.8%). High-sediment disturbance leaf packs on day 32 also contained Proteobacteria (54.8%), Acidobacteria (3.1%), CFB group (8.9%), and Actinobacteria (3.6%). However, these samples had a significantly higher proportion of taxa affiliated with the phylum Firmicutes (25.1%; p=0.046), 93% of which were from the class Clostridia, mostly of the family Clostridiaceae. Low- and high-sediment disturbance leaf packs after 32 days contained similar proportions of taxa that affiliated with the phyla #-Proteobacteria (41.4%, 56.5%) and '-Proteobacteria (16.5%, 16.5%). Low-sediment disturbance leaf packs contained a higher proportion of &-Proteobacteria (38.6% vs. 19.0%; p=0.75) than high-sediment disturbance leaf packs, and high-sediment disturbance leaf packs contained a higher, although not significant (p=0.854), proportion of (-Proteobacteria (7.4% ! 143 vs. 2.9%) than low-sediment disturbance leaf packs. Approximately three-fourths (73.9%) of (-Proteobacteria in high-sediment disturbance leaf packs came from the family Geobacteraceae. Folowing 64 days instream incubation, low-sediment disturbance leaf packs contained taxa that affiliated with the phyla Proteobacteria (79.5%), Acidobacteria (10.0%), CFB group (5.2%), Firmicutes (1.8%), and Actinobacteria (0.9%). The taxa that affiliated within the phylum Proteobacteria could be further classified at the class level with 46.0% #-Proteobacteria, 45.2% &- Proteobacteria, 5.1% '-Proteobacteria, and 3.2% (-Proteobacteria. After 64 days in stream, high-sediment disturbance leaf packs contained similar proportions of taxa that affiliated with the phyla Proteobacteria (74.9%), Actinobacteria (3.1%), and CFB group (3.7%) to those of low-sediment disturbance leaf packs. The proportion of taxa within the phylum Firmicutes droped to 4.0%, compared to 25.1% after 32 days (p=0.16). The Proteobacteria taxa identified within the high- sediment disturbance leaf packs, similar to the low-sediment disturbance leaf packs, consisted mostly of #-Proteobacteria (34.7%), &-Proteobacteria (41.6%), and '-Proteobacteria (5.8%) but had a significantly higher proportion of representatives from the class (-Proteobacteria than low-sediment disturbance leaf packs (17.1%; p=0.040). Most of these were from the suborders Sorangineae (25.8%) and Cystobacterineae (24.0%) and the family Geobacteraceae (25.6%). Overal, low-sediment disturbance leaf packs contained a significantly higher proportion of taxa that affiliated with the phylum Acidobacteria than high- ! 144 sediment disturbance leaf packs (p=0.041). The relative abundance of Acidobacteria taxa in leaf packs was significantly negatively correlated to streamwater pH at the #=0.10 level (p=0.058, r=-0.61). High-sediment disturbance leaf packs tended to have a higher relative abundance of Firmicutes taxa than low-sediment disturbance leaf packs, and the relative abundance of Firmicutes taxa was significantly positively correlated to the amount of sediment in leaf packs (p=0.049, r=0.67). For both low- and high-sediment disturbance leaf packs on day 64, there was a decrease (p=0.38), in the proportion of representatives from '-Proteobacteria, particularly in the family Pseudomonadaceae, which was concomitantly associated with an increase in the proportion of #- and &-Proteobacteria in low-sediment and &- and (- Proteobacteria in high-sediment disturbance leaf packs. For both disturbance and time treatments, comon #-Proteobacteria taxa came from the orders Caulobacterales, Rhizobiales, Rhodospirilales, and Sphingomonadales. Common &-Proteobacteria taxa for both disturbance and time treatments were members of the order Burkholderiales including the families Oxalobacteraceae and Comamonadaceae. For the leaf pack bacterial assemblages that were assessed using pyrosequencing, a global RDA test using al physicochemical variables was significant (p=0.017). Folowing this, forward stepwise regression was performed using six physicochemical variables as explanatory variables (site disturbance intensity, amount of sediment in pack, water temperature, pH, depth, and current velocity). This analysis resulted in the selection of pH (p=0.02) as the significant ! 