DISRUPTION OF EMBRYONIC DEVELOPMENT IN CHANNEL CATFISH, 
ICTALURUS PUNCTATUS, USING ?STERILE-FERAL? GENE CONSTRUCTS 
 
Except where reference is made to the work of others, the work described in this thesis 
is my own or was done in collaboration with my advisory committee.  
This thesis does not include proprietary or classified information. 
 
 
______________________________________ 
Christopher Michael Templeton 
 
 
Certificate of Approval: 
 
___________________________ ___________________________ 
Ronald P. Phelps       Rex A. Dunham, Chair 
Associate Professor       Alumni Professor 
Fisheries and Allied Aquacultures Fisheries and Allied Aquacultures 
 
 
 
___________________________ ___________________________ 
Covadonga R. Arias                                                       Stephen L. McFarland 
Assistant Professor  Dean 
Fisheries and Allied Aquacultures  Graduate School
 
DISRUPTION OF EMBRYONIC DEVELOPMENT IN CHANNEL CATFISH, 
ICTALURUS PUNCTATUS, USING ?STERILE-FERAL? GENE CONSTRUCTS 
 
 
Christopher Michael Templeton 
 
 
 
A Thesis 
Submitted to 
the Graduate Faculty of 
Auburn University  
in Partial Fulfillment of the  
Requirements for the  
Degree of 
Master of Science 
 
 
Auburn, Alabama 
August 8, 2005 
 iii
DISRUPTION OF EMBRYONIC DEVELOPMENT IN CHANNEL CATFISH, 
ICTALURUS PUNCTATUS, USING ?STERILE- FERAL? GENE CONSTRUCTS 
 
 
 
Christopher Michael Templeton 
Permission is granted to Auburn University to make copies of this thesis 
 at its discretion, upon request of individuals or institutions and at their expense.  
 The author reserves all publication rights. 
 
 
 
 
     
   ______________________________ 
    Signature of Author 
 
 
      
  ______________________________ 
          Date 
 
 
 
 
Copy sent to: 
 
Name _______________________________________ Date_____________ 
 
 iv
VITA 
 Christopher Michael Templeton, son of James Christopher and Sharon Rebecca  
(Wheelock) Templeton, was born April 15, 1976 in Kingsport, Tennessee.  He graduated 
from Sullavin North High School, Kingsport, Tennessee in 1994.  In August 1994, 
entered Maryville College, Maryville, TN and received the degree of Bachelor of Arts in 
the Biological Sciences May 1998.  He married Brooke West, daughter of Bill and 
Suzanne West, on July 8, 2000.  He and his wife had their first son, Malachi Michael 
Templeton on June 10, 2001.  He began his graduate studies at Auburn University in the 
Department of Fisheries and Allied Aquacultures, January 2004, to pursue a Master of 
Science Degree.  Their second child, Isabelle Noor Templeton, was born April 29, 2004.
 v
THESIS ABSTRACT 
DISRUPTION OF EMBRYONIC DEVELOPMENT IN CHANNEL CATFISH, 
ICTALURUS PUNCTATUS, USING ?STERILE- FERAL? GENE CONSTRUCTS 
 
Christopher Michael Templeton 
Master of Science, August 8, 2005 
(B. A., Maryville College, 1998) 
 
 
 103 Typed Pages 
Directed by Rex A. Dunham 
 Channel catfish, Ictalurus punctatus, embryos were electroporated with a sterile feral 
3 (SF3), sterile feral 4 (SF4), glutamate decarboxylase (GAD) or CAB constructs and a 
blank control.  Doxycycline was applied to some of the SF3 and SF4 replicates, at 50, 
100 and 150 ppm, from 4 hours post fertilization to first hatch.  The developing embryos 
were observed from fertilization up through hatch.  Dead and deformed embryos were 
removed during this period.  Mortality rates were computed for different time intervals 
during development as well as for the entire development period.  Deformity rates were 
also analyzed.  Objectives were to determine the efficacy of the sterile feral constructs to 
disrupt embryonic development, the optimal concentration of doxycycline to prevent 
embryonic disruption and the time interval most crucial for application of doxycycline to 
prevent the expression of the developmental disrupter. 
 vi
 Embryos electroporated with the SF3 construct consistently demonstrated a higher 
mean mortality than that of the control groups.  The overall mean mortality of the SF3 
groups for all three experiments was 35.9% higher than the overall mean mortality of the  
three blank control groups.  Similar results were obtained for the SF4 construct, whose 
overall mean mortality for the three experiments was 27.2% higher than that of the 
overall mean mortality for the three experiments with the blank control groups.  The 
GAD construct produced an overall mean mortality that was 34.5% lower than that of the 
mean mortality of SF3 and 25.6% lower than that of SF4.  The single mortality rate 
produce by the CAB construct was 82.3% lower than that of the mean mortality of SF3 
and 79.9% lower than SF4.  These consistently lower mortality rates produced by both 
CAB and GAD  demonstrated that the introduction of exogenous DNA was not 
responsible for the larger mortality rates observed in the SF4 group and SF3 groups, but 
the action of the SF3 and SF4 constructs.  The concentration of doxycycline most optimal 
for blocking the expression of the blocker in the SF3 construct was 100 ppm and 150 
ppm for blocking the developmental disrupter of the SF4 construct.  Some portion of the 
18-61 hour post fertilization time period appears to be the time interval in which it is 
most crucial to apply doxycycline to prevent embryonic disruption.   
 vii
ACKNOWLEDGEMENTS 
 
 The author would like to thank Fajin Wang for the many hours of assistance that he 
gave during a large portion of this study.  I would like to thank Yi Hahn for the assistance 
he provided in the statistical analyses.  The Aqua Bounty Technologies Inc. team also 
deserves recognition for conducting the PCR analyses of the eggs and embryos.  I would 
like to thank Dr. Cova Arias, Assistant Professor, Auburn University, and Curtis 
Phillipson, Aqua Bounty Technologies Inc., who were responsible for growing the gene 
constructs used in this study.  I would like to thank Vance Trudeau, University of Ottawa, 
for providing the GAD construct used in this study.  I would also like to thank Ron 
Thresher and CSIRO for the Sterile Feral constructs used in this study.  Also, I would like 
to thank Aqua Bounty Technologies Inc. for their support that provided my research 
assistantship throughout the entirety of this study.  
 Lastly, the author would like to thank his wonderful wife for her continued help.  The 
parents and sister of the author have also been a source of encouragement and help 
throughout this process. 
 viii
Style manual or journal used: Aquaculture
Computer software used: Windows Microsoft Word 2002, Windows Microsoft Excel 
2002, SAS version 9.1. 
 
 ix
TABLE OF CONTENTS 
LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .x 
LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .xiii 
I. INTRODUCTION . . . . .  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 
II. MATERIALS AND METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 
  Sterile Feral Constructs 
  Other Constructs 
  Brood Stock 
  Spawning 
  Gene Transfer 
  Incubation and Hatching 
  Egg and Embryo Observations 
  Water Analysis 
  PCR Analysis 
  Data Analysis 
 
III. RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25 
 Mortality Rates 
 Mortality Over Time 
 Deformities 
 
IV. DISCUSSION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 
 Disrupting Embryonic Development 
 Doxycycline Concentration 
 Critical Period of Development 
 Conclusion 
 
V. REFERENCES . . . . .  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 
VI. APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 
 
 x
 
 
 
LIST OF TABLES 
1. Treatments and number of replicates employed in each experiment. . . . . . . . . . . . . . 23 
 
2. Mean mortality of channel catfish, Ictalurus punctatus, embryos electroporated with 
Sterile Feral (SF3 and SF4) gene constructs, glutamate decarboxylase (GAD), CAB 
and a electroporated blank control.  Dead eggs and embryos were identified and 
removed beginning from 14 hours post fertilization up to 153 hours post fertilization.  
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .26 
 
3. Mean mortality of channel catfish, Ictalurus punctatus, embryos electroporated with 
the Sterile Feral (SF3) gene construct, and treated with three different concentrations 
of doxycycline (50, 100 and 150 ppm) from 4 hours post fertilization to first hatch and 
a electroporated control group.  Dead eggs and embryos were identified and removed 
beginning from 14 hours post fertilization up to 153 hours post fertilization . . . . . . .31 
 
4. Mean mortality of channel catfish, Ictalurus punctatus, embryos electroporated with 
the Sterile Feral (SF4) gene construct, and treated with three different concentrations 
of doxycycline (50, 100 and 150 ppm) from 4 hours post fertilization to first hatch and 
a electroporated control.  Dead eggs and embryos were identified and removed 
beginning from 14 hours post fertilization up to 153 hours post fertilization . . . . . . .34 
 
5. Mean mortality at various time intervals (hours) for channel catfish, Ictalurus 
punctatus, embryos electroporated with the Sterile Feral (SF3 and SF4) gene 
constructs, and a electroporated control in experiment one.  Dead eggs and embryos 
were identified and removed beginning from 14 hours post fertilization to 153 hours 
post fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .38 
 
6. Mean mortality percentages at various time intervals (hours) for channel catfish, 
Ictalurus punctatus, embryos electroporated with the Sterile Feral (SF3 and SF4) gene 
constructs, and a electroporated control in experiment two.  Dead eggs and embryos 
were identified and removed beginning from 18 hours post fertilization to 137 hours 
post fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41 
 
7. Comparison of the mean mortality percentages at various time intervals (hours) for 
channel catfish, Ictalurus punctatus, embryos electroporated with the Sterile Feral 
(SF3 and SF4) gene constructs, and a electroporated control in experiment three.  
Dead eggs and embryos were identified and removed beginning from 18 hours post 
fertilization to 137 hours post fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .44 
 xi
 
8. Comparison of the mean mortality percentages at various time intervals (hours) for 
channel catfish, Ictalurus punctatus, embryos electroporated with the Sterile Feral 
(SF3 and SF4) gene constructs, glutamate decarboxylase (GAD) and CAB in 
experiment one.  Dead eggs and embryos were identified and removed beginning from 
14 hours post fertilization to 153 hours post fertilization . . . . . . . . . . . . . . . . . . . . . . 46 
 
9. Comparison of the mean mortality percentages at various time intervals (hours) for 
channel catfish, Ictalurus punctatus, embryos electroporated with the Sterile Feral 
(SF3 and SF4) gene constructs and glutamate decarboxylase (GAD) in experiment 
two.  Dead eggs and embryos were identified and removed beginning from 18 hours 
post fertilization to 146 hours post fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49 
 
10. Comparison of the mean mortality percentages at various time intervals (hours) for  
channel catfish, Ictalurus punctatus, embryos electroporated with the Sterile Feral 
(SF3) gene construct, and treated with doxycycline (100 ppm) from 4 hours post 
fertilization to first hatch and a electroporated blank control in experiment one.  Dead 
eggs and embryos were identified and removed beginning from 14 hours post 
fertilization to 153 hours post fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 
 
11. Comparison of the mean mortality percentages at various time intervals (hours) for 
channel catfish, Ictalurus punctatus, embryos electroporated with the Sterile Feral 
(SF3) gene construct, and treated with different concentrations of doxycycline (50, 
100 and 150 ppm) from 4 hours post fertilization to first hatch and a electroporated 
control in experiment two.  Dead eggs and embryos were identified and removed 
beginning from 12 hours post fertilization to 137 hours post fertilization . . . . . . . . .55 
 
12. Comparison of the mean mortality percentages at various time intervals (hours) for 
channel catfish, Ictalurus punctatus, embryos electroporated with the Sterile Feral 
(SF4) gene construct, and treated with doxycycline (100 ppm) from 4 hours post 
fertilization to first hatch and a electroporated control in experiment one.  Dead eggs 
and embryos were identified and removed beginning from 14 hours post fertilization 
to 153 hours post fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .59 
 
13. Comparison of the mean mortality percentages at various time intervals (hours) for 
channel catfish, Ictalurus punctatus, embryos electroporated with the Sterile Feral 
(SF4) gene construct, and treated with different concentrations of doxycycline (50, 
100 and 150 ppm) from 4 hours post fertilization to first hatch and a electroporated 
control in experiment two.  Dead eggs and embryos were identified and removed 
beginning from 24 hours post fertilization to 144 hours post fertilization . . . . . . . . .62 
 
14. Mean deformity percentages of channel catfish, Ictalurus punctatus, embryos 
electroporated with Sterile Feral (SF3 and SF4) gene constructs, a electroporated 
control, glutamate decarboxylase (GAD) and CAB.  Dead eggs and embryos were 
 xii
identified and removed beginning from 14 hours post fertilization up to 153 hours 
post fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 
 
15. Mean deformity percentages of channel catfish, Ictalurus punctatus, embryos 
electroporated with Sterile Feral (SF3) gene construct, and treated with different 
concentrations of doxycycline (50, 100 and 150 ppm) from 4 hours post fertilization 
to first hatch and a electroporated control.  Dead eggs and embryos were identified 
and removed beginning from 14 hours post fertilization up to 153 hours post 
fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 
 
16. Mean deformity percentages of channel catfish, Ictalurus punctatus, embryos 
electroporated with Sterile Feral (SF4) gene construct, and treated with different 
concentrations of doxycycline (50, 100 and 150 ppm) from 4 hours post fertilization 
to first hatch and a electroporated control.  Dead eggs and embryos were identified 
and removed beginning from 14 hours post fertilization up to 153 hours post 
fertilization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .68 
 
17. Mean mortality of channel catfish, Ictalurus punctatus, embryos electroporated with 
Sterile Feral (SF3 and SF4) gene constructs, PH3, PAB and a electroporated control 
in experiment one.  Dead eggs and embryos were identified and removed beginning 
from 14 hours post fertilization up to 153 hours post fertilization. . . . . . . . . . . . . . . 84 
 
18. Mean mortality of channel catfish, Ictalurus punctatus, embryos electroporated with 
the Sterile Feral (SF3 & SF4) gene constructs treated with doxycycline (100 ppm) 
from 4 hours post fertilization to first hatch, PAB, PH3 and a electroporated control in 
experiment one.  Dead eggs and embryos were identified and removed beginning 
from 14 hours post fertilization up to 153 hours post fertilization. . . . . . . . . . . . . . . 86 
 
 
 
 
 
 
 xiii
 
 
 
 
LIST OF FIGURES 
1. Flow chart of the steps of the Sterile Feral mechanism . . . . . . . . . . . . . . . . . . . . . 9 
2. The BD? Tet-Off and Tet-On Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12 
3. Sterile Feral 3 Construct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15 
4. Sterile Feral 4 construct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 
5. Glutamate decarboxylase construct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 
6. CAB construct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19 
7. Mean mortality of channel catfish, Ictalurus punctatus, embryos electroporated 
with Sterile Feral (SF3 and SF4) gene constructs and a electroporated control in 
experiment one.  Dead eggs and embryos were identified and removed beginning 
from 14 hours post fertilization up to 153 hours post fertilization . . . . . . . . . . . .27 
 
8. Mean mortality of channel catfish, Ictalurus punctatus, embryos electroporated 
with Sterile Feral (SF3 and SF4) gene constructs and a electroporated control in 
experiment two.  Dead eggs and embryos were identified and removed beginning 
from 18 hours post fertilization up to 137 hours post fertilization . . . . . . . . . . . .28 
 
9. Mean mortality of channel catfish, Ictalurus punctatus, embryos electroporated 
with Sterile Feral (SF3 and SF4) gene constructs and a electroporated control in 
experiment three.  Dead eggs and embryos were identified and removed 
beginning from 18 hours post fertilization up to 137 hours post fertilization . . . 29 
 
10. Mean mortality of channel catfish, Ictalurus punctatus, in experiment one. 
Embryos were electroporated with the Sterile Feral (SF3) gene construct, and 
treated with doxycycline (100 ppm) from 4 hours post fertilization to first hatch 
and a electroporated control.  Dead eggs and embryos were identified and 
removed beginning from 14 hours post fertilization up to 153 hours post 
fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 
 
 xiv
11. Mean mortality of channel catfish, Ictalurus punctatus, in experiment two. 
Embryos were electroporated with the Sterile Feral (SF3) gene construct, and 
treated with three concentrations of doxycycline (50, 100, and 150 ppm) from 4 
hours post fertilization to first hatch and a electroporated control.  Dead eggs and 
embryos were identified and removed beginning from 18 hours post fertilization 
up to 137 hours post fertilization . . . . . .  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33 
 
12. Mean mortality of channel catfish, Ictalurus punctatus, in experiment one. 
Embryos were electroporated with the Sterile Feral (SF4) gene construct, and 
treated with doxycycline (100 ppm) from 4 hours post fertilization to first hatch 
and a electroporated control.  Dead eggs and embryos were identified and 
removed beginning from 14 hours post fertilization up to 153 hours post 
fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 
 
13. Mean mortality of channel catfish, Ictalurus punctatus, in experiment two. 
Embryos were electroporated with the Sterile Feral (SF4) gene construct, and 
treated with three concentrations of doxycycline (50, 100 and 150 ppm) from 4 
hours post fertilization to first hatch and a electroporated control.  Dead eggs and 
embryos were identified and removed beginning from 14 hours post fertilization 
up to 153 hours post fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 
 
14. Mean mortality at various time intervals (hours) for channel catfish, Ictalurus 
punctatus, embryos electroporated with the Sterile Feral (SF3 and SF4) gene 
constructs, and a electroporated control in experiment one.  Dead eggs and 
embryos were identified and removed beginning from 14 hours post fertilization 
to 153 hours post fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39 
 
15. Mean mortality at various time intervals (hours) for channel catfish, Ictalurus 
punctatus, embryos electroporated with the Sterile Feral (SF3 and SF4) gene 
constructs, and a electroporated blank control in experiment two.  Dead eggs and 
embryos were identified and removed beginning from 18 hours post fertilization 
to 137 hours post fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .42 
 
16. Mean mortality at various time intervals (hours) for channel catfish, Ictalurus 
punctatus, embryos electroporated with the Sterile Feral (SF3 and SF4) gene 
constructs, and a electroporated control in experiment three.  Dead eggs and 
embryos were identified and removed beginning from 18 hours post fertilization 
to 137 hours post fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45 
 
17. Comparison of the mean mortality percentages at various time intervals (hours) 
for channel catfish, Ictalurus punctatus, embryos electroporated with the Sterile 
Feral (SF3 and SF4) gene constructs, glutamate decarboxylase (GAD) and CAB 
in experiment one.  Dead eggs and embryos were identified and removed 
beginning from 14 hours post fertilization to 153 hours post fertilization . . . . . .47 
 
 xv
18. Comparison of the mean mortality percentages at various time intervals (hours)for 
channel catfish, Ictalurus punctatus, embryos electroporated with the Sterile Feral 
(SF3 and SF4) gene constructs and glutamate decarboxylase (GAD) in experiment 
two.  Dead eggs and embryos were identified and removed beginning from 18 
hours post fertilization to 146 hours post fertilization. . . . . . . . . . . . . . . . . . . . . 50 
 
19. Comparison of the mean mortality percentages at various time intervals (hours) 
for channel catfish, Ictalurus punctatus, embryos electroporated with the Sterile 
Feral (SF3) gene construct, and treated with doxycycline (100 ppm) from 4 hours 
post fertilization to first hatch and a electroporated control in experiment one.  
Dead eggs and embryos were identified and removed beginning from 14 hours 
post fertilization to 153 hours post fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . 53 
 
20. Comparison of the mean mortality percentages at various time intervals (hours) 
for channel catfish, Ictalurus punctatus, embryos electroporated with the Sterile 
Feral (SF3) gene construct, and treated with three different concentrations of 
doxycycline (50, 100 and 150 ppm) from 4 hours post fertilization to first hatch 
and a electroporated control in experiment two.  Dead eggs and embryos were 
identified and removed beginning from 14 hours post fertilization to 153 hours 
post fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 
 
21. Comparison of the mean mortality percentages at various time intervals (hours) 
for channel catfish, Ictalurus punctatus, embryos electroporated with the Sterile 
Feral (SF4) gene construct, and treated with doxycycline (100 ppm) from 4 hours 
post fertilization to first hatch and a electroporated control in experiment one.  
Dead eggs and embryos were identified and removed beginning from 14 hours 
post fertilization to 153 hours post fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . 60 
 
22. Comparison of the mean mortality percentages at various time intervals (hours) 
for channel catfish, Ictalurus punctatus, embryos electroporated with the Sterile 
Feral (SF4) gene construct, and treated with three different concentrations of 
doxycycline (50, 100 and 150 ppm) from 4 hours post fertilization to first hatch 
and a electroporated control in experiment two.  Dead eggs and embryos were 
identified and removed beginning from 18 hours post fertilization to 137 hours 
post fertilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 
 
23. Pab-sGNRH-antisense construct. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 
 
24. Mean mortality of channel catfish, Ictalurus punctatus, in experiment one. 
Embryos were electroporated with the Sterile Feral (SF3 & SF4) gene constructs, 
PAB, PH3 and a electroporated blank control.  Dead eggs and embryos were 
identified and removed beginning from 14 hours post fertilization up to 153 hours 
post fertilization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .85 
 
 xvi
25. Mean mortality of channel catfish, Ictalurus punctatus, embryos electroporated 
with the Sterile Feral (SF3 & SF4) gene constructs treated with doxycycline (100 
ppm) from 4 hours post fertilization to first hatch, PAB, PH3 and a electroporated 
control in experiment one.  Dead eggs and embryos were identified and removed 
beginning from 14 hours post fertilization up to 153 hours post fertilization. . . .87 
 
 
 
 
 
 
 
 
 
 
 1
 
 
 
INTRODUCTION 
 
 Genetically Modified Organisms (GMOs) are no longer a novelty for the 
agricultural sector of the United States or the World.  However, their application to 
commercial agriculture continues to be an issue of major concern.  A GMO or 
transgenic organism is one that has been transformed by the insertion of one or more 
transgenes (http://www.fao.org/biotech/find-formalpha-n.asp).  Transgenes are 
specific pieces of foreign DNA from another species or endogenous gene sequences 
that are re-inserted into the organism (Dunham, 2004). 
Palmiter et al. (1982) are credited with the creation of the first transgenic 
animals.  They demonstrated that increased growth could be achieved through 
insertion of growth hormone genes in mice.  This motivated researchers in the field of 
aquaculture to apply these findings to fish.  Progress in transgenic fish research was 
rapid from 1985 to 1987 with the creation of transgenic goldfish, Carassius auratus, 
in China, transgenic medaka, Oryzias latipes, in Japan, transgenic rainbow trout, 
Oncorhynchus mykiss, in France, and transgenic channel catfish, Ictalurus punctatus, 
in the U.S. (Dunham, 2004).  There have been gene transfer studies conducted in over 
35 species of fish, and half of these are significant for aquaculture (Zbikowska, 2003). 
 These advances have the potential to enable aquaculture to meet the growing 
demands of an increasing world population in the area of food production as well as 
the potential to impact the maintenance and improvement of the biodiversity of 
natural fisheries along with other natural resources (Melamed et al., 2002; Dunham, 
 2
2004).  In addition to these benefits, there is the potential to use transgenic fish in the 
area of biomedical research, and for monitoring potentially harmful chemicals in 
aquatic environments (Zbikowska, 2003).  Major advances in the field of genetic 
improvement in aquaculture, via gene transfer have been for the traits, increased 
growth rate (Devlin et al., 2004), feed conversion efficiency (Cook et al., 2000), 
disease resistance (Sarmasik et al., 2002; Dunham et al., 2002a), tolerance of low 
water quality (Dunham et al. 2002b), cold tolerance (Wang et al., 1995), body shape 
(Dunham et al. 2002b), dress-out percentage (Dunham et al. 2002b), and carcass 
quality (Dunham et al. 2002b). 
 Currently, no aquatic GMOs are commercially grown for human consumption 
(FAO/WHO, 2004).  Research populations and commercial brood stocks are 
maintained, but not marketed.  Transgenic ornamental fish, such as the GloFish?, 
have been sold to the public in the U.S. and Asia (Dunham, 2003).  The extent to 
which GMOs are incorporated into commercial aquaculture will primarily depend 
upon the risk they pose to wild populations of conspecifics and to natural ecosystems, 
as well as the ability to confine these transgenic aquatic organisms (Zbikowska, 
2003).  Government regulations currently prohibit application of transgenic aquatic 
organisms except for the small ornamental fish (Dunham, 2004).   
 The prohibition placed on commercially growing transgenic fish is mostly 
attributable to the assumption that these transgenic fish could have superior fitness 
traits as compared to those of naturally occurring populations and cause demise to 
biodiversity or aquatic ecosystems (Hallerman et al., 1993).   The fitness traits, 
reproductive performance, foraging ability, swimming ability and predator avoidance, 
are key factors in determining the ability of transgenic fish to out compete 
conspecifics.  Models evaluating the genetic risk of transgenic fish have been 
 3
developed by Hedrick (2001) and Muir and Howard (1999; 2001).  Each of these 
models demonstrated the ability of transgenic fish to negatively impact conspecifics, 
but these models have weaknesses.  These include utilization of an artificial 
environment, assumption of size related mating preferences, which do not exist for 
many species of fish, assumption of no genotype-environment interactions, no 
foraging and no predation (Dunham, 2004).   
In most actual measurements of transgenic fish, their fitness is lower than that 
of wildtype individuals (Hedrick, 2001; Dunham, 2004).  This is supported by studies 
that demonstrate transgenic fish have equal or lower rates of sexual reproduction and 
sexual maturity (Dunham et al., 1992; Chatakondi, 1995), equal foraging abilities 
(Chitmanat, 1996), inferior predator avoidance (Dunham, 1995) and equal or inferior 
swimming ability (Farrell, 1997).  Based upon the majority of the environmental risk 
data gathered on transgenic fish, in the event of an escape, ecological impacts should 
be minimal because of the overall reduced fitness of transgenic fish (Dunham and 
Devlin, 1998).  However, this alone will not be enough to change current prohibition 
of transgenic fish from commercial application.  Both physical and biological 
confinement, including genetic sterilization practices, will likely be necessary to 
ensure the absence of negative environmental impacts and government approval for 
application of these fish. 
 One method to ensure protection against negative environmental impacts is 
physical confinement.  Physical confinement cannot guarantee the complete 
elimination of escapees.    The continuous application of chlorine, rotenone or other 
approved fish toxins to a facility?s discharge is chemical confinement.  Apart from the 
obvious environmental impact of continuous application of these chemicals, this 
method is also unable to guarantee complete confinement.  Biological confinement 
 4
such as removal of escapees via predators is another alternative.  Each method of 
confinement discussed fails to provide a guarantee against environmental impact and 
also fails to address the issue of feral populations.  Sterilization is the key to reducing 
the impact that transgenic and exotic species could exert on ecosystems. 
Potentially, sterilization can be accomplished through surgery, immunology, 
radiation or with chemicals.  Sterilization can also be accomplished genetically 
through hybridization, ploidy manipulation or transgenesis.  The high labor costs and 
possibility for mortality make surgical sterilization only feasible for small numbers of 
valuable fish.  Another pitfall is gonadal regeneration.  Underwood et al. (1986) have 
demonstrated that fish have the ability to regenerate functional testes and ovaries.  
Ictalurid catfish (Bart, 1994) and Nile tilapia (Akhtar, 1984) have demonstrated the 
ability to regenerate whole or partial gonads following complete gonadectomies.  The 
combination of its inability to guarantee permanent sterility and high labor costs make 
surgical sterilization of transgenics and exotics on a commercial level unfeasible.  
Sterilization via immunology and irradiation have also failed to provide permanent 
sterilization (Dunham, 2004).  These methods, although potentially useful in specific 
situations and on a small-scale basis, all fall short of the need for a commercial scale, 
permanent, 100% effective sterilization.  
Chemical sterilization has demonstrated some potential and needs further 
investigation, but to date has proven to be at best a temporary solution (Stanley, 
1979).  Chemicals that interfere with the production, release or utilization of sex 
hormones that interrupt mating behavior or cause sex reversal have all been used.  
Those that serve as hormone antagonists have demonstrated the most potential.  
Methallibure, cyproterone, trenbolone acetate, ?-aminobutyric acid (GABA) and 
recently bisazir are all hormone antagonists that have demonstrated varying levels of 
 5
efficacy (Dunham, 2004).  A recent study by Young et al. (2004) employed sterile 
male sea lampreys, Petromyzon marinus, created using bisazir, in a study that 
evaluated this chemical?s effect on production of sex steroids in this species.  The 
study demonstrated the absence of endocrine disruption in this species when treated 
with bisazir.  More studies are needed to further evaluate chemical sterilization?s 
efficacy, effects on treated species, food safety issues and other important topics 
before chemicals such as bisazir could be commercially applied for sterilization 
purposes. 
Genetic approaches show tremendous potential to supply a 100% effective and 
reversible method of sterilization.  Monosex populations produced through hormonal 
sex reversal offer one potential method of limiting reproduction within a transgenic 
population (Pandian, 2003).  This approach is only feasible when working with exotic 
species and not with conspecifics that are present in the watershed.   Even if working 
with exotics, the escape of both sexes from a monosex/monogenetic culture unit can 
still have a short-term impact lasting up to two generations.   
Genetic sterilization via polyploidy is another option for the production of 
genetically sterilized fish stocks.  Although triploid mammals and birds are not viable, 
triploid fish are not only viable, but are also usually sterile (Chourrout et al., 1986).  
Triploid fish also have the potential for increased growth (Chourrout et al., 1986), 
increased carcass yield and increased survival (Dunham, 1996).   
Triploid induction via heat shock, cold shock, pressure shock, or through 
application of chemicals has been the method of choice, but does not always provide 
100% efficacy (Abdul-Razak et al., 1999; Dunham, 2004).  Triploidy is not 
commercially practical for all fish because it can lead to slower growth in some 
species (Dunham, 2004). 
 6
Cotter et al. (2000) demonstrated the reduced ability of triploid Atlantic 
salmon to compete with naturally occurring conspecifics.  Triploidy also decreases 
growth to the point that it negates some of the gain due to the growth hormone gene 
insertion (Abdul Razak et al., 1999; Devlin et al., 2004).  Another problem with 
sterility via polyploidy is the necessity of fertile diploid brood stock, so environmental 
risk cannot be completely eliminated.   
To date, the only genetic sterilization method that has the potential to produce 
100% effective sterilization with the potential for reversibility is transgenic 
sterilization.  Transgenic approaches are the best option for genetic sterilization and 
for confinement.  Gene constructs need to be developed that have the capability for 
sterilization or lethality that can be temporarily reversed as needed to allow culture of 
these fish. 
 To date use of fish that are transgenics, domestics, interspecific hybrids, and 
exotic species is hindered by community and scientific concerns over the impact these 
fish could have or in some cases have had on wild stocks if introduced into natural 
environments.  Reversible transgenic sterilization is the ultimate answer to virtually 
eliminate these risk issues for both aquaculture and recreational fishing.  If successful, 
this technology could be applied in many situations.  Sterilization has the potential to 
increase yields with precocious species, such as tilapia.  Exotic species that were 
transgenically sterilized would be unable to establish feral populations.  Sterile 
interspecific hybrids would be unable to backcross with parental species (Curtis et al., 
1987).  Also, declining or decimated natural stocks could be supplemented without 
concern of genetic introgression from the introduced conspecifics since they were 
transgenically sterilized (Dunham, 2004).  Lastly, the ability to transgenically sterilize 
fish transgenic for other constructs would eliminate all most all environmental risks 
 7
encountered when applying these fish to commercial aquaculture.  Impacts of non-
transgenic aquacultured fish on wild fish stocks, such as the common controversy 
with aquacultured salmon impacting natural populations could be greatly minimized 
with this technology.  This technology has the potential to even benefit recreational 
fishing by incorporating exotics and hybrids that were successfully sterilized. 
The Commonwealth Scientific Industrial Research Organization (CSIRO), 
Australia, has iniated research on transgenic sterilization and developed a method 
termed ?sterile feral.?  The method employs the insertion of a gene that can reversibly 
interrupt gamete and/or embryonic development (Dunham, 2004).  CSIRO has 
identified genes in zebrafish, Danio rerio, and Pacific oysters, Crassostrea gigas, that 
are essential to and active only during embryonic development and/or gametogenesis 
(Thresher et al., 2001).  In its most basic form, the construct contains a blocker to 
embryonic development and/or gametogenesis and a genetic switch to control it.  The 
sterile feral constructs are made of a species-specific promoter coupled with a 
repressible element, which drives expression of a blocker gene.  For construction of 
sterile feral 3 (SF3) and sterile feral 4 (SF4), a developmental stage-specific promoter 
was coupled to components derived from the commercially available Tet-Off 
controllable expression system marketed by CLONTECH ( http://clontech.com/, 
Protocol # PT3001-1, version # PR95962).  The blocker can be inactivated in 
captivity by signalling for the promoter to allow production of the repressor protein, 
which prevents transcription of the blocker and subsequent knockout (Figure 1).  The 
promoter can be triggered to produce the repressor protein through the incorporation 
of doxycycline or tetracycline into the diet or through soluble form (Dunham, 2004).     
The basis of this technology is the Tet-Off and Tet-On system developed by 
BD? Biosciences.  In its most basic form, the tetracycline repressor (tetR) binds to its 
 8
operators within the promoter region of the Tn10-specified tetracycline-resistance 
operon of E. coli and prevents transcription when tetracycline is absent from its 
environment.  In the presence of tetracycline, the tetR does not bind to its operators 
and thus allows for transcription (Gossen and Bujard, 1992).  The tetR is bound to the 
activating domain of the virion protein 16 (VP16) of herpes simplex virus (HSV).  
The resulting combination is referred to as a ?tetracycline-controlled transactivator? 
(tTA) (Gossen and Bujard, 1992).  This transactivator used to regulate the targeted 
gene is attached to a cytomegalovirus (CMV) promoter.  In the Tet-Off system, the 
tTA transactivator binds to the Tet Response Element (TRE) contained within the 
CMV promoter and activates transcription when tetracycline is absent. In the Tet-On 
system, which employs the rtTa transactivator, the Tet repressor protein with four 
amino acid changes, also binds to the TRE and induces transcription in the presence 
of tetracycline (Figure 2).  Gossen and Bujard (1992) studied this system and its be  
control over a luciferase gene.  They found that the luciferase activity could regulated 
to over five orders of magnitude depending upon the concentration of the antibiotic in 
the culture medium.  The Tet-On and Tet-Off systems have the capacity to be 
controlled at varying levels, thus it is not a choice between fully on or off.  Yin and 
Schimke (1995) found that predictable levels of gene expression could be produced 
based upon the amount of tetracycline in the culture medium.  
In the case of the SF3 or SF4 constructs, doxycycline or tetracycline act upon 
the tetracycline-responsive transcriptional activator protein (tTA), which is a fusion of 
TetR and VP16  from the pTet-Off regulatory plasmid (Thresher et al., 2001).  A 
zebrafish promoter drives expression of the tTA, and tTA regulates the tetracycline 
(Tet)-responsive human cytomegalovirus promoter (PhCMV*-1), which controls 
expression of the blocker gene.  PhCMV*-1 contains the tetracycline responsive 
 9
Figure 1.  Flow chart of the steps of the Sterile Feral mechanism. 
 