145 explanatory variable, explaining 31.17% of the variance in bacterial comunity composition. A Mantel correlation betwen bacterial assemblage distance and the environmental distance of al six physicochemical variables was significant (p=0.03; r=0.47). If only pH was examined in this maner, the correlation betwen bacterial assemblage and the environmental distance matrix was significant (p=0.02, r=0.72). For the sites and samples examined via pyrosequencing, streamwater pH was also significantly correlated to disturbance intensity (p<0.015; r=0.7), and disturbance intensity was significantly correlated to sediment (p=0.035; r=0.70). E. DISCUSSION In this study the breakdown of maple liter, a medium-degrading leaf species, was significantly decreased in the presence of high-sediment disturbance. Given the smaler particle size of these streams (0.56-0.89 m) and the increased sediment in high-sediment disturbance leaf packs, it is possible that the decreased breakdown of maple was due to burial by sediment, which has ben sugested in other leaf breakdown studies (Bun, 198; Benfield et al., 201). The lack of a similar efect of increased sedimentation on oak breakdown could be due to its chemical structure that results in a slower breakdown rate. It is possible that a greater efect of sedimentation would have ben observed for oak leaf liter if this study had ben conducted for a longer period of time that captured the ful range of breakdown for oak. ! 146 Increased sediment disturbance and sediment within leaf packs led to was associated with decreases in many macroinvertebrate metrics (taxa richness, macroinvertebrate abundance and density, Shanon?s diversity, and biomass), which also could have contributed to decreased leaf breakdown, especialy given the large role of macroinvertebrates in leaf breakdown (Cumins et al., 1973). Similar to previous studies involving sediment disturbance (Hagen et al., 2006; Jones et al., 201), the leaf packs incubated in high-sediment disturbance sites had decreased macroinvertebrate abundance and density. And, although not significant for both leaf species, there was an overal trend towards decreased shreder abundance and density in high-sediment disturbance leaf packs. As the amount of sediment in leaf packs increased, decreases in macroinvertebrate taxa richness and diversity (H?) were observed, a trend that has also generaly ben observed in streams with increased sedimentation (Jones et al., 201). Taken together with the decrease in leaf breakdown, the macroinvertebrate assemblage characteristics observed in this study indicate that our low- and high-sediment disturbance sites were similar to those typicaly observed by other sediment disturbance studies (Jones et al., 2011; Maloney et al., 201; Sponseler and Benfield, 201). Effects of increased sediment disturbance on leaf liter bacterial assemblages were observed folowing one month of instream incubation for both medium- (maple) and slow-degrading (oak) leaf species where bacterial assemblage similarity betwen low- and high-sediment disturbance leaf packs was decreased. This sugests that sediment disturbance can alter bacterial ! 147 assemblage composition during the later periods of bacterial succession on leaf liter. Maple leaves demonstrated an efect of high-sediment disturbance on both leaf breakdown rate and the associated bacterial assemblage, and both later time points were selected for bacterial assemblage characterization via pyrosequencing. Although no significant diferences were found in measures of alpha diversity betwen low- and high-sediment disturbance bacterial assemblages, significant diferences in bacterial assemblage composition (beta diversity) were observed between low- and high-sediment disturbance sites, further indicating that high-sediment disturbance alters bacterial assemblage composition during later stages of breakdown. Low-sediment disturbance leaf packs shared similarity with findings of previous studies, being dominated by mostly gram-negative bacteria affiliated with the Proteobacteria and Acidobacteria phyla (Das et al., 206; Suberkrop and Klug, 1976). Both disturbance treatments revealed decreases over time in the relative abundance of taxa afiliated with the '-Proteobacteria (particularly Pseudomonas spp.), with members of this genus often described as being oportunistic, r- strategists due to their relatively fast growth (Juteau et al., 199; Yang and Lou, 201). Along with this decrease, there were increases in the relative abundance of many #- and (-Proteobacteria taxa, with al comon orders and suborders found (e.g. Caulobacterales, Rhizobiales, Rhodospirilales, and Sphingomonadales, Sorangineae, Cystobacterineae, Geobacteraceae) being eficient users of environmental resources and typicaly considered K-strategists (as compared to other bacteria such as Pseudomonas spp.) due to their slower ! 148 colonization and growth rates and higher energy investment in maintenance rather than reproduction (Mikkonen, 201; Bastian et al., 209). The K-strategists colonizing leaf packs difered in relative abundance betwen low- and high- sediment disturbance leaf packs. Low-sediment disturbance leaf packs saw increases in #-Proteobacteria orders, whereas high-sediment disturbance leaf packs had increases in (-Proteobacteria suborders. High-sediment disturbance leaf packs had a significantly diferent bacterial assemblage composition from that of low-sediment disturbance leaf packs folowing just one month of instream incubation. These high-sediment disturbance leaf packs had high relative abundances of taxa within the phyla Firmicutes (93% from the class Clostridia, mostly of the family Clostridiaceae), which are endospore-forming heterotrophs and obligate anaerobes (Griebler and Lueders, 209), and probably K-strategist taxa within the (-Proteobacteria (Geobacter spp. [order Desulfuromonadales] and members of the suborders Sorangineae and Cystobacterineae [both from the order Myxococcales]). All of these taxa are capable of surviving either as endospores or as cels with lower metabolic activity in environments where nutrients are exhausted and oxygen is absent (Garrity et al., 205). These results point towards a shift under high- sediment disturbance conditions to a bacterial assemblage that is beter capable of withstanding harsh environmental conditions, particularly low oxygen. This efect of sedimentation on oxygen availability has also ben postulated by observations made by Sponseler and Benfield (201). An alteration in bacterial assemblage composition via environmental conditions could consequently alter ! 149 the bacterial metabolic activity within leaf packs. Navel et al. (2010), in an in vitro study, showed fine sediment deposition led to decreases with depth in both oxygen concentration and the percentage of active bacteria, leading to a 30% decrease in leaf breakdown in their study. However, it is not known in this study exactly what percentage of decreased breakdown was due to altered bacterial assemblage composition alone, given the diferences in macroinvertebrate metrics and burial of leaf packs in high-sediment disturbance leaf packs both of which can drasticaly impact leaf breakdown. A major diference betwen high- and low-sediment disturbance leaf pack bacterial assemblages that occurred regardless of time was an increased relative abundance of taxa affiliated with the phylum Acidobacteria in low-sediment disturbance leaf packs. This high relative abundance of Acidobacteria in low- sediment disturbance leaf packs is surprising given that Acidobacteria are comonly found in terrestrial sediments (Rape and Giovanoni, 203) and would therefore be predicted to be higher in high-sediment disturbance leaf packs. However, it is possible that the relative abundance of Acidobacteria taxa is increased in low-sediment disturbance leaf packs due to the decreased pH found at low-sediment disturbance sites, especialy since Acidobacteria taxa have ben shown to thrive in lower pH environments (Jones et al., 209; Sait et al., 206), and were also significantly negatively correlated with streamwater pH in this study. Previous studies on these sites have also found a significant correlation betwen streamwater pH and disturbance intensity via land use and increased sedimentation (Houser et al., 206). ! 