Early Developmental Promoter 
? 
Early Developmental Promoter 
? 
Regulatory Protein tTA 
? 
Regulatory Protein tTA 
? 
No Tetracycline 
? 
Add Tetracycline 
? 
tTA and TRE bind 
(tTA turns on TRE) 
? 
TTA and TRE do not bind 
 
? 
Promoter 2 on 
? 
Promoter 2 off 
? 
Blocker Gene (Knockout) 
? 
Blocker Gene (Knockout) off 
? 
Embryo Dies Embryo Lives 
 
 
 
 
 
 
 
Figure 2.  The BD? Tet-Off and Tet-On Systems (Taken from BD? Tet-Off and Tet-On Systems by BD Biosciences)
 10
 11
element (TRE), which consists of seven copies of the tet operator sequence (tetO) 
located just upstream of the minimal CMV promoter (PminCMV).  PhCMV*-1 does 
not function in the absence of binding of transactivator protein to the tetO.  The 
tetracycline-sensitive element is described by Gossen and Bujard (1992) for Tet-Off 
and Tet-On is described by Gossen et al. (1995) and Kistner et al. (1996).  In the Tet-
Off system, addition of tetracycline (Tet) or doxycycline (Dox; a Tet derivative) 
prevents the binding of tTA to the Tet-responsive element. Expression of the blocker 
gene is controlled by TRE/CMV, and is repressed until tetracycline is removed from 
the incubation water.  In the absence of tetracycline or doxycycline, the blocker gene 
is transcribed if the developmental promoter continues to express tTA, thus killing the 
developing embryo.   
If successful, this system would prevent any escapees from reproducing or 
producing viable offspring thus preventing the establishment of feral populations 
among native wild stocks.  Reproduction would only be possible under appropriate 
culture conditions.  A repressor compound is necessary at an embryonic stage in the 
hatchery to allow the organism to live (Dunham, 2004).  This technology has its 
beginnings in a procedure that is used to genetically sterilize transgenic plants.  Much 
controversy was raised over the use of this technology because it forced farmers to 
depend upon the vendor for seed supply, thus leading to its abandonment, at least for 
now (Niiler, 1999).  The application of principles taken from this technology and 
applied to aquatic organisms appears to be very promising and more acceptable 
because of the increased mobility and risk of transgenic fish compared to transgenic 
plants. 
  Our objective was to evaluate this technology?s potential for application in 
channel catfish.  The channel catfish, Ictalurus punctatus, is the primary cultured fish 
 12
for food production in the U.S. (Engle, 2003).  The catfish industry in the United 
States boasts the most successful aquaculture business to date.  Catfish sales to 
processors for 2004 was estimated at 286 thousand to 288 thousand metric tons for 
2004, at an average of 1.50 to 1.56 dollars per kilogram (Harvey, 2004).  The industry 
is currently struggling to keep pace with the increasing cost of inputs and competition 
with inexpensive imported fish.  Also, efficiency of production, profitability and 
sustainability need improvement.   
Transgenic channel catfish have been developed that demonstrate increased 
growth rates and increased disease resistance (Dunham et al., 1992; Dunham et al., 
2002c).  The possible risks and concerns rising from the use of these transgenic 
aquatic animals has prevented their application to date.  Transgenic sterilization, if 
successful, would facilitate approval of this technology for the catfish industry and 
others. 
The primary objective of this research was to evaluate the sterile feral 
technology to transgenically sterilize channel catfish, Ictalurus punctatus, possessing 
transgenes, and to temporarily reverse the sterility as needed to produce sterile 
progeny for production.  Specific objectives were to:  1) determine the ability of the 
SF3 and SF4 gene constructs to interrupt embryonic development, 2) to determine the 
concentration of doxycycline necessary to counteract the SF3 and SF4 constructs thus 
allowing for normal development and 3) to determine the period of development at 
which SF3 and SF4 disrupts development. 
 13
MATERIALS AND METHODS 
 
 
 
Sterile Feral Constructs 
 
Sterile Feral gene constructs (SF3 and SF4) were obtained from CSIRO.  For 
construction of SF3 and SF4, a developmental stage-specific promoter was coupled to 
components derived from the commercially available Tet-Off controllable expression 
system marketed by CLONTECH ( http://clontech.com/, Protocol # PT3001-1, 
version # PR95962).   
SF3 Construct 
 This construct contains the zebrafish zSMad5 promoter whose expression 
drives the tTA (Figure 3).  Mutations in the zebrafish SMad5 gene result in a 
dorsalized mutation designated somitabun and the dorsalised mutant phenotype has 
been shown to be rescued by injection of SMad5 mRNA at the single cell stage.  The 
next component is the SV40PA enhancer, which is taken from a eukaryotic cell virus.  
This enhancer acts on the zSMad5 promoter to increase the level of transcription 
(Kriegler, 1990).  The next sequence is the enhanced green fluorescent protein 
(EGFP) coding region.  The next section of the plasmid contains a bi-directional tet-
responsive element (BI-TRE), which contains seven copies of the 42 bp tet operator 
sequence (tetO).  When tetracycline is present, the tTA does not bind to the TRE 
which blocks gene expression from the TRE.  The next sequence in the construct is a 
single tet responsive double stranded RNA blocker construct under the regulation of 
zBMP2 promoter.  The dsBmp2 gene serves to block expression of the zSMAD5 
Figure 3.  Sterile Feral 3 Construct. (Thresher et al., 2001) 
 
 
 
 
 
 
 
 
 
 14
 15
promoter.  The plasmid was linearized by Aqua Bounty Technologies Inc. using the 
Xhol restriction enzyme.  The entire length of the construct is 7998 bp (Thresher et 
al., 2001).   
SF4 Construct 
The only difference between the SF3 plasmid and SF4 plasmid is that the 
dsBMP2 construct was replaced with a pBI-bmp2-Sense construct that served as a 
blocker on the zSMAD5 promoter (Figure 4).  This plasmid was also linearized using 
the Xhol restriction enzyme.  The entire length of the construct is 8611 bp.  (Thresher 
et al., 2001). 
Other Constructs 
Glutamate Decarboxylase (GAD) Construct   
The GAD construct is made up of the GAD65 gene (Accession number 
AF045594) put into the BsrGI, which is a isoschizomer of Kpnl, site of pFV-3 Cat 
(Figure 5).  The GAD65 gene was obtained from the University of Ottawa and the 
carp ?-actin promoter was obtained from Auburn University.  Aqua Bounty 
Technologies Inc. fused these two components to make the GAD construct used in 
this study.  The GAD65 gene is 2,653 bp and the pFV-3 Cat carp ?-actin promoter is 
6,782 bp.     
CAB Construct 
 The pPCab-IfS plasmid contains the GnRHsense (long fragment) under the 
control of the B-actin promoter pPCab-IfA (Figure 6).  This construct was employed 
in the first experiment.  The entire length of both plasmids was 8546 bp. 
Brood Stock 
 Potentially transgenic brood stock were used for this research.  Brood stock 
were obtained from the Genetics Research Station of Auburn University in Auburn, 
Figure 4.  Sterile Feral 4 construct (Thresher et al., 2001). 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 16
Figure 5.  Glutamate decarboxylase construct (Caldovic et. al, 1995). 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 17
Figure 6.  CAB construct. 
 
 
 
 
 
 
 
 
 18
 19
AL, USA.  The fish were collected from two separate ponds, G-32 and G-39.  The 
 
brood stock collected from pond G-32 were potentially transgenic for rainbow trout  
 
metallothionein promoter coupled with channel catfish growth hormone gene.  The 
 
brood stock collected from G-39 were potentially transgenic for an all fish promoter 
coupled with the cecropin B gene.  Two separate spawnings were conducted.  
Procedures were the same each time except as noted.  For the first round spawning, 6 
females and 4 males were collected from pond G-32.  Seven females and 4 males 
were collected from pond G-39.  The second round of spawning used 12 females from 
pond G-32 and 12 males from G-39.   
Spawning 
 The fish were separated by sex and held in separate flow-through tanks.  They 
were then weighed and the females were placed in individual mesh bags.  The females 
received two doses of luteinizing hormone releasing hormone (LHRHa).  The first 
injection was with a concentration of 200 ?g/ml at a rate of 0.1 ml/kg (20 ?g/kg).  
The second injection had a concentration of 1000 ?g/ml also at a rate of 0.1 ml/kg 
(100 ?g/kg).  The injections were given approximately 12 hours apart and at a water 
temperature of 25-27?C.  Testes of four G-32 males and four G-39 males were 
harvested.  The testes were harvested and placed in two separate clean sealed bags for 
use upon arrival of the eggs.  The first eggs were released by a G-39 female 
approximately 35 hours post second injection.  Eggs from two females from each 
pond were randomly separated into multiple greased pie pans to prepare for 
fertilization.  Testes from four G-32 and four G-39 males were manually ruptured to 
release milt over the eggs.  Water was applied at 0036 in order to stimulate 
fertilization of these eggs.    
 20
 The second spawning began approximately twenty days after the first round 
began.  G-32 females were crossed with G-39 males.  The females received LHRHa 
implants at a dosage rate of 75 ?g/kg.  Eggs from five G-32 females were randomly 
fertilized with twelve males from G-39.          
Gene transfer 
 Fertilized eggs were loaded into 50-mm plastic petri dishes for 
electroporation.  Random samples of eggs were taken from multiple pie pans and 
placed in each petri dish.  Approximately 2 ml of each genetic construct media (50 
?g/ml) was placed in the dish containing the eggs.  The first trial had six replicates of 
the two sterile feral constructs and three replicates of GAD and CAB.  There were 
also 3 electroporated blank control replicates.  The eggs were electroporated at 0106, 
60-100 minutes post-fertilization (Dunham et al., 1992; Powers et al., 1992).  
Electroporation was performed with the Baekon 2000 macromolecule transfer system.  
Its parameters were set at 6 kV, 2
7
 pulses, 0.8 sec burst, 4 cycles, 160 ?sec (Powers et 
al., 1992).  The electrode was situated 1-2 mm above the egg layer in each dish.  The 
eggs were then incubated in the DNA solution or nanopure water for 10 minutes 
before transferring to prepared incubation containers.  Each incubation container was 
initially filled with 4 L of 100% Holtzfretter?s solution.  Three out the six sterile feral 
constructs, both SF3 and SF4, were incubated in 100% Holtzfretter?s solution treated 
with doxycycline (100 mg/ml).  The doxycycline rate was increased to 150 mg/L 
approximately 60 hours post fertilization.  
 The second experimental trial of catfish followed the same electroporation 
protocol, except for the amount of time the eggs remained in the DNA solution.  Post 
electroporation, the eggs remained covered in the petri dish for 30 minutes before 
being divided into multiple incubation containers.  The electroporated eggs were 
 21
divided into six replicates for the sterile feral construct groups and three for the GAD 
and a electroporated blank control group.  For the purpose of studying the amount of 
antibiotic necessary to repress the blocker, three different rates of doxycycline were 
used in the second experiment.  In three of the six sterile feral replicates of both SF3 
and SF4, from both the first and second batch, one replicate was incubated in 50 
mg/L, 100 mg/L, and 150 mg/L.  Thus, for the second experiment there were six SF3 
replicates, three treated with one level of the doxycycline concentration and three 
without doxycycline.  The same protocol was followed for the SF4 construct.  There 
was also three replicates of the GAD construct.  A third experiment was conducted 
using only the sterile feral constructs and a electroporated blank control group.  The 
sterile feral constructs were replicated three times each while a the electroporated 
blank control was replicated three times (Table 1).  All treatments were initially 
incubated in 4 L of 100% Holzfretter?s solution. 
Incubation and Hatching 
 After the electroporation procedure was completed, the eggs from each genetic 
group were equally distributed to the incubation containers.  Each container was 
initially filled with 4 L of 100% Holzfretter?s solution.  The incubation solution was 
maintained as 100% Holzfretter?s until 48 hours post fertilization, at which time it 
was reduced by 25% each 24 hour period dechlorinated city water until it was 
replaced with 100% dechlorinated city water.  Doxycycline treatment was terminated 
around 24 hours prior to hatch.  Each replicate was sampled at approximately 24-hour 
intervals.  Each sampling involved a random sample of eggs or the complete number 
of eggs.  Eggs were observed under a dissecting microscope.  Dead and deformed 
eggs or embryos were removed and frozen for future PCR analysis.  Normal fry were 
acclimated and transferred, 3 days post-hatch, to a recirculating system located at the  
 22
Table 1.  Treatments and number of replicates employed in each experiment. 
Treatment Experiment 1 Experiment 2 Experiment 3 
SF3 
 
3 3 3 
SF4 
 
3 3 3 
SF3 Dox 50 
 
- 2 - 
SF3 Dox 100 
 
3 2 - 
SF3 Dox 150 
 
- 2 - 
SF4 Dox 50 
 
- 2 - 
SF4 Dox 100 
 
3 2 - 
SF4 Dox 150 
 
- 2 - 
Control 
 
3 3 3 
CAB 
 
3 - - 
GAD 
 
3 3 - 
 
 23
Genetics Research Station, Auburn University, Auburn, AL.  Fish in each replicate 
were observed again 1 month post-hatch for deformities.  Fish were enumerated and 
weighed.   
The second round of catfish also had two replicates each of varying 
concentrations of doxycycline, 50 mg/L, 100 mg/L, and 150 mg/L.  The sampling 
intervals were slightly longer due to the increased number of replicates, and were 
maintained at intervals ranging from 6-48 hours.    
Egg and Embryo Observations 
 As soon as the eggs were transferred to the incubation containers they were 
observed and inspected for death and deformity.  The observations were made with a 
dissecting microscope and lamp as well as gross observations.  Initially, dead eggs, 
white or opaque in color, were removed and frozen.  The second experiment was 
began with a total fertilized egg count to establish a baseline on total fertilized eggs in 
each replicate.  Fertilization was based upon the observation, starting at 2 hours post 
fertilization, of a rounded blastodisc, which was moderately raised from the surface of 
the yolk (Mansueti and Hardy, Jr., 1967).  Throughout the incubation phase, dead or 
deformed were removed and frozen for future PCR analysis.  Embryos with any 
structural deformities, most notably in the posterior region, were removed and frozen.  
Embryos or fry that were significantly behind in development as compared to others 
in the group or completely arrested were also removed and frozen.   
Water Analysis 
 An alcohol thermometer was used to monitor temperature throughout the 
experiment.  Multi-test strips from Mardel were used to monitor overall water quality 
parameters, such as total hardness, total alkalinity, nitrates, nitrites and pH.  The pH 
was also monitored through pH strips. A liquid ammonia test kit, Aquarium 
 24
Pharmaceuticals, Inc., was used to monitor total ammonia levels in the incubation 
tubs. 
PCR Analysis 
 DNA anaylysis was conducted at Aqua Bounty Technologies Inc. (San 
Diego).  DNA was extracted from about 50 milligrams of catfish fin or up to 4 
embryos using the DNAzol reagent.  Samples were resuspended in 8 mM NaOH and 
further diluted 5 fold with H
2
O.  To check the integrity of the DNA the actin gene 
from catfish was used as a standard.  Specific primers against the actin gene were 
designed.  The primers are called CFaActin-3461F/3551R and have the sequences 5?-
TACTCCGTTTGGATCGGTGG-3? and 5?-GACCGGCCTCGTCGTAGTC-3?  
respectively.   
Samples were run in duplicate on the Corbett Research RG-3000 real time 
PCR system using Applied Biosciences SyBR Green master mix.  Transgenic fish or 
embryos were determined by using one of two sets of primers one for the Bi-TRE 
promoter region and the other from the tTA gene.  The primers are as follows: Bi-Tre-
114R - 5?-GACCTATATAAGCAGAGCTCGTTTAGTG-3?, Bi-Tre-14F- 5?-
GGCTGGATCGGTCCCG-3? and SF3-4 tTA-493F -5?-
CTGATAGTATGCCGCCATTATTACG-3?, SF3-4 tTA-593R ? 5?-
TCCGCATATGATCAATTCAAGG-3?.  Each sample was run in duplicate and the 
melt curves were analyzed for positives. 
Data Analysis   
 Chi-square tests were performed on cumulative mortalities, timeline 
mortalities and cumulative deformities to calculate significance in comparing the 
sterile feral constructs with various controls and doxycycline treated groups.   
  
 25
RESULTS 
 
 
 
Mortality rates (dead and deformed dead) 
 
SF constructs versus controls 
 
For experiment one, the mean mortality percent for both SF3 (79.2%) and SF4 
(70.6%) were greater (P < 0.05) than that of the electroporated control group (28.1%) 
(Table 2) (Figure 7).  The mean mortality percent was higher (P < 0.05) in the SF3 
group as compared to SF4.  Mean mortality for both SF3 and SF4 were greater (P < 
0.05) than both CAB (13.9%) and GAD (34.5%) DNA control groups (Table 2).  A 
higher (P < 0.05) mean mortality percentage was demonstrated by the GAD group as 
compared to CAB.    
Mean mortality percentages in experiment two for SF3 (70.1%) and SF4 
(59.2%) were greater (P < 0.05) than the electroporated control group (27.5%) (Table 
2) (Figure 8).  The SF3 group in this experiment had a mean mortality that was 
greater (P < 0.05) than that of the SF4 group.  The mean mortality percentages for 
both SF3 and SF4 were greater (P < 0.05) than the GAD (53.5%) DNA control group 
(Table 2). 
In experiment three, mean mortality for both SF3 (85.5%) and SF4 (84.0%) 
were greater (P < 0.05) than the electroporated control group (69.4%) (Table 2) 
(Figure 9).  The mean mortality was not different (P > 0.05) in SF4 as compared to 
SF3 (Table 2). 
Table 2.  Mean mortality of channel catfish, Ictalurus punctatus, embryos 
electroporated with Sterile Feral (SF3 and SF4) gene constructs, glutamate 
decarboxylase (GAD), CAB and a electroporated control.  Dead eggs and embryos 
were identified and removed beginning from 14 hours post fertilization up to 153 
hours post fertilization. 
 
Mean Mortality (%) 
 
Treatment 
 
 
Experiment 
 1
a
Experiment
2
 a
Experiment
3
  a
Grand  
Mean 
SF3 
 
79.2
 
70.1
 
85.5
 
78.5
1
SF4 
 
70.6 59.2 84.0 69.6
2
Control 
 
28.1 27.5 69.4 50.3
3
GAD 
 
34.5 53.5 - 51.4
3
CAB 
 
13.9 - - 13.9
4 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a
 ? SF3 was greater than SF4 (P < 0 .05); SF3 and SF4 were greater than control, 
GAD and CAB (P < 0.05), chi-square. 
1 2 3 4
 ? Mean percentages with different letters within the column are different (P < 
0.05), chi-square. 
 
 
 
 
 
 
 
 
 
 
 
 
 26
Figure 7.  Mean mortality of channel catfish, Ictalurus punctatus, embryos 
electroporated with Sterile Feral (SF3 and SF4) gene constructs and a electroporated 
control in experiment one.  Dead eggs and embryos were identified and removed 
beginning from 14 hours post fertilization up to 153 hours post fertilization.   
0
10
20
30
40
50
60
70
80
90
Me
a
n
 Mo
r
t
a
l
i
t
y
 P
e
r
cen
t
SF3 SF4 Control
 
 
 
 
 
 
 
 
 
 
 27
Figure 8.  Mean mortality of channel catfish, Ictalurus punctatus, embryos 
electroporated with Sterile Feral (SF3 and SF4) gene constructs and a electroporated 
control in experiment two.  Dead eggs and embryos were identified and removed 
beginning from 18 hours post fertilization up to 137 hours post fertilization.   
0
10
20
30
40
50
60
70
80
Me
a
n
 Mo
r
t
a
l
i
t
y
 P
e
r
cen
t
SF3 SF4 Control
 
 
 
 
 
 
 
 
 
 
 28
Figure 9.  Mean mortality of channel catfish, Ictalurus punctatus, embryos 
electroporated with Sterile Feral (SF3 and SF4) gene constructs and a electroporated 
control in experiment three.  Dead eggs and embryos were identified and removed 
beginning from 18 hours post fertilization up to 137 hours post fertilization.  
0
10
20
30
40
50
60
70
80
90
Me
a
n
 Mo
r
t
a
l
i
t
y
 P
e
r
cen
t
SF3 SF4 Control
 
 
 
 
 
 
 
 
 
 
 29
 30
Effect of doxycycline 
The first channel catfish experiment produced a mean mortality percentage for 
SF3 (79.2%) that was higher (P < 0.05) than the SF3 doxycycline 100 group (47.3%) 
(Table 3) (Figure 10).  Thus, the SF3 doxycycline 100 group reduced the relative 
mean mortality percentage of SF3, which was 179% higher than the electroporated 
control, to a 68.3% increase.  The mean mortality percentage was higher (P < 0.05) in 
the SF3 doxycycline group as compared to the non-electroporated control (28.1%) 
(Table 3).   
In the second experiment, a mean mortality percentage for SF3 (70.1%) was 
greater (P < 0.05) than both SF3 doxycycline 50 (56.4%), and 100 (55.6%).  It was 
less (P < 0.05) than that of SF3 doxycycline 150 (77.5%) (Table 3) (Figure 11).  
Therefore, the SF3 doxycycline 50 group reduced the relative mean mortality 
percentage for SF3, 154.9%, to 105.1%.  The SF3 doxycycline 100 group reduced it 
to 102.2% and lastly, the SF3 doxycycline 150 group produced a 181.8% increase 
over that of the electroporated control group.  Mean mortalities were higher (P < 0.05) 
in all SF3 doxycycline groups as compared to the electroporated control group 
(27.5%).  The mean mortality percentages were not different (P > 0.05) in the SF3 
doxycycline 50 and 100 groups.  SF3 doxycycline 150 had a higher (P < 0.05) mean 
mortality percentage than both SF3 doxycycline 50 and 100 (Table 3).   
In the first experiment, the mean mortality percentage for SF4 (70.6%) was 
not different (P > 0.05) than the SF4 doxycycline 100 group (63.0%) (Table 4) 
(Figure 12).  SF4 had a higher (P < 0.05) mean mortality as compared to the 
electroporated control group (28.1%).  The mean mortality percentage of SF4 relative 
to that of the electroporated control was 151.2% higher and was reduced to 124.2% by 
the SF4 doxycycline 100 group.   
Table 3.  Mean mortality of channel catfish, Ictalurus punctatus, embryos 
electroporated with the Sterile Feral (SF3) gene construct, and treated with three 
different concentrations of doxycycline (50, 100 and 150 ppm) from 4 hours post- 
fertilization to first hatch and a electroporated control group.  Dead eggs and embryos 
were identified and removed beginning from 14 hours post fertilization up to 153 
hours post fertilization. 
a b c
 - mean percentages with different letters within columns are different (P < 0.05), 
chi-square. 
Mean Mortality (%) 
 
Treatment 
 
 
Experiment 1 Experiment 2 Experiment 3 Grand 
Mean 
SF3 
 
79.2
a
70.1
b
85.5
a
78.5
a
SF3 Dox 50 ppm 
 
- 56.4
c
- 56.4
b
SF3 Dox 100 ppm 
 
47.3
b
55.6
 c
- 52.9
b
SF3 Dox 150 ppm 
 
- 77.5
 a
- 77.5
a
Control 
 
28.1
c
27.5
d
69.4
c
50.3
b 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 31
Figure 10.  Mean mortality of channel catfish, Ictalurus punctatus, in experiment one. 
Embryos were electroporated with the Sterile Feral (SF3) gene construct, and treated 
with doxycycline (100 ppm) from 4 hours post fertilization to first hatch and a 
electroporated control.  Dead eggs and embryos were identified and removed 
beginning from 14 hours post fertilization up to 153 hours post fertilization. 
0
10
20
30
40
50
60
70
80
90
M
e
an
 M
o
r
t
al
i
t
y
 Pe
r
c
e
n
t
SF3 SF3 Dox Control
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 32
 
Figure 11.  Mean mortality of channel catfish, Ictalurus punctatus, in experiment two. 
Embryos were electroporated with the Sterile Feral (SF3) gene construct, and treated 
with three concentrations of doxycycline (50, 100, and 150 ppm) from 4 hours post 
fertilization to first hatch and a electroporated control.  Dead eggs and embryos were 
identified and removed beginning from 18 hours post fertilization up to 137 hours 
post fertilization. 
0
10
20
30
40
50
60
70
80
90
Me
a
n
 Mo
r
t
a
l
i
t
y
 P
e
r
cen
t
SF3 SF3 Dox 50 SF3 Dox 100 SF3 Dox 150 Control
 
 
 
 
 
 
 
 
 
 33
Table 4.  Mean mortality of channel catfish, Ictalurus punctatus, embryos 
electroporated with the Sterile Feral (SF4) gene construct, and treated with three 
different concentrations of doxycycline (50, 100 and 150 ppm) from 4 hours post 
fertilization to first hatch and a electroporated control.  Dead eggs and embryos were 
identified and removed beginning from 14 hours post fertilization up to 153 hours 
post fertilization. 
a b c
 - mean percentages with different letters within columns are different (P < 0.05), 
chi-square. 
Mean Mortality (%) 
 
Treatment 
 
 
Experiment 1 Experiment 2 Experiment 3 Grand 
Mean 
SF4 
 
70.6
a
59.2
a
84.0
a
69.1
a
SF4 Dox 50 ppm 
 
- 64.2
a
- 64.2
b
SF4 Dox 100 ppm 
 
63.0
a
62.3
a
- 62.6
b
SF4 Dox 150 ppm 
 
- 52.8
b
- 52.8
c
Control 
 
 
28.1
b
27.5
c
69.4
b
50.3
c
 
 
 
 
 
 
 
 
 
 
 
 34
Figure 12.  Mean mortality of channel catfish, Ictalurus punctatus, in experiment one. 
Embryos were electroporated with the Sterile Feral (SF4) gene construct, and treated 
with doxycycline (100 ppm) from 4 hours post fertilization to first hatch and a 
electroporated control.  Dead eggs and embryos were identified and removed 
beginning from 14 hours post fertilization up to 153 hours post fertilization. 
0
10
20
30
40
50
60
70
80
M
e
an
 M
o
r
t
al
i
t
y
 Pe
r
c
e
n
t
SF4 SF4 Dox Control
 
 
 
 
 
 
 
 
 
 35
 
 36
Mean mortality percentages for the second experiment for SF4 (59.2%) were 
not different (P > 0.05) than that of SF4 doxycycline 50 (64.2%) and SF4 doxycycline 
100 (62.3%) (Table 4) (Figure 13).  SF4 did have a greater (P < 0.0125) mean 
mortality percentage than SF4 doxycycline 150 (52.8%) (Table 4).  The SF4 
doxycycline 150 group was the only SF4 doxycycline group in this experiment that 
demonstrated a reduction in the relative mean mortality percent of SF4.  SF4 
doxycycline 150 reduced the relative mean mortality percent of SF4, 115.3%, to 
92.0%.  Mean mortality percentages were higher (P < 0.05) in all SF4 doxycycline 
groups as compared to the electroporated control group (27.5%) (Table 4).  SF4 also 
had a higher (P < 0.05) mean mortality percentage than the electroporated control 
group.  Mean mortality percentages were not different (P > 0.05) in the SF4 
doxycycline 50 and 100 groups.  SF4 doxycycline 100 generated a higher (P < 0.05) 
mortality than SF4 doxycycline 150.  SF4 doxycycline 50 also had a higher (P < 0.05) 
mean mortality percentage than that of SF4 doxycycline 150 (Table 4).   
Mortality over time 
 