150 The results of this study agree with the r/K selection continum (MacArthur and Wilson, 1967; Andrews and Harris, 1986) with observed decreases over time in Pseudomonas spp. (r-strategists) and increases in many taxa considered to be K-strategists and endospore-formers. For bacterial assemblages in high-sediment leaf packs, changes in the instream environment associated with high-sediment disturbance (i.e. increased pH, leaf pack burial via sedimentation) was associated with a shift towards a leaf pack bacterial assemblage dominated by taxa capable of surviving under harsher environmental conditions (i.e. low oxygen, high pH) (Figure 5.9). The results of this study ilustrate that high-sediment disturbance can potentially lead to long- term efects on leaf liter bacterial assemblage composition through its efects on the instream environment, particularly pH. ! 151 F. LITERATURE CITED Abelho, M. 201. From literfal to breakdown in streams: a review. The Scientific World Journal, 1, 656-680. Al Shaw, E. and Richardson, J.S. 201. Direct and indirect efects of sediment pulse duration on stream invertebrate assemblages and rainbow trout Oncorhynchus mykiss. growth and survival. Canadian Journal of Fisheries and Aquatic Sciences, 58, 2213-2221. Allan, J.D. 204. Landscapes and riverscapes: the influence of land use on stream ecosystems. Annual Review of Ecology, Evolution, and Systematics, 35, 257-284. Andrews, J.H. and Harris, R.F. 200. The ecology and biogeography of microorganisms on plant surfaces. Annual Review of Phytopathology, 38, 145-180. Ausubel, F.M. 194. Preparation of plant DNA using CTAB. Current Protocols in Molecular Biology, Supl 27, 2.3-2.3.7. Barbour, M.T., Gerritsen, J., Snyder, B.D. and Stribling, J.B. 199. Rapid bioassessment protocols for use in streams and wadeable rivers. USEPA, Washington. Bastian, F., Bouziri, L., Nicolardot, B. and Ranjard, L. 209. Impact of wheat straw decomposition on succesional paterns of soil microbial comunity structure. Soil Biology and Biochemistry, 41, 262-275. ! 152 Benfield, E.F., Webster, J.R., Tank, J.L. and Hutchens, J.J. 201. Long Term Paterns in Leaf Breakdown in Streams in Response to Watershed Logging. International Review of Hydrobiology, 86, 467-474. Benke, A.C., Huryn, A.D., Smock, L.A. and Walace, J.B. 199. Length-mass relationships for freshwater macroinvertebrates in North America with particular reference to the southeastern United States. Journal of the North American Benthological Society, 18, 308-343. Boulton, A.J. and Boon, P.I. 191. A review of methodology used to measure leaf litter decomposition in lotic environments: time to turn over an old leaf? Marine and Freshwater Research, 42, 1-43. Bray, J.R. and Curtis, J.T. 1957. An ordination of the upland forest comunities of southern Wisconsin. Ecological monographs, 27, 325-349. Bun, S.E. 198. Processing of leaf liter in two northern jarrah forest streams, Western Australia: I. The role of macroinvertebrates and the influence of soluble polyphenols and inorganic sediment. Hydrobiologia, 162, 211-223. Caporaso, J.G., Bitinger, K., Bushman, F.D., Desantis, T.Z., Andersen, G.L. and Knight, R. 209. PyNAST: a flexible tol for aligning sequences to a template alignment. Bioinformatics, 26, 266. Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bitinger, K., Bushman, F.D., Costelo, E.K., Fierer, N., Pe?a, A.G., Godrich, J.K. and Gordon, J.I. 2010. QIME alows analysis of high-throughput comunity sequencing data. Nature methods, 7, 335-336. ! 