SF3-SF4-control 
 
In the first experiment, there was no difference (P > 0.05) in the mean 
mortality percentages among SF3 (2.1%), electroporated control (3.5%) and SF4 
(2.8%) for the 0-14 hour time period (Table 5) (Figure 14).  There were no mortalities 
observed during the 14-31 hour time period in the same experiment.  During the 31-
61 hour time period, mean mortality percentages for both SF3 (73.5%) and SF4 
(62.4%) were greater (P < 0.05) than the control group (16.3%) (Table 5).  Also 
during this same time period, SF3 produced a greater (P < 0.05) mean mortality 
percentage than did the SF4 group (Table 5).  No mortalities were observed during the 
61-85 hour time period.  For the 85-109 hour interval, SF3 (2.7%) had a mean 
Figure 13.  Mean mortality of channel catfish, Ictalurus punctatus, in experiment two. 
Embryos were electroporated with the Sterile Feral (SF4) gene construct, and treated 
with three concentrations of doxycycline (50, 100 and 150 ppm) from 4 hours post 
fertilization to first hatch and a electroporated control.  Dead eggs and embryos were 
identified and removed beginning from 14 hours post fertilization up to 153 hours 
post fertilization. 
0
10
20
30
40
50
60
70
Me
a
n
 Mo
r
t
a
l
i
t
y
 P
e
r
cen
t
SF4 SF4 Dox 50 SF4 Dox 100 SF4 Dox 150 Control
 
 
 
 
 
 
 
 
 
 37
Table 5.  Mean mortality at various time intervals (hours) for channel catfish, 
Ictalurus punctatus, embryos electroporated with the Sterile Feral (SF3 and SF4) gene 
constructs, and a electroporated blank control in experiment one.  Dead eggs and 
embryos were identified and removed beginning from 14 hours post fertilization to 
153 hours post fertilization. 
a b c
 ? mean percentages with different letters within columns are different (P < 0.05), 
chi-square. 
Mortality (%)  in Hour Intervals 
 
Treatment 
 
0-14 14-31 31-61 61-85 85-109 109-133 133-153 
SF3 
 
2.1
a
0.0
a
73.5
a
0.0
a
2.7
b
0.0
a
0.9
a
SF4 
 
2.8
a
0.0
a
62.4
b
0.0
a
4.7
a,b
0.8
a
0.0
a
Control 
 
 
3.5
a
0.0
a
16.3
c
0.0
a
7.5
a
0.0
a
0.8
a
 
 
 
 
 
 
 
 
 
 
 
 
 
 38
Figure 14.  Mean mortality at various time intervals (hours) for channel catfish, 
Ictalurus punctatus, embryos electroporated with the Sterile Feral (SF3 and SF4) gene 
constructs, and a electroporated control in experiment one.  Dead eggs and embryos 
were identified and removed beginning from 14 hours post fertilization to 153 hours 
post fertilization. 
0
10
20
30
40
50
60
70
80
0-14 14-31 31-61 61-85 85-109 109-133 133-153
TIme Post Fertilization (Hours)
M
ean
 M
o
r
t
al
i
t
y
 P
e
r
cen
t
SF3 SF4 Control
 
 
 
 
 
 
 
 
 
 
 39
 40
mortality percentage that was lower (P < 0.05) than the electroporated control group 
(7.5%) (Table 5).  For the same time period, the mean mortalities were not different 
(P < 0.05) between SF4 (4.7%) and the electroporated control group or between SF4 
and SF3 (Table 5).  During the 109-133 time period, there was no difference (P > 
0.05) among the treatments.  For the final time interval, 133-153 hours, of this 
experiment, there were also no difference (P < 0.05) observed among any of the 
treatments (Table 5).   
During the 0-18 hour time interval in the second experiment, the SF3 (5.8%) 
group produced a higher (P < 0.05) mean mortality percentage than that of the 
electroporated control group (2.5%) (Table 6) (Figure 15).  There was no difference 
(P > 0.05) between SF4 (1.3%) and the electroporated control group during this time 
period.  Both SF3 (57.5%) and SF4 (54.5%) had higher (P < 0.05) mean mortality 
percentages than the electroporated control group (22.7%) during the next time period 
of 18-59 hours (Table 6).  The mean mortality percentage of SF3 was also greater (P 
< 0.05) than SF4 during this same time period.   
The mean mortality percentage during the 59-96 hour time interval for the SF3 
(4.4%) group was greater (P < 0.05) than that of the electroporated control (2.0%) 
group and the SF4 (2.3%) group (Table 6).  During this same time period, SF4 did not 
produce a mean mortality percentage that was greater than the electroporated control 
group (P > 0.05) (Table 6).   
For the final time interval, 96-137 hours, SF3 (2.4%) produced a mean 
mortality percentage that was greater (P < 0.05) than both the electroporated control 
group (0.3%) and SF4 (1.1%) (Table 6).  SF4 did not produce a mean mortality 
percentage that was greater (P < 0.05) than the electroporated control group (Table 6).
  
 41
Table 6.  Mean mortality percentages at various time intervals (hours) for channel 
catfish, Ictalurus punctatus, embryos electroporated with the Sterile Feral (SF3 and 
SF4) gene constructs, and a electroporated control in experiment two.  Dead eggs and 
embryos were identified and removed beginning from 18 hours post fertilization to 
137 hours post fertilization.  
Mean Mortality (%) in Hour Intervals 
 
Treatment 
 
0-18 18-59 59-96 96-137
SF3 
 
5.8
a
57.5
a
4.4
a
2.4
a
SF4 
 
1.3
b
54.5
a
2.3
b
1.1
b
Control 
 
2.5
b
22.7
b
2.0
b
0.3
b
a b c
 ?mean percentages with different letters within columns are different (P < 0.05), 
chi-square. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 15.  Mean mortality at various time intervals (hours) for channel catfish, 
Ictalurus punctatus, embryos electroporated with the Sterile Feral (SF3 and SF4) gene 
constructs, and a electroporated blank control in experiment two.  Dead eggs and 
embryos were identified and removed beginning from 18 hours post fertilization to 
137 hours post fertilization. 
0
10
20
30
40
50
60
70
0-18 18-59 59-96 96-137
Time Post Fertilization (Hours)
M
ean
 M
o
r
t
al
i
t
y
 Pe
r
c
e
n
t
SF3 SF4 Control
 
 
 
 
 
 
 
 
 
 42
 
 43
During the 0-18 hour time period for the third experiment the mean mortality 
percentage for the electroporated control group (10.6%) was greater (P < 0.05) than 
that of SF3 (3.5%) but not different (P > 0.05) than that of SF4 (9.7%) (Table 7)  
(Figure 16).  SF4 had a higher (P < 0.05) mean mortality percentage than the SF3 
group during this same time period (Table 7).  The SF3 (73.9%) group produced a 
greater (P < 0.05) mean mortality percentage than did the electroporated control group 
(54.2%) but was not different (P > 0.05) than the SF4 (71.0%) group during the 18-59 
hour time interval (Table 7).  In the same time period, SF4 had a higher (P < 0 .05) 
mean mortality percentage as compared to the electroporated control group (Table 7).  
The mean mortality percentage during the 59-96 hour time interval for SF4 (1.8%) 
was less than (P < 0.05) that of both SF3 (5.9%) and the electroporated control group 
(4.2%) (Table 7).  There was no difference (P > 0.05) between SF3 and the 
electroporated control group during the same time period (Table 7).  Next, 
during the 96-137 hour time period, SF3 (2.2%) produced a mean mortality 
percentage that was greater (P < 0.05) than that of the electroporated control group 
(0.4%) but not greater (P > 0.05) than that of SF4 (1.5%) (Table 7).  SF4 did have a 
higher (P < 0.05) mean mortality percentage as compared to the electroporated control 
group in this final time period of 96-137 hours (Table 7).   
SF3-SF4-DNA Control 
During the 0-14 hour time interval for the first experiment, there was no 
difference (P > 0.05) between the mean mortality percentages of SF3 (2.1%), SF4 
(2.8%), CAB (1.1%), or GAD (0.0%) (Table 8) (Figure 17).  There were no 
differences in any of the comparisons made between the treatments.    
There were no mortalities for any of the groups during the 14-31 hour time 
interval.  During the 31-61 hour time period, the mean mortality percentage of SF3 
 44
Table 7.  Comparison of the mean mortality percentages at various time intervals 
(hours) for channel catfish, Ictalurus punctatus, embryos electroporated with the 
Sterile Feral (SF3 and SF4) gene constructs, and a electroporated blank control in 
experiment three.  Dead eggs and embryos were identified and removed beginning 
from 18 hours post fertilization to 137 hours post fertilization. 
Mean Mortality (%) in Hourly Intervals 
 
Treatment 
 
0-18 18-59 59-96 96-137 
SF3 
 
3.5
b
73.9
a
5.9
a
2.2
a
SF4 
 
9.7
a
71.0
a
1.8
b
1.5
a
Control 
 
10.6
a
54.2
b
4.2
a
0.4
b
a b c
 ? mean percentages with different letters within columns are different (P < 0.05), 
chi-square. 
  
 
 
 
 
 
 
 
 
 
 
 
 
Figure 16.  Mean mortality at various time intervals (hours) for channel catfish, 
Ictalurus punctatus, embryos electroporated with the Sterile Feral (SF3 and SF4) gene 
constructs, and a electroporated control in experiment three.  Dead eggs and embryos 
were identified and removed beginning from 18 hours post fertilization to 137 hours 
post fertilization. 
0
10
20
30
40
50
60
70
80
0-18 18-59 59-96 96-137
Me
a
n
 Mo
r
t
a
l
i
t
y
 P
e
r
cen
t
SF3 SF4 Control
  
 
 
 
 
 
 
 
 
 45
 
 46
Table 8.  Comparison of the mean mortality percentages at various time intervals 
(hours) for channel catfish, Ictalurus punctatus, embryos electroporated with the 
Sterile Feral (SF3 and SF4) gene constructs, glutamate decarboxylase (GAD) and 
CAB in experiment one.  Dead eggs and embryos were identified and removed 
beginning from 14 hours post fertilization to 153 hours post fertilization.   
Mean Mortality (%) in Hourly Intervals 
 
Treatment 
 
0-14 14-31 31-61 61-85 85-109 109-133 133-153 
SF3 
 
2.1
a
0.0
a
73.5
a
0.0
a
2.7
ab
0.0
a
0.9
a
SF4  
 
2.8
a
0.0
a
62.4
b
0.0
 a
4.7P
b
0.8
a
0.0
a
CAB 
 
1.1
a
0.0
a
12.2
c
0.0
 a
0.0P
c
0.4
a
0.0
a
GAD 
 
 
0.0
a
0.0
a
72.3
ab
0.0
 a
0.0P
ac
0.0P
a
0.0P
a
a b c
 ? mean percentages with different letters within columns are different (P < 0.05), 
chi-square. 
  
 
 
 
 
 
 
 
 
 
 
 
Figure 17.  Comparison of the mean mortality percentages at various time intervals 
(hours) for channel catfish, Ictalurus punctatus, embryos electroporated with the 
Sterile Feral (SF3 and SF4) gene constructs, glutamate decarboxylase (GAD) and 
CAB in experiment one.  Dead eggs and embryos were identified and removed 
beginning from 14 hours post fertilization to 153 hours post fertilization.   
0
10
20
30
40
50
60
70
80
0-14 14-31 31-61 61-85 85-109 109-133 133-153
Me
a
n
 Mo
r
t
a
l
i
t
y
 P
e
r
cen
t
SF3 SF4 CAB GAD
  
 
 
 
 
 
 
 
 
 47
 
 48
(73.5%) was greater (P < 0.05) than SF4 (62.4%) and CAB (12.2%) but not different 
(P > 0.05) from the mean mortality percentage of GAD (72.3%) (Table 8).  In the 
same time period, SF4 had a mean mortality percentage that was greater (P < 0.05) 
than CAB but not different (P > 0.05) as compared to GAD (Table 8).  GAD had a 
mean mortality percentage that was greater (P < 0.05) than that of CAB (Table 8).  
There were no mortalities observed during the 61-85 hour time period for any of the 
groups in this experiment.   
There was no difference (P > 0.05) between the mean mortality percentages of 
SF3 (2.7%) and SF4 (4.7%) during the 85-109 hour time interval (Table 8).  During 
this same time period, SF3 produced a mean mortality percentage that was greater (P 
< 0.05) than that of CAB (0.0%) but was not different (P > 0.05) as compared to that 
of GAD (0.0%) (Table 8).  The mean mortality percentage of SF4 was greater (P < 
0.05) than both CAB and GAD (Table 8).  For the 85-109 hour time 
period, there was no difference between the mean mortality percentages of CAB and 
GAD for neither group had mortalities recorded for this period.   
During the 109-133 hour time interval, there were no differences (P > 0.05) in 
the comparison of the mean mortality percentage of SF4 (0.8%), SF3 (0.0%), CAB 
(0.4%) and GAD (0.0%) (Table 8).  There were also no difference (P > 0.05) among 
any of the treatments in the last period of this experiment, 133-153 hours (Table 8). 
In the second experiment, during the 0-18 hour time period, SF3 (5.8%) 
produced a mean mortality percentage that was not different (P > 0.05) as  
compared to GAD (4.2%) but was greater (P < 0.05) than SF4 (1.3%)  (Table 9) 
(Figure 18).  GAD had a higher (P < 0.05) mean mortality percentage during this 
period than SF4 (Table 9).   
 49
 Table 9.  Comparison of the mean mortality percentages at various time intervals 
(hours) for channel catfish, Ictalurus punctatus, embryos electroporated with the 
Sterile Feral (SF3 and SF4) gene constructs and glutamate decarboxylase (GAD) in 
experiment two.  Dead eggs and embryos were identified and removed beginning 
from 18 hours post fertilization to 146 hours post fertilization.   
Mean Mortality (%) in Hourly Intervals 
 
Treatment 
 
0-18 18-60 60-98 98-146
SF3 
 
5.8
a
57.5
a
4.4
a
2.4
a
SF4 
  
1.3
b
54.5
a
2.3
b
1.1
b
GAD 
 
 
4.2
a
45.4
b
1.4
b
0.3
c
a b c
 ? mean percentages with different letters within columns are different (P < 0.05), 
chi-square. 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 18.  Comparison of the mean mortality percentages at various time intervals 
(hours)for channel catfish, Ictalurus punctatus, embryos electroporated with the 
Sterile Feral (SF3 and SF4) gene constructs and glutamate decarboxylase (GAD) in 
experiment two.  Dead eggs and embryos were identified and removed beginning 
from 18 hours post fertilization to 146 hours post fertilization.   
 