153 Cardinale, M., Bruseti, L., Quatrini, P., Borin, S., Puglia, A.M., Rizzi, A., Zanardini, E., Sorlini, C., Corseli, C. and Dafonchio, D. 204. Comparison of Diferent Primer Sets for Use in Automated Ribosomal Intergenic Spacer Analysis of Complex Bacterial Comunities. Applied and Environmental Microbiology, 70, 6147-6156. Clarke, K. R. 193. Non-parametric multivariate analysis of changes in comunity structure. Australian Journal of Ecology 18, 117-143. Cubaka Kabagale, A., Cornu, B., Van Vliet, F., Meyer, C.L., Mergeay, M., Lumbu Simbi, J.B., Droogmans, L., Vander Wauven, C. and Verbrugen, N. 2010. Diversity of endophytic bacteria from the cuprophytes Haumaniastrum katangense and Crepidorhopalon tenuis. Plant and soil, 334, 461-474. Cumins, K.W., Petersen, R.C., Howard, F.O., Wuycheck, J.C. and Holt, V.I. 1973. The utilization of leaf liter by stream detritivores. Ecology, 336-345. Dangles, O., Gessner, M.O., Guerold, F. and Chauvet, E. 204. Impacts of stream acidification on liter breakdown: implications for assessing ecosystem functioning. Journal of Applied Ecology, 41, 365-378. Das, M., Royer, T.V. and Lef, L.G. 206. Fungal, bacterial and actinomycete diversity on leaves decomposing in a stream. Applied and Environmental Microbiology, 73, 756-767. Downes, B.J., Lake, P.S., Glaister, A. and Bond, N.R. 206. Effects of sand sedimentation on the macroinvertebrate fauna of lowland streams: are the efects consistent? Freshwater biology, 51, 144-160. ! 154 Fisher, M.. and Triplet, E.W. 199. Automated Approach for Ribosomal Intergenic Spacer Analysis of Microbial Diversity and Its Application to Freshwater Bacterial Comunities. Applied and Environmental Microbiology, 65, 4630-4636. Fisher, S.G. and Likens, G.E. 1973. Energy flow in Bear Brook, New Hampshire: an integrative aproach to stream ecosystem etabolism. Ecological monographs, 43, 421-439. Garrity, G.M., Brener, D.J. and Krieg, N.R. 205. Bergey's manual of systematic bacteriology, Springer. Gore, J.A. 1985. Mechanisms of colonization and habitat enhancement for benthic macroinvertebrates in restored river chanels. The Restoration of Rivers and Streams: Theories and Experience, Buterworth Publishers, Boston Massachusets. 1985. p 81-101, 7 fig, 80 ref. Griebler, C. and Lueders, T. 209. Microbial biodiversity in groundwater ecosystems. Freshwater biology, 54, 649-677. Hagen, E.M., Webster, J.R. and Benfield, E.F. 206. Are leaf breakdown rates a useful measure of stream integrity along an agricultural landuse gradient? Journal of the North American Benthological Society, 25, 330-343. Henley, W.F., Paterson, M.A., Neves, R.J. and Lemly, A.D. 200. Effects of sedimentation and turbidity on lotic fod webs: a concise review for natural resource managers. Reviews in Fisheries Science, 8, 125-139. ! 155 Hieber, M. and Gessner, M.O. 202. Contribution of stream detrivores, fungi, and bacteria to leaf breakdown based on biomass estimates. Ecology, 83, 1026-1038. Hil, B.H., Hal, R.K., Husby, P., Herlihy, A.T. and Dune, M. 200. Interregional comparisons of sediment microbial respiration in streams. Freshwater biology, 44, 213-222. Houser, J.N., Mulholand, P.J. and Maloney, K.O. 205. Catchment disturbance and stream etabolism: paterns in ecosystem respiration and gross primary production along a gradient of upland soil and vegetation disturbance. Journal of the North American Benthological Society, 24, 538-552. Houser, J.N., Mulholand, P.J. and Maloney, K.O. 206. Upland disturbance afects headwater stream nutrients and suspended sediments during baseflow and stormflow. Journal of Environmental Quality, 35, 352-365. Jones, J.I., Murphy, J.F., Colins, A.L., Sear, D.A., Naden, P.S. and Armitage, P.D. 201. The impact of fine sediment on macro invertebrates. River Research and Applications. DOI: 10.102/rra.1516 Jones, R.T., Robeson, M.S., Lauber, C.L., Hamady, M., Knight, R. and Fierer, N. 209. A comprehensive survey of soil acidobacterial diversity using pyrosequencing and clone library analyses. The ISME journal, 3, 442-453. Joseph, S.J., Hugenholtz, P., Sangwan, P., Osborne, C.A. and Janssen, P.H. 203. Laboratory cultivation of widespread and previously uncultured soil bacteria. Applied and Environmental Microbiology, 69, 7210-7215. ! 