0
10
20
30
40
50
60
70
0-18 18-60 60-98 98-146
Me
a
n
 Mo
r
t
a
l
i
t
y
 Pe
r
c
e
n
t
SF3 SF4 GAD
 
 
 
 
 
 
 50
 51
The mean mortality percentage for SF3 (57.5%) was greater (P < 0.05) as 
compared to GAD (45.4%) but not different (P > 0.05) as compared to SF4 (54.5%) 
during the 18-60 hour time period (Table 9).  In this same time period, SF4 also had a 
higher (P < 0.05) mean mortality percentage than GAD (Table 9).  The mean 
mortality percentage for SF3 (4.4%) during the 60-98 hour time interval was greater 
(P < 0.05) than both SF4 (2.3%) and GAD (1.4%) (Table 9).  During this same period, 
SF4 had a mean mortality that was not different (P > 0.05) from GAD (Table 9).  
During the final time period, 98-146 hours, SF3 (2.4%) produced a higher (P < 0.05) 
mean mortality percentage as compared to GAD (0.3%) and SF4 (1.1%) (Table 9).  
SF4 also produced a mean mortality percentage that was higher (P < 0.0441) than 
GAD (Table 9). 
SF constructs and the effects of doxycyline 
 
During the first experiment, the group SF3 (2.1%) exhibited a mean mortality 
percentage that was greater (P < 0.05) than SF3 doxycycline (0.0%) but not greater (P 
< 0.05) than the electroporated control group (3.5%) during the 0 to 14 hour time 
period (Table 10) (Figure 19).  The electroporated control group had a mean mortality 
for this same time interval that was greater (P < 0.05) than SF3 doxycycline 100 
(Table 10).  For the 14 to 31 hour time period, SF3 doxycycline 100 (2.5%) had a 
higher (P < 0.05) mean mortality percentage as compared to both SF3 (0.0%) and the 
electroporated control group (0.0%) (Table 10).  The mean mortality percentages for 
the SF3 group (73.5%) were greater (P < 0.05) compared to both the SF3 doxycycline 
100 group (33.3%) and the control group (16.3%) during the 34 to 61 hour time 
interval (Table 10).  SF3 doxycycline 100 had a higher (P < 0.05) mean mortality 
percentage than the electroporated control group (Table 10).   
 52
Table 10.  Comparison of the mean mortality percentages at various time intervals 
(hours) for  channel catfish, Ictalurus punctatus, embryos electroporated with the 
Sterile Feral (SF3) gene construct, and treated with doxycycline (100 ppm) from 4 
hours post fertilization to first hatch and a electroporated blank control in experiment 
one.  Dead eggs and embryos were identified and removed beginning from 14 hours 
post fertilization to 153 hours post fertilization. 
Mean Mortality (%) in Hourly Intervals 
 
Treatment 
 
0-14 14-31 31-61 61-85 85-109 109-133 133-153
SF3 
 
2.1
a
0.0
b
73.5
a
0.0
a
2.7
b
0.0
a
0.9
a
SF3 Dox 100 ppm 
 
0.0
b
2.5
a
33.3
b
0.0
a
11.4
a
0.0
a
0.0
a
Control 
 
3.5
a
0.0
b
16.3
c
0.0
a
7.5
a
0.0
a
0.8
a
a b c
 - percentages with different letters within columns signifies a difference (P < 
0.05), chi-square test. 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 19.  Comparison of the mean mortality percentages at various time intervals 
(hours) for channel catfish, Ictalurus punctatus, embryos electroporated with the 
Sterile Feral (SF3) gene construct, and treated with doxycycline (100 ppm) from 4 
hours post fertilization to first hatch and a electroporated control in experiment one.  
Dead eggs and embryos were identified and removed beginning from 14 hours post 
fertilization to 153 hours post fertilization. 
0
10
20
30
40
50
60
70
80
0-14 14-31 31-61 61-85 85-109 109-133 133-153
Me
a
n
 Mo
r
t
a
l
i
t
y
 P
e
r
cen
t
SF3 SF3 Dox 100 Control
 
 
 
 
 
 
 
 
 53
 54
There were no mortalities observed during the next time period, which was 61 
to 85 hours (Table 10).  SF3 doxycycline 100 (11.4%) produced a higher (P < 0.05) 
mean mortality percentage than did SF3 (2.7%) during the 85 to 109 hour time 
interval (Table 10).  During this same time period, the mean mortality percentage of 
the electroporated control group (7.5%) was higher (P < 0.05) than that of SF3 but not 
different (P > 0.05) than that of the SF3 doxycycline 100 group (Table 10).   
For the 109 to 133 time interval, no mortalities were observed for any of the 
groups.  There was no difference (P > 0.05) between SF3 (0.9%) and SF3 doxycycline 
100 (0.0%), SF3 and the electroporated control group (0.8%) or between SF3 
doxycycline 100 and the control group (Table 10).   
For the second experiment within the time interval of 0 to 12 hours, SF3 
(5.8%) produced a mean mortality percentage that was not different (P > 0.05) than 
that observed for SF3 doxycycline 50 (3.2%), SF3 doxycycline 100 (3.7%) or SF3 
doxycycline 150 (4.9%) (Table 11) (Figure 20).  There was no difference (P > 0.05) 
between the mean mortality percentage of the control group (2.5%) and SF3 
doxycycline 50, SF3 doxycycline 100, but the mortalities of the electroporated control 
group were lower (P < 0.05) than that of SF3 doxycycline 150 (Table 11).  The mean 
mortality percentage of SF3 was higher (P < 0.05) than that of the electroporated 
control group (Table 11).  SF3 doxycycline 50 produced a mean mortality percentage 
that was not different (P > 0.05) from that of SF doxycycline 100 or of SF3 
doxycycline 150 (Table 12).  Lastly, the mean mortality percentage of SF3 
doxycycline 100 as compared to SF3 doxycycline 150 was not different (P > 0.05) 
(Table 11).   
For the next time period, 12 to 58 hours, SF3 (57.5%) exhibited a mean 
mortality percentage that was not different (P > 0.05) than that of SF3 doxycycline 50  
 55
Table 11.  Comparison of the mean mortality percentages at various time intervals 
(hours) for channel catfish, Ictalurus punctatus, embryos electroporated with the 
Sterile Feral (SF3) gene construct, and treated with different concentrations of 
doxycycline (50, 100 and 150 ppm) from 4 hours post fertilization to first hatch and a 
electroporated control in experiment two.  Dead eggs and embryos were identified 
and removed beginning from 12 hours post fertilization to 137 hours post fertilization.   
Mean Mortality (%) in Hourly Intervals 
 
Treatment 
 
0-12 12-58 58-96 96-137
SF3 
 
5.8
a
57.5
a
4.4
b
2.4
a
SF3 Dox 50 ppm 
 
3.2
ab
52.5
a
0.0
d
0.7
ab
SF3 Dox 100 ppm 
 
3.7
ab
47.7
b
3.7
bd
0.5
b
SF3 Dox 150 ppm 
 
4.9
a
53.1
ab
11.4
a
2.5
a
Control 
 
 
2.5
b
22.7
c
2.0
c
0.3
b
a b c d
 - percentages with different letters within columns signifies a difference (P < 
0.05), chi-square test. 
 
 
 
 
 
 
 
 
 
Figure 20.  Comparison of the mean mortality percentages at various time intervals 
(hours) for channel catfish, Ictalurus punctatus, embryos electroporated with the 
Sterile Feral (SF3) gene construct, and treated with three different concentrations of 
doxycycline (50, 100 and 150 ppm) from 4 hours post fertilization to first hatch and a 
electroporated control in experiment two.  Dead eggs and embryos were identified 
and removed beginning from 14 hours post fertilization to 153 hours post fertilization. 
0
10
20
30
40
50
60
70
0-12 12--58 58-96 96-137
M
ean
 M
o
r
t
al
i
t
y
 Pe
r
c
e
n
t
SF3 SF3 Dox 50 SF3 Dox 100 SF3 Dox 150 Control
  
 
 
 
 
 
 
 
 56
 
 57
(52.5%) and SF3 doxycycline 150 (53.1%) (Table 11).  There was a difference (P < 
0.05) between the mean mortality percentages of SF3 and SF3 doxycycline 100 
(47.7%). The electroporated control group (22.7%) had a lower (P < 0.05) mean 
mortality percentage than that of SF3 doxycycline 50 and 100 (Table 11).  There was 
no difference (P > 0.05) between the mean mortality percentage of the electroporated 
control group and SF3 doxycycline 150 (Table 11).  During the same time period, SF3 
had a higher (P < 0.05) mean mortality percentage as compared to the electroporated 
control group.  SF3 doxycycline 50 had a higher (P < 0.05) mean mortality percentage 
than that of SF3 doxycycline 100 but was not different (P > 0.05) from that of SF3 
doxycycline 150.  SF3 doxycycline 100 had a higher (P < 0.05) mean mortality 
percentage as compared to that of SF3 doxycycline 150 (Table 11).   
During the 58 to 96 hour time interval, SF3 (4.4%) had a mean mortality 
percentage that was higher (P < 0.05) than that of SF3 doxycycline 50 (0.0%), but not 
different (P > 0.05) from that of SF3 doxycycline 100 (3.7%) and lower (P < 0.05) 
than that of SF3 doxycycline 150 (11.4%) (Table 11).  The mean mortality of the 
electroporated control group (2.0%) was greater (P < 0.05) than that of SF3 
doxycycline 50, not different (P > 0.05) than that of SF3 doxycycline 100 and less (P 
< 0.05) than that of SF3 doxycycline 150 (Table 12).  SF3 had a mean mortality that 
was higher (P < 0.05) than that of the electroporated control group.  SF3 doxycycline 
50 had no mortalities for this time period and thus had a lower (P < 0.05) mean 
mortality percentage as compared to SF3 doxycycline 100 and SF3 doxycycline 150.  
During this same time period, SF3 doxycycline 150 had a higher (P < 0.05) mean 
mortality percentage than SF3 doxycycline 100 (Table 11).   
During the last time interval, 96 to 137 hours, the mean mortality percentage 
of SF3 (2.4%) was not different (P > 0.05) from SF3 doxycycline 50 (0.7%) or SF3 
 58
doxycycline 150 (2.5%) but was higher (P < 0.05) than SF3 doxycycline 100 (0.5%) 
(Table 11).  The electroporated control group (0.3%) had a mean mortality percentage 
that was not different (P > 0.05) from SF3 doxycycline 50 or from SF3 doxycycline 
100 but was lower (P < 0.05) than that of SF3 doxycycline 150 (Table 11).  SF3 
produced a mean mortality percentage that was higher (P < 0.05) than that of the 
electroporated control group.  There was no difference (P > 0.05) between the mean 
mortality percentages of SF3 doxycycline 50 and 100 or between SF3 doxycycline 50 
and 150 (Table 11).  There was a difference (P < 0.05) between the mean mortality 
percentages of SF3 doxycycline 100 and 150 (Table 11). 
 SF4-Dox Comparison 
The initial time period for the first experiment with channel catfish was 0 to 14 
hours.  During this interval, the mortality pattern was atypical, for there were no 
treatments that experienced more than 3.5% mortality (Table 12) (Figure 21).  The 
next time interval was 14 to 31 hours, and mortality rates were no higher than 0.3% 
for any treatment (Table 12).  Mortalities peaked during the 31 to 61 hour time period.  
SF4 (62.4%) had a mean mortality rate that was greater (P < 0.05) than that of SF4 
doxycycline 100 (42.0 %) and the electroporated control group (16.3%) (Table 12).  
The mortality rate of the SF4 doxycycline 100 group was greater (P < 0.05) than that 
of the electroporated control group (Table 12).   
The time interval 61-85 hours had no mortalities observed.  During the next 
time period, 85 to 109 hours, the mean mortality percentage of SF4 doxycycline 100 
(12.6%) was higher (P < 0.05) than that of SF4 (4.7%) and the electroporated control 
group (7.5%) (Table 12).  There was no difference (P > 0.05) between the mean 
mortality percentages of SF4 and the electroporated control group (Table 12).  For the 
109 to 133 hour time period, SF4 (0.8%) produced a mean mortality percentage that  
 59
Table 12.  Comparison of the mean mortality percentages at various time intervals 
(hours) for channel catfish, Ictalurus punctatus, embryos electroporated with the 
Sterile Feral (SF4) gene construct, and treated with doxycycline (100 ppm) from 4 
hours post fertilization to first hatch and a electroporated control in experiment one.  
Dead eggs and embryos were identified and removed beginning from 14 hours post 
fertilization to 153 hours post fertilization.   
Mean Mortality (%) in Hourly Intervals 
 
Treatment 
 
0-14 14-31 31-61 61-85 85-109 109-133 133-153
SF4 
 
2.8
ab
0.0
a
62.4
a
0.0
a
4.7
c
0.8
a
0.0
b
SF4 Dox 100 ppm 
 
1.0
b
0.3
a
42.0
b
0.0
a
12.6
a
0.0
a
7.2
a
Control 
 
3.5
a
0.0
a
16.3
c
0.0
a
7.5
b
0.0
a
0.8
b
a b c
 ? mean percentages with different letters within columns are different (P < 0.05), 
chi-square test. 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 21.  Comparison of the mean mortality percentages at various time intervals 
(hours) for channel catfish, Ictalurus punctatus, embryos electroporated with the 
Sterile Feral (SF4) gene construct, and treated with doxycycline (100 ppm) from 4 
hours post fertilization to first hatch and a electroporated control in experiment one.  
Dead eggs and embryos were identified and removed beginning from 14 hours post 
fertilization to 153 hours post fertilization. 
0
10
20
30
40
50
60
70
0-14 14-31 31-61 61-85 85-109 109-133 133-153
Me
a
n
 Mo
r
t
a
l
i
t
y
 P
e
r
cen
t
SF4 SF4 Dox 100 Control
 
 
 
 
 
 
 
 
 
 60
 61
was not greater (P > 0.05) than that of SF4 doxycycline 100 (0.0%) or of the control 
group (0.0%) (Table 12).  SF4 doxycycline 100 (7.2%) had a mean mortality greater 
(P < 0.05) than that of both SF4 (0.0%) and the electroporated control group (0.8%) 
during the final time interval, 133 to 153 hours (Table 12).  
In the 0 to 24 hour time period for the second experiment, the SF4 (1.3%) 
group had a lower (P < 0.05) mean mortality percentage as compared to that of SF4 
doxycycline 50 (10.2%), 100 (5.3%) and 150 (4.8%) (Table 13) (Figure 22).  There 
 
was no difference (P > 0.05) between the mean mortality percentages of SF4 and 
 
the control group (2.5%) (Table 13).  The electroporated control group had a lower (P 
< 0.05) mean mortality percentage as compared to SF4 doxycycline 50 and 100 and 
also lower (P < 0.05) than that of SF4 doxycycline 150 (Table 13). 
Mortality peaked at 24 to 60 hours.  SF4 doxycycline 50 (39.0%) and 150 
(46.1%) had lower (P < 0.05) mean mortalities as compared to SF4 (Table 13).  There 
was no difference (P > 0.05) between the mean mortality of SF4 and SF4 doxycycline 
100 (52.2%) (Table 13).  The SF4 group had higher (P < 0.05) mortality than the 
electroporated control group (22.7%) (Table 13).  The mean mortality of the control 
group was less (P < 0.05) than all doxycycline treatments (Table 13). 
In the 60 to 98 hour time period, SF4 (2.3%) had higher (P < 0.05) mean 
mortality than both SF4 doxycycline 50 (0.0%) and 150 (0.0%) (Table 13).  There 
was no difference (P > 0.05) between the mean mortality percentages of SF4 and the 
electroporated control group (2.0%) (Table 13).   
 The final time interval for the second experiment with channel catfish was 96 
to 144 hours.  During the 96 to 144 hour period, there was no difference (P > 0.05) 
between the mean mortality percentages of SF4 (1.1%) and SF4 doxycycline 50  
 
 
Table 13. Comparison of the mean mortality percentages at various time intervals 
(hours) for channel catfish, Ictalurus punctatus, embryos electroporated with the 
Sterile Feral (SF4) gene construct, and treated with different concentrations of 
doxycycline (50, 100 and 150 ppm) from 4 hours post fertilization to first hatch and a 
electroporated control in experiment two.  Dead eggs and embryos were identified 
and removed beginning from 24 hours post fertilization to 144 hours post fertilization.   
 