156 Juteau, P., Rho, D., Larocque, R. and Leduy, A. 199. Analysis of the relative abundance of diferent types of bacteria capable of toluene degradation in a compost biofilter. Applied microbiology and biotechnology, 52, 863-868. Kim, S.Y., Yoo, K.S., Kim, J.E., Kim, J.S., Jung, J.Y., Jin, Q., Eom, H.J. and Han, N.S. 2010. Diversity analysis of lactic acid bacteria in Korean rice wines by culture-independent method using PCR-denaturing gradient gel electrophoresis. Fod Science and Biotechnology, 19, 749-755. Kruskal, W.H. and Walis, W.A. 1952. Use of ranks in one-criterion variance analysis. Journal of the American statistical Association, 47, 583-621. Legendre, P. and Galagher, E.D. 201. Ecologicaly meaningful transformations for ordination of species data. Oecologia, 129, 271-280. Lemly, A.D. 1982. Modification of benthic insect comunities in poluted streams: combined efects of sedimentation and nutrient enrichment. Hydrobiologia, 87, 229-245. Lockaby, B.G., Governo, R., Schiling, E., Cavalcanti, G. and Hartsfield, C. 205. Effects of Sedimentation on Soil Nutrient Dynamics in Riparian Forests. Journal of Environmental Quality, 34, 390-396. Macarthur, R.H. and Wilson, E.O. 1967. The theory of island biogeography, Princeton Univ Pr. Maloney, K., Mulholand, P. and Feminela, J. 205. Influence of Catchment- Scale Military Land Use on Stream Physical and Organic Mater Variables in Smal Southeastern Plains Catchments USA. Environmental Management, 35, 677-691. ! 157 Maloney, K.O. and Feminela, J.W. 206. Evaluation of single- and multi-metric benthic macroinvertebrate indicators of catchment disturbance over time at the Fort Bening Military Instalation, Georgia, USA. Ecological Indicators, 6, 469-484. Maloney, K.O., Munguia, P. and Mitchel, R.M. 201. Anthropogenic disturbance and landscape paterns afect diversity paterns of aquatic benthic macroinvertebrates. Journal of the North American Benthological Society, 30, 284-295. Maloney, K.O. and Weler, D.E. 201. Anthropogenic disturbance and streams: land use and land-use change afect stream ecosystems via multiple pathways. Freshwater biology, 56, 611-626. Merrit, R.W. and Cumins, K.W. 196. An introduction to the aquatic insects of North America, Kendal Hunt. Mikkonen, A., Lapi, K., Walenius, K., Lindstr?m, K. and Suominen, L. 201. Ecological inference on bacterial succession using curve based comunity fingerprint data analysis, demonstrated with rhizoremediation experiment. FEMS Microbiology Ecology. DOI: 10.1111/j.1574- 6941.2011.01187.x Naiman, R.J., Decamps, H. and Mcclain, M.E. 205. Riparia: ecology, conservation, and management of streamside comunities, Academic Press. Navel, S., Mermilod-Blondin, F., Montuele, B., Chauvet, E., Simon, L. and Marmonier, P. 2010. Water-Sediment Exchanges Control Microbial ! 158 Processes Associated with Leaf Liter Degradation in the Hyporheic Zone: a Microcosm Study. Microbial Ecology, 61, 968-979. Newcombe, C.P. and Macdonald, D.. 191. Effects of Suspended Sediments on Aquatic Ecosystems. North American Journal of Fisheries Management, 11, 72-82. Nossa, C.W., Oberdorf, W.E., Yang, L., Aas, J.A., Paster, B.J., Desantis, T.Z., Brodie, E.L., Malamud, D., Poles, M.A. and Pei, Z. 208. Design of 16S rRNA gene primers for 454 pyrosequencing of the human foregut microbiome. World Journal of Gastroenterology: WJG, 16, 4135-4144. Petersen, R.C. and Cumins, K.W. 1974. Leaf processing in a wodland stream. Freshwater biology, 4, 343-368. Rape, M.S. and Giovanoni, S.J. 203. The uncultured microbial majority. Annual Reviews in Microbiology, 57, 369-394. Reder, J. and Knight, R. 2010. Rapidly denoising pyrosequencing amplicon reads by exploiting rank-abundance distributions. Nature methods, 7, 668- 669. Reice, S.R. 1974. Environmental patchiness and the breakdown of leaf liter in a wodland stream. Ecology, 55, 1271-1282. Resh, V.H., Brown, A.V., Covich, A.P., Gurtz, M.E., Li, H.W., Minshal, G.W., Reice, S.R., Sheldon, A.L., Walace, J.B. and Wissmar, R.C. 198. The role of disturbance in stream ecology. Journal of the North American Benthological Society, 7, 433-455. ! 159 Ryan, P.A. 191. Environmental efects of sediment on New Zealand streams: a review. New Zealand Journal of Marine and Freshwater Research, 25, 207-221. Sait, M., Davis, K.E.R. and Jansen, P.H. 206. Efect of pH on isolation and distribution of members of subdivision 1 of the phylum Acidobacteria occurring in soil. Applied and Environmental Microbiology, 72, 1852-1857. Saito, A., Ikeda, S., Ezura, H. and Minamisawa, K. 207. Microbial comunity analysis of the phytosphere using culture-independent methodologies. Microbes and Environments, 22, 93-105. Santmire, J.A. and Lef, L.G. 207. The influence of stream sediment particle size on bacterial abundance and comunity composition. Aquatic Ecology, 41, 153-160. Sponseler, R.A. and Benfield, E.F. 201. Influences of land use on leaf breakdown in southern Appalachian headwater streams: a multiple-scale analysis. Journal of the North American Benthological Society, 20, 44-59. Suberkrop, K. and Klug, M.J. 1976. Fungi and bacteria associated with leaves during processing in a wodland stream. Ecology, 57, 707-719. Suzuki, M.T. and Giovanoni, S.J. 196. Bias caused by template annealing in the amplification of mixtures of 16S rRNA genes by PCR. Applied and Environmental Microbiology, 62, 625-630. Wang, Q., Garrity, G.M., Tiedje, J.M. and Cole, J.R. 207. Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Applied and Environmental Microbiology, 73, 5261-5267. ! 160 Webster, J.R. and Benfield, E.F. 1986. Vascular plant breakdown in freshwater ecosystems. Annual Review of Ecology and Systematics, 17, 567-594. Webster, J.R. and Waide, J.B. 1982. Effects of forest clearcuting on leaf breakdown in a southern Appalachian stream. Freshwater biology, 12, 331-344. Witkamp, M. 1963. Microbial populations of leaf liter in relation to environmental conditions and decomposition. Ecology, 44, 370-377. Witkamp, M. 196. Decomposition of leaf liter in relation to environment, microflora, and microbial respiration. Ecology, 194-201. Zar, J.H. 199. Biostatistical analysis, Prentice hal Uper Sadle River, NJ. ! 161 D i st u rb a n ce i n t e n si t y 3.67 4.63 8.38 10.46 13.56 14.66 St re a m o rd e r 2 2 3 2 1 2 M i l i t a ry l a n d u se I n f a n t ry / ra n g e r I n f a n t ry / ra n g e r H e a v y m a ch i n e ry I n f a n t ry / ra n g e r/ i m p a ct H e a v y m a ch i n e ry H e a v y m a ch i n e ry UTM 0715377N, 3597908E 0720701N, 3600036E 0716349N, 3585850E 0710893N, 3588286E 0716005N, 3584889E 0714935N, 3585249E Ab b re v i a t i on LC KM SB1 BC SB2 SB3 M i l i t a ry co m p a rt m e n t K1 3 K1 1 F3 D12 F1 D6 Tab l e 5. 1. St u d y st re a m s a t F o rt Be n i n g M i l i t a ry I n st a l a t i o n (F BM I ) i d e n t i f i e d b y t h e i r m i l i t a ry co m p a rt m e n t , U T M co o rd i n a t e s, p re d o m i n a n t l a n d u se p ra ct i ce s, a n d d i st u rb a n ce i n d e x v a l u e (p ro p o rt i o n o f w a t e rsh e d a s b a re g ro u n d a n d ro a d co v e r) . D i st u rb a n ce i n t e n si t y v a l u e s f ro m M a l o n e y et al. (2 0 5 ) . St re a m L o i s C re e k Ki n g s M i l C re e k T ri b u t a ry Sa l y Bra n ch T ri b u t a ry Bo n h a m T ri b u t a ry Sa l y Bra n ch T ri b u t a ry Sa l y Bra n ch T ri b u t a ry ! 162 Table 5.2. MID barcodes used to tag each PCR product during pyrosequencing of bacterial assemblage. Primer Name MID Barcode MID 29 ATCAGACACG MID 30 TGATACGTCT MID 31 CGTCTAGTAC MID 32 TACTCTCGTG MID 33 ACTGTACAGT MID 34 CAGTAGACGT MID 35 TAGTGTAGAT MID 36 ATATCGCGAG MID 37 TACTGAGCTA MID 38 TCTACGTAGC MID 39 TAGAGACGAG MID 40 AGACTATACT ! 163 SB3 0.07 ? 0.01 0.10 ? 0.01 5.31 10.18 ? 0.27 SB2 0.11 ? 0.01 0.08 ? 0.01 4.22 10.54 ? 0.28 H i g h Se d i m e n t BC 0.05 ? 0.01 0.13 ? 0.01 3.92 10.35 ? 0.29 SB1 0.18 ? 0.01 0.11 ? 0.01 5.13 10.09 ? 0.33 LC 0.06 ? 0.01 0.15 ? 0.01 4.04 10.11 ? 0.30 L o w Se d i m e n t KM 0.10 ? 0. 01 0.12 ? 0.01 4.33 10.03 ? 0.31 Tab l e 5. 3. M e a n ? 1 SE cu rre n t v e l o ci t y , d e p t h , a n d t e m p e ra t u re f o r e a ch si t e a n d se d i m e n t d i st u rb a n ce t re a t m e n t . Va r i a b l e Si t e C u rre n t v e l o ci t y (m / s) D e p t h (m ) pH* T e m p e ra t u re (? C ) * D u p l i ca t e v a l u e s w e re n o t re co rd e d f o r st re a m w a t e r p H . ! 164 Figure 5.1. Leaf liter ash-free dry mass (AFDM) remaining (%) over time from low (solid) and high (holow) sediment disturbance sites for both maple (circles) and oak (triangles). Ploted points are means (? 