 62
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a b c d
 ? mean percentages with different letters within columns are different (P < 0.05), 
chi-square test. 
Mean Mortality (%) in Hourly Intervals 
 
Treatment 
 
0-24 24-60 60-96 96-144
SF4 
 
1.3
c
54.5
a
2.3
b
1.1
ac
SF4 Dox 50 ppm 
 
10.2
a
39.0
c
0.0
c
1.5
a,b
SF4 Dox 100 ppm 
 
5.3
b
52.2
a
4.3
a
0.5
bcd
SF4 Dox 150 ppm 
 
4.8
b
46.1
b
0.0
c
1.9
a
Control 
 
 
2.5
c
22.7
d
2.0
b
0.3
c
  
 
 
 
 
 
 
 
Figure 22.  Comparison of the mean mortality percentages at various time intervals 
(hours) for channel catfish, Ictalurus punctatus, embryos electroporated with the 
Sterile Feral (SF4) gene construct, and treated with three different concentrations of 
doxycycline (50, 100 and 150 ppm) from 4 hours post fertilization to first hatch and a 
electroporated control in experiment two.  Dead eggs and embryos were identified 
and removed beginning from 18 hours post fertilization to 137 hours post fertilization. 
0
10
20
30
40
50
60
0-24 24-60 60-96 96-144
M
ean
 M
o
r
t
al
i
t
y
 Pe
r
c
e
n
t
SF4 SF4 Dox 50 SF4 Dox 100 SF4 Dox 150 Control
 
 
 
 
 
 
 
 
 63
 64
(1.5%) or between SF4 and SF4 doxycycline 100 (0.5%) (Table 13).  Mortality for all 
groups was 1.9% or less.   
Deformities 
 
SF constructs versus controls 
  
The predominant deformity observed was that of scoliosis or lordosis.  
Embryos were also observed that had enlarged cardiac sacs and hearts and also with  
other varieties of posterior curvatures or abnormally formed tails.  During the first 
experiment, mean deformity percentages for SF3 (3.7%) were greater (P < 0.05) than 
that of the control group (0.3%) (Table 14).  No difference (P > 0.05) was observed in 
the deformity rates when compared between SF4 (1.5%) and control or SF4 compared 
to SF3 (Table 14).  In the second experiment, SF3 (1.3%) had a higher (P < 0.05) 
mean deformity percentage than the electroporated control group (0.1%) (Table 14).  
There was not a difference (P > 0.05) between SF4 (0.5%) and electroporated control 
or between SF3 and SF4 (Table 14).  In the third experiment, there were no 
differences (P > 0.05) between the mean deformity percentages of SF3 (0.1%) and 
SF4 (0.1%) as compared to the electroporated control group (0.0%) (Table 14).  
Again, there was no difference (P < 0.05) between the mean deformity percentages of 
SF3 and SF4 (Table 14). 
In the first experiment, SF3 had a higher (P < 0.05) mean deformity 
percentage than both CAB (0.0%) and GAD (0.0%) (Table 14).  The mean deformity 
percentage of SF4 was higher (P < 0.05) than CAB but not different (P > 0.05) than 
GAD (Table 14).  In the second experiment, deformity percentages were not different 
(P > 0.05) between SF3 and GAD (0.0%) or between SF4 and GAD (Table 14).   
SF constructs versus doxycycline  
In the first experiment, there was no difference (P > 0.05) between the mean  
Table 14.  Mean deformity percentages of channel catfish, Ictalurus punctatus, 
embryos electroporated with Sterile Feral (SF3 and SF4) gene constructs, a 
electroporated control, glutamate decarboxylase (GAD) and CAB.  Dead eggs and 
embryos were identified and removed beginning from 14 hours post fertilization up to 
153 hours post fertilization.   
a
 ? SF3 was greater than control, GAB and GAD (P < 0.05), chi-square; SF4 was 
greater than GAD and CAB (P < 0.05), chi-square. 
Mean Deformity (%) 
 
Treatment 
 
Experiment 1
a
Experiment 2
 b
Experiment 3
 
SF3 
 
3.7
 
1.3
 
0.1
 
SF4 
 
1.5 0.5 0.1 
Control 
 
0.3 0.1 0.0 
GAD 
 
0.0 0.0 - 
CAB 
 
 
0.0 - - 
b
 ? SF3 was greater than control (P < 0.05), chi-square. 
 
 
 
 
 
 
 
 
 
 65
 66
deformity percentages of SF3 (3.7%) as compared to SF3 doxycycline 100 (3.4%) 
(Table 15).  SF3 and SF3 doxycycline 100 did have higher (P < 0.05) mean deformity 
percentages than that of the mean deformity of the control (0.3%) (Table 15).  The 
mean deformity of SF3 for the second experiment (1.3%) was greater (P < 0.05) than 
the deformity rates of SF3 doxycycline 50 (0.0%), 100 (0.0%) 150 (0.0%) and control  
(0.1%) (Table 15).  There was no difference (P > 0.05) between the mean deformity 
percentages of the electroporated control group (0.1%) as compared to that of SF3 
doxycycline 50, 100 and 150 (Table 15).   
SF4-SF4 doxycycline-control 
 
There was no difference (P > 0.05) between the mean deformity percentages 
of SF4 (1.5%) and SF4 doxycycline 100 (2.7%) or between SF4 and the 
electroporated control group (0.3 %) in the first experiment (Table 16).  SF4 
doxycycline 100 did have a higher (P < 0.05) mean deformity percentage as compared 
to that of the electroporated control group (Table 16).  There was no difference (P > 
0.05) between the mean deformity percentages of SF4 (0.5%) and SF4 doxycycline 50 
(0.0%), 100 (0.0%) and 150 (0.0%) in the second experiment (Table 16).  There was 
also no difference (P > 0.05) between the mean deformity percentages of the control 
group (0.1%) as compared to SF4 doxycycline 50, 100 and 150 (Table 16).   
 
 
 
 
 
 
 
 
 
 
 
 
Table 15.  Mean deformity percentages of channel catfish, Ictalurus punctatus, 
embryos electroporated with Sterile Feral (SF3) gene construct, and treated with 
different concentrations of doxycycline (50, 100 and 150 ppm) from 4 hours post 
fertilization to first hatch and a electroporated control.  Dead eggs and embryos were 
identified and removed beginning from 14 hours post fertilization up to 153 hours 
post fertilization.   
a b c
 ? mean percentages with different letters within columns are different (P < 0.05), 
chi-square. 
Mean Deformity (%) 
 
 
Treatment 
 
Experiment 1 Experiment 2 Experiment 3 
SF3 
 
3.7
a
1.3
a
0.1
a
SF3 DOX 50 ppm 
 
- 0.0
b
- 
SF3 DOX 100 ppm 
 
3.4
a
0.0
b
- 
SF3 DOX 150 ppm 
 
- 0.0
b
- 
Control 
 
 
0.3
b
0.1
b
0.0
a
 
 
 
 
 
 
 
  
 67
Table 16.  Mean deformity percentages of channel catfish, Ictalurus punctatus, 
embryos electroporated with Sterile Feral (SF4) gene construct, and treated with 
different concentrations of doxycycline (50, 100 and 150 ppm) from 4 hours post 
fertilization to first hatch and a electroporated control.  Dead eggs and embryos were 
identified and removed beginning from 14 hours post fertilization up to 153 hours 
post fertilization.  
  
Mean Deformity (%) 
 
Treatment 
 
Experiment 1 Experiment 2 Experiment 3 
SF4 
 
1.5
ab
0.5
a
0.1
a
SF4 DOX 50 ppm 
 
- 0.0
a
- 
SF4 DOX 100 ppm 
 
2.7
a
0.0
a
- 
SF4 DOX 150 ppm 
 
- 0.0
a
- 
Control 
 
 
0.3
b
0.1
a
0.0
a
a b c
 ? mean percentages with different letters within columns are different (P < 0.05), 
chi-square. 
 
 
 
 68
 69
DISCUSSION 
 
 
 
Disrupting embryonic development 
The ability to interrupt embryonic development is contingent upon 
manipulating the expression of essential genes that are active only during 
embryogenesis.  An approach to achieve embryonic gene disruption, or knockout, is 
the use of mRNA specifically targeted for interference of gene function.  Mechanisms 
used to achieve mRNA knockout include expression of ribozymes, antisense mRNA, 
and double stranded mRNA (dsRNA) have all succeeded in several organisms 
(Thresher et al., 2001, Dunham 2004).  The sterile feral constructs investigated in this 
study employ one of these mechanisms and an additional one.  The SF3 construct uses 
dsRNA to achieve gene disruption, and the SF4 construct uses sense cDNA (Thresher 
et al., 2001). 
The Sterile Feral gene constructs, SF3 and SF4, were successful in their ability 
to interrupt embryonic development in the channel catfish, Ictalurus punctatus.  The 
mean mortality percent of the SF3 construct (78.5 %) from the three experiments 
conducted with the channel catfish was 56.1 % higher than the mean mortality percent 
of the electroporated control group (50.3%).  Likewise the SF4 construct caused a 
mean mortality percent (69.1 %) that was 37.4 % higher than that of the 
electroporated control group (50.3%).  Thus, both SF3 and SF4 demonstrated the 
potential to disrupt embryonic development at a level that was greater than the 
 70
electroporated control group showing that electroporation alone was not 
responsible for the increased mortality rate. 
 These differences would have been even greater except that the third 
experiment had a relatively high mortality rate in the electroporated blank control 
group as compared to the two other experiments.  The ability of the sterile feral 
constructs to disrupt embryonic development in the channel catfish was further 
confirmed by again causing mortality rates that were significantly greater than that of 
embryos electroporated with the glutamate decarboxylase construct (51.4%) (GAD).  
Thus, SF3 demonstrated a 52.7 % increase and SF4 a 37.4 % increase as compared to 
the mean mortality percent of GAD.  The mortality of SF3 was over 400% greater and 
SF4 was just under 400% greater than another electroporated control containing the 
CAB construct (13.9%), which was used in the first experiment.  This demonstrates 
that it was not simply the action of electroporation or the introduction of exogenous 
DNA that caused the higher mortality rates in the SF3 and SF4 groups.   
Through PCR analysis, dead and deformed embryos were confirmed to 
contain SF3 and SF4 constructs.  Based upon the PCR analysis approximately 100% 
of the embryos sampled that were treated with the SF3 or SF4 constructs were found 
to contain the bi-directional TRE element or the tTA element.  This demonstrates the 
efficacy of the electroporation in introducing the exogenous DNA into to the 
embryos. 
In terms of percent mortality, the mortality rates in this study were greater in 
magnitude but correlate with similar studies conducted on other organisms.  Thresher 
et al., 2001 used the same dsRNA blocker (SF3) and 27 of 190 (14.2%) 1-4 cell stage 
zebrafish, Danio rerio, embryos died after being injected with this dsRNA blocker.  
Four of 51 control embryos that were injected died (0.1%).  In this same study, the 
 71
SF-4 construct injected killed 141 of 234 zebra fish embryos within 24 hours post-
injection (60.3%).  Sixty-one of these individuals that were injected with the SF4 
construct were positive for the EGFP, which was a component of SF-4, and 33 
individuals were deformed.  Fifty-six of 118 control embryos died (47.5%), and only 
3 were deformed (Thresher et al., 2001).   
Thresher et al., 2001 also used three variants of two Hox genes to disrupt 
larval development in Pacific oysters.  Sense, anti-sense and double stranded RNA 
were increasingly effective in that order for disrupting development of Pacific oysters.  
The dsRNA specific to the HoxCG1 gene was the most effective, and almost 80% of 
treated embryos failed to reach the trocophore stage (Thresher et al., 2001).  In this 
same study, transfected oyster embryos with the pHSP-oHoxDS/BH plasmid and 29% 
of the embryos experienced arrested development with no heat shock and more than 
65% had arrested development with heat shock.  In this same experiment only 5% of 
control oyster embryos (not heat shocked) and 4% (heat shocked) had arrested 
development (Thresher et al., 2001). 
The disruption mechanism for the SF3 construct is based on double stranded 
mRNA (dsRNA), which has been shown to be superior in disrupting normal gene 
expression as compared to single stranded RNA (Fire et al., 1998).  In the first (79.2% 
vs. 70.6%) and second (70.1% vs. 59.2%) experiment with channel catfish, SF3 
demonstrated a higher efficacy for disrupting embryonic development than SF4.   
Deformed individuals are another indicator of disruption of embryonic 
development.  The efficacy of the SF3 construct is further substantiated by Thresher 
et al. (2001), which obtained forty-three deformed (22.6%) of the 190 zebrafish 
embryos injected with the zBMP2-dsRNA construct, which is the same disruption 
mechanism used for the SF3 construct.  In this same experiment, the sham injected 
 72
control embryos with buffer had 0 out of 37 embryos deformed and the uninjected 
controls had 1 out of 174 that was deformed.  These results correlate with the amount 
of deformities observed in the current study.  Although, no deformity rate in any 
treatment in any experiment was greater than 3.7%.  Even with this depressed 
deformity rate, the deformity rate for the sterile feral constructs was consistently 
greater than that of the electroporated control, GAD and CAB construct in the first 
two experiments.  It was only in the third experiment that the deformity rate for GAD 
surpassed that of both sterile feral constructs. 
Doxycycline concentration 
Overall, channel catfish electroporated with the SF3 construct had a higher 
percent mortality than the same group treated with doxycycline during incubation 
from 4 hours post hatch to first hatch.  The concentration of 150 ppm doxycycline 
used with channel catfish electroporated with the SF3 construct was less effective in 
reducing mean mortality percentages than both the 50 and 100 ppm concentrations.  
The relative percent mortality (56.1%) for the SF3 construct was reduced to only 
12.1% by the 50 ppm concentrations and to 5.2% by the 100 ppm doxycycline 
concentration.  The doxycycline concentration of 150 ppm had a 54.1% relative 
percent mortality, which was almost equal to that of the SF3 group.  Thresher et al., 
(2001) also noted higher mortalities with the SF3 construct when treated with 
concentrations in excess of 100 ppm.  Thresher et al. (2001) injected 52 zebrafish 
embryos with the SF3 construct and treated them with 125 ppm doxycycline and the 
results were that 25 (48.1%) died  by 24 hours post hatch.  In the same experiment, he 
injected 60 zebrafish embryos with the SF3 construct and did not treat them with 
doxycycline, and the results were that 22 (36.7%) died by 24 hours post hatch 
(Thresher et al. 2001).  In the current study, the 100 ppm concentration of 
 73
doxycycline produced the lowest mortality rates with the SF3 construct.  The relative 
effectiveness of the doxycycline treatments in the current study were higher than that 
of Thresher et al. (2001). 
The results with doxycycline were not as consistent with the SF4 construct as 
compared to the SF3 construct.  The first experiment demonstrated that 100 ppm 
doxycycline (63.0%) counteracted the SF4 construct (70.6%) by producing a lower 
mean mortality percentage.  In the second experiment, a lower mean mortality 
percentage occurred in the SF4 group (59.2%) as compared to the 50 (64.2%) and 100 
(62.3%) ppm doxycycline groups, but SF4 had a higher mean mortality percentage 
than the 150 ppm (52.8%) doxycycline group.  In the third experiment, the mean 
mortality percentage of all three doxycycline groups (64.2%, 62.3% and 52.8%) were 
lower than that of the SF4 group (84.0%).  These results, which demonstrated that the 
150 ppm doxycycline concentration to have the lowest mean mortality percent, 
contradict the previous results making the same comparisons with the mean mortality 
percent of the SF3 construct.  The possibility exists that different genes require 
different doxycycline treatments.  
The ability of the Tet-Off? system, which is the basis of doxycycline 
mediated control, to control transgene expression was also demonstrated in 
experiments with the Pacific oyster, Crassostrea gigas, (Thresher et al., 2001).  These 
experiments showed that transfected Pacific oyster embryos had expression levels 
equal to that of non-transfected embryos when they were treated with doxycycline 
(Thresher et al., 2001).  A second experiment in the same study with the same species 
also confirmed the ability of doxycycline to prevent the expression of dsRNA, for the 
doxycycline treated groups had lower fluorescence than the two non-doxycycline 
treated groups (Thresher et al, 2001).  Percent survival for control groups were 77% 
 74
and 71% while doxycycline treated groups were 95% and 92%, demonstrating the 
ability of doxycycline to suppress expression of the blocker thus allowing for a higher 
survival rate.   
Critical period of development 
 