1SE). ! 165 p 0.006 0.012 0.004 0.004 0.002 0.009 0.015 0.914 0.928 High 10.78 ? 7.63 2.73 ? 1.89 3.58 ? 1.50 0.80 ? 0.26 1.62 ? 1.28 3.45 ? 2.45 0.86 ? 0.61 65.66 ? 4.92 15.50 ? 3.88 Oak Low 23. 46 ? 11.42 8.26 ? 6.00 6.42 ? 1.84 1.26 ? 0.23 5.64 ? 3.14 7.48 ? 3.87 2.70 ? 1.99 68.32 ? 4.05 12.05 ? 5.21 p 0.024 0.211 0.014 0.01 0.001 0.015 0.172 0.876 0.999 High 17.11 ? 10.81 5.43 ? 3.85 4.56 ? 1.80 0.98 ? 0.24 3.05 ? 2.32 7.90 ? 5.21 2. 54 ? 1.88 64.47 ? 3.98 13.50 ? 3.66 Map l e Low 26.56 ? 11.68 7.14 ? 4.85 7.10 ? 2.08 1.40 ? 0.24 10.74 ? 3.84 14.04 ? 6.90 3.93 ? 2.81 64.14 ? 2.04 15.82 ? 1.54 Tab l e 5. 4. M a cro i n v e rt e b ra t e m e t ri cs (m e a n ? 1 SE) f o r e a ch l e a f sp e ci e s a n d d i st u rb a n ce t re a t m e n t co m b i n a t i o n d u ri n g o u r 6 4 - d a y i n cu b a t i o n . Bo l d e d v a l u e s re p re se n t si g n i f i ca n t d i f e r e n ce s ( P < 0 . 0 5 ) between low - and high - se d i m e n t d i st u rb a n ce si t e s. M a cro i n v e rt e b ra t e a b u n d a n ce Sh re d e r a b u n d a n ce R i ch n e ss Sh a n o n ? s d i v e rsi t y (H ? ) Bi o m a ss (m g ) M a cro i n v e rt e b ra t e d e n si t y (i n d . / g AF D M ) Sh re d e r d e n si t y (i n d . / g AF D M ) % C h i ro n o m i d a e % EPT ! 166 Figure 5.2. Diversity of macroinvertebrates (H?) and associated sediment (grams) in maple (solid circles) and oak (holow circles) leaf packs. Spearman?s rank correlations were used to describe the macroinvertebrate diversity to sediment relationships. Trend lines shown indicate significant relationships (p<0.05). r = -0.61 p = 0.05 r = -0.57 p = 0.09 ! 167 Figure 5.3. Macroinvertebrate taxon richness (S) and associated sediment (grams) in maple (solid circles) and oak (holow circles) leaf packs. Spearman?s rank correlations were used to describe the macroinvertebrate taxa richness to sediment relationships. Trend lines shown indicate significant relationships (p<0.05). r = -0.58 p = 0.08 r = -0.64 p = 0.03 ! 168 Figure 5.4. Nonmetric multi-dimensional scaling (NMDS) plots based on Bray- Curtis similarities betwen maple leaf liter samples from low- (!) and high- ()) sediment disturbance sites for days 8, 16, 32, and 64. ! 169 Figure 5.5. Nonmetric multi-dimensional scaling (NMDS) plots based on Bray- Curtis similarities betwen oak leaf liter samples from low- (!) and high- ()) sediment disturbance sites for days 8, 16, 32, and 64. ! 170 High 6.2 ? 0.4 255.9 ? 2.3 100.9 ? 13.9 Day 64 Low 5.6 ? 0.5 169.4 ? 27.7 77 ? 15.4 High 5.7 ? 0.1 156.8 ? 12.3 76.2 ? 0.9 Day 32 Low 5.8 ? 0.6 182.1 ? 62.5 81.5 ? 18.7 Tab l e 5.5 . M e a n ? 1 SE d i v e rsi t y , ri ch n e ss, a n d O T U e st i m a t e s f o r e a ch t i m e a n d se d i m e n t d i st u rb a n ce t re a t m e n t sa m p l e d d u ri n g p y ro se q u e n ci n g . Index a Sh a n o n (H ? ) Chao1 OTUs a I n d e x v a l u e s a re n o rm a l i ze d b a se d o n t re a t m e n t w i t h sm a l e st n u m b e r o f se q u e n ce s (D a y 6 4 , H i g h ). ! 171 Figure 5.6. Nonmetric multi-dimensional scaling (NMDS) plots derived from Bray-Curtis dissimilarity values betwen maple leaf liter samples from pyrosequenced low- (!) and high- ()) sediment disturbance sites for days 32 and 64. Stress level = 0.081. ! 172 Figure 5.7. Comparison of relative abundance of the most comon bacterial phyla found in pyrosequenced maple leaf pack samples grouped by days in stream and sediment disturbance (high vs. low). ! 173 Figure 5.8. Comparison of relative abundance of Proteobacteria classes found in pyrosequenced maple leaf pack samples grouped by days in stream and sediment disturbance (high vs. low). ! 174 Figure 5.9. Conceptual model ilustrating the predicted shift in leaf pack bacterial assemblage composition under increased sedimentation and catchment disturbance intensity (see Maloney et al., 205 for disturbance intensity estimation) toward a bacterial assemblage dominated by taxa capable of surviving harsher environmental conditions (e.g. increased streamwater pH and decreased dissolved oxygen).