 For the first experiment with channel catfish, the 31-61 hour post fertilization 
time period, had the highest rate of mortality when treated with SF3 or SF4.  For both 
the second and third experiment, the 18-59 hour post fertilization time period 
produced the highest mortality rates.  These varying results were due to the frequency 
of sampling and seems to indicate the sensitive time period is more accurately 
represented by the 18-61 hour post fertilization time period.  More detailed 
experimentation is needed to determine if the critical development time that is 
sensitive to the SF-3 and SF-4 constructs is narrower than the 18-61 hour time period.  
Further investigation needs to be made into this 18-61 hour time period by making 
more frequent observations within this interval in order to more narrowly define the 
most critical time. 
Conclusion   
 The current study demonstrates the potential of both sterile feral constructs to 
disrupt embryonic development.  SF3 produced a mean mortality percent that was 
56.1% higher than that of the electroporated control group.  The SF4 construct also 
showed potential by causing a 37.4% increase in the mean mortality percent relative 
to that of the electroporated control group.  In terms of effectiveness, the SF3 
construct demonstrated consistently higher mortality rates.  The ability of doxycycline 
to reduce expression of the blocker, thus allowing for normal development and 
survival, was also demonstrated.  In the study of doxycycline, the 100 ppm was the 
optimal concentration level to counteract SF3.  Overall, 150 ppm was the most 
 75
effective treatment to prevent expression of SF4.  The crucial time period for the 
action of the sterile feral constructs appears to be somewhere between 18 and 61 
hours post fertilization but further investigation also needs to be made in order to 
narrow this interval.  However, the application of the doxycycline would need to 
precede this significantly in order to counteract the disruption of the SMad5 gene, 
which was shown to be active as soon as the single cell stage in zebrafish (Thresher et 
al., 2001). 
 Further study is also needed into methods of applying the doxycycline in terms 
of commercial reproducibility and environmental sustainability.  The ability of the 
channel catfish to pass the constructs on to future offspring needs to be evaluated and 
the efficacy of these constructs in a transgenic fish needs to be determined.  The goal 
of achieving an effective and reversible sterilization technique appears feasible based 
on our initial results.   
  
 
 
 
 76
 
 
 
 
 
 
REFERENCES 
 
Abdul Razak, S., Hwang, G.L., Rahman, M.A., Maclean, N.  1999.  Growth 
performance and gonadal development of growth enhanced transgenic 
tilapia, Oreochromis niloticus, following heat-shock-induced triploidy.  
Marine Biotechnology.  1, 533-544. 
 
Akhtar, N.  1984.  Anesthesia, abdominal surgery and efficacy of ovariectomy and 
subsequent androgen treatments in inducement of sex reversal in Tilapia 
nilotica.  PhD dissertation, Auburn University, Auburn, Alabama, USA. 
 
Bart, A.N.  1994.  Effects of sperm concentration, egg number, fertilization 
method, nutrition and cryopreservation of sperm on fertilization efficiency 
with channel catfish eggs and blue catfish sperm.  PhD dissertation, Auburn 
University, Alabama, USA. 
 
Chatakondi, N.G.  1995.  Evaluation of transgenic common carp, Cyprinus carpio, 
containing rainbow trout growth hormone in ponds.  PhD dissertation, 
Auburn University, Auburn, Alabama, USA. 
 
Chitmanat, C.  1996.  Predator avoidance of transgenic channel catfish containing 
salmonids growth hormone genes.  MSc thesis, Auburn University, Auburn, 
Alabama, USA. 
 
Chourrout, D., Chevassus, B., Krieg, F., Happe, A., Burger, G., Renard, P.  1986.  
Production of second generation triploid and tetraploid rainbow trout by 
mating tetraploid males and diploid females:  potential of tetraploid fish.  
Theoretical and Applied Genetics.  72, 193-206.   
 
Cook, J.T., McNiven, M.A., Richardson, G.F., Sutterlin, A.M.  2000.  Growth 
rate, body composition and feed digestibility/conversion of growth-
enhanced transgenic Atlantic salmon (Salmo salar).  Aquaculture.  188, 15-
32. 
 
Cotter, D., O?Donovan, V., O?Maoil?idigh, N., Rogan, G., Roche, N., Wilkins, 
N.P.  2000.  An evaluation of the use of triploid Atlantic salmon, Salmo 
salar, in minimizing the impact of escaped farmed salmon on wild 
populations.  Aquaculture.  186, 61-75. 
 
Curtis, T.A., Sessions, F.W., Bury, D., Rezek, M., Dunham R.A.  1987.  Induction 
of polyploidy in striped bass, white bass and their hybrids with hydrostatic 
pressure.  Proceedings Annual Conference Southeastern Association of Fish 
Wildlife Agencies.  41, 63-69. 
 77
 
Devlin, R.H., Biagi, C.A., Yesaki, T.Y.  2004.  Growth, viability and genetic 
characteristics of GH transgenic coho salmon strains.  Aquaculture.  236, 
607-632. 
 
 Dunham, R.A.  1995.  Predator avoidance, spawning and foraging ability of 
transgenic catfish.  In:  Levin, M., Grim, C. and Angle, J.S.  (eds)  
Proceedings Biotechnology Risk Assessment Symposium, 22-24 June 1994.  
USDA, EPA, College Park, Maryland, pp. 151-169. 
 
Dunham, R.A.  1996.  Contribution of genetically improved aquatic organisms to 
global food security.  In:  International Conference on Sustainable 
Contribution of Fisheries to Food Security.  Government of Japan and FAO, 
Rome, Italy, 50 pp. 
 
Dunham, R.A.  1999.  Utilization of transgenic fish in developing countries:  
potential benefits and risks.  Journal of the World Aquaculture Society.  30, 
1-11. 
 
Dunham, R.A.  2003.  Status of genetically modified (transgenic) fish:  research 
and application.  
http://www.fao.org/es/ESN/food/risk_biotech_animal_en.stm 
 
Dunham, R.A.  2004.  Aquaculture and Fisheries Biotechnology:  Genetic 
Approaches.  CABI Publishing, Cambridge, MA, 372 pp. 
 
Dunham, R.A., Chatakondi, N., Nichols, A., Chen, T.T., Powers, D.A., Kucuktas, 
H.  2002a.  Survival of F
2 
transgenic common carp, Cyprinus carpio, 
containing pRSVrtGH
1 
complementary DNA when subjected to low 
dissolved oxygen.  Marine Biotechnology.  4, 323-327. 
 
Dunham, R.A., Chatakondi, N., Nichols, A.J., Kucuktas, H., Chen, T.T., Powers, 
D.A., Weete, J.D., Cummins, K., Lovell, R.T.  2002b.  Effect of rainbow 
trout growth hormone complementary DNA on body shape, carcass yield, 
and carcass composition of F
1
 and F
2
 transgenic common carp, Cyprinus 
carpio.  4, 604-611. 
 
Dunham, R.A., Devlin, R.  1998.  Comparison of traditional breeding and 
transgenesis in farmed fish with implications for growth enhancement and 
fitness.  In:  Murray, J.D., Anderson, G.B., Oberbauer, A.M. and 
McGloughlin, M.N.  (eds) Transgenic Animals in Agriculture.  CAB 
International, Wallingford, UK, pp. 209-229. 
 
Dunham, R.A., Ramboux, A.C., Duncan, P.L., Hayat, M., Chen, T.T., Lin, C.M., 
Kight, K., Gonzalex-Villasenor, I., Powers, D.A.  1992.  Transfer, 
expression, and inheritance of salmonids growth hormone genes in channel 
catfish, Ictalurus punctatus, and effects on performance traits.  Molecular 
Marine Biology and Biotechnology.  1, 380-389. 
 
 78
Dunham, R.A., Warr, G.W., Nichols, A., Duncan, P.L., Argue, B., Middleton, D., 
Kucuktas, H.  2002c.  Enhanced bacterial disease resistance of transgenic 
channel catfish, Ictalurus punctatus, possessing cecropin genes.  Marine 
Biotechnology.  4, 338-344. 
 
Engle, C.  2003.  The evolution of farm management, production efficiencies, and 
current challenges to catfish production in the United States.  Aquaculture 
Economics and Management.  7, 67-84.  
 
FAO/WHO.  2004.  Safety Assessment of Foods Derived from Genetically 
Modified Animals, including Fish.  FAO Food and Nutrition Paper 79.  
Report of the FAO/WHO Expert Consultation Rome, 17-21 November 
2003.  World Health Organization, Food and Agriculture Organization of 
the United Nations, Rome. 
 
Farrell, A.P., Bennett, W., Devlin, R.H.  1997.  Growth-enhanced transgenic 
salmon can be inferior swimmers.  Canadian Journal of Zoology.  75, 335-
337. 
 
Gossen, M., Bujard, H.  1992.  Tight control of gene expression in mammalian 
cells by tetracycline-responsive promoters.  Proc. Natl. Acad. Sci. USA.  89, 
5547-5551.  
 
Gossen, M., Freundlied, S., Bender, G., Muller, G., Hillen, W., Bujard, H.  1995.  
Transcriptional activation by tetracycline in mammalian cells.  Science.  
268, 1766-1769.  
 
Hallerman, E.M, Kapuscinski, A.R.  1993.  Potential impacts of transgenic and 
genetically manipulated fish on wild populations:  addressing the 
uncertainties through field testing.  In:  Cloud, J.G., Thorgaard, G.H. (Eds.) 
Genetic Conservation of Salmonid Fishes.  Plenum Press, New York, 93-
112. 
 
Harvey, D.J.  Aquaculture Outlook.  United States Department of Agriculture.  
Electronic Outlook Report from the Economic Research Service.  
http://usda.mannlib.cornell.edu/reports/erssor/livestock/ldp-
aqs/2004/aqs20.pdf.  Oct. 8, 2004.  1-26. 
 
Hedrick, P.W.  2001.  Invasion of transgenes from salmon or other genetically 
modified organisms into natural populations.  Canadian Journal of Fisheries 
and Aquatic Sciences 58, 841-844. 
 
Kistner, A., Gossen, M., Simmermann, F., Jerecic, J., Ullmer, C., Lubbert, H., 
Bujard, H.  1996.  Doxycycline-mediated quantitative and tissue-specific 
control of gene expression in transgenic mice.  Proceedings of the National 
Academy of Sciences, USA.  93, 10933-10938. 
 
 
 79
Kriegler, M.  1990.  Assembly of enhancers, promoters, and splice signals to 
control expression of transferred genes.  Methods in Enzymology.  185, 
512-527. 
 
Mananos, E.L., Anglade, I., Chyb, J., Saligaut, C., Breton, B., Kah, O.  1999.  
Involvement of ?-aminobutyric acid in the control of GtH-1 and GtH-2 
secretion in male and female rainbow trout.  Neuroendocrinology.  69, 269-
280. 
 
Mansueti, A.J., Hardy Jr., J.D.  1967.  Development of Fishes of the Chesapeake 
Bay Region:  An Atlas of Egg, Larval, and Juvenile Stages.  Edited by Earl 
E. Deubler, Jr.  Published by the Natural Resources Institute, University of 
Maryland.  Port City Press, Baltimore, Maryland.  
 
Melamed, P., Gong, Z., Fletcher, G., Hew, C.L.  2002.  The potential impact of 
modern biotechnology on fish aquaculture.  Aquaculture.  204, 255-269. 
 
Muir, W.M., Howard, R.D.  1999.  Possible ecological risks of transgenic 
organism release when transgenes affect mating success:  sexual selection 
and the Trojan gene hypothesis.  Proceedings of the National Academy of 
Science, USA.  96, 13853-13856. 
 
Muir, W.M., Howard, R.D.  2001.  Fitness components and ecological risk of 
transgenic release:  a model using Japanese medaka, Oryzias latipes.  
American Naturalist.  158, 1-16. 
 
Niiler, E.  1999.  Terminator technology temporarily terminated.  Nature 
Biotechnology.  17, 1054. 
 
Palmiter, R.D., Brinster, R.L., Hammer, R.E., Trumbauer, M.E., Rosenfeld, M.G., 
Birnberg, N.C., Evans, R.M.  1982.  Dramatic growth of mice that develop 
from eggs microinjected with metallothionein-human growth hormone 
fusion genes.  Nature.  300, 611-615. 
 
Pandian, T.  2003.  Recent advances in hormonal induction of sex-reversal in fish.  
Journal of Applied Aquaculture.  13, 205-230. 
 
Powers, D.A., Hereford, L., Cole, T., Chen, T.T., Lin, C.M., Kight, K., Creech, 
K., Dunham, R.A..  1992.  Electroporation:  a method for transferring genes 
into the gametes of zebrafish, Brachydanio rerio, channel catfish, Ictalurus 
punctatus, and common carp, Cyprinus carpio.  Molecular Marine Biology 
and Biotechnology.  1, 301-308. 
 
Sarmasik, A., Warr, G., Chen, T.T.  2002.  Production of transgenic medaka with 
increased resistance to bacterial pathogens.  Marine Biotechnology.  4, 310-
322. 
 
 80
Stanley, J.G.  1979.  Control of sex in fishes, with special reference to the grass 
carp.  In:  Shireman, J.V. (ed.) Proceedings Grass Carp Conference.  
University of Florida Press, Gainesville, Florida, p.21. 
 
Sun, X., Liang, L., Yan, X., Cao, D.  2002.  The faster growth effect of transgenic 
common carp with growth hormone gene and its progenies.  Journal of 
fisheries of China/Shuichan Xuebao.  26, 391-395. 
 
Thresher, R., Hinds, L., Hardy, C., Whyard, S., Vignarajan, S., Grewe, P., Patil, J.  
2001.  Repressible Sterility of Animals.  International Application Published 
Under the Patent Cooperation Treaty.   
 
Underwood, J.L., Hestand, R.S. III, Thompson, B.Z. 1986.  Gonad regeneration in 
grass carp following bilateral gonadectomy.  Progressive Fish-Culturist.  48, 
54-56. 
 
Uzbekova, S., Chyb, J., Ferriere, F., Bailhache, T., Prunet, P., Alestrom, P., 
Breton, B.  2000.  Transgenic rainbow trout expressed sGnRH-antisense 
RNA under the control of sGnRH promoter of atlantic salmon.  Journal of 
Molecular Endocrinology.  25, 337-350. 
 
Wang, R., Zhang, P., Gong, Z., Hew, C.L.  1995.  Expression of the antifreeze 
protein gene in transgenic goldfish, Carassius auratus, and its implication in 
cold adaptation.  Molecular Marine Biology and Biotechnology.  4, 20-26. 
 
Yin, D.X., Schimke, R.T.  1995.  BCL-2 expression delays drug-induced 
apoptosis but does not increase clongenic survival after drug treatment in 
HeLa cells.  Cancer Research.  55, 4922-4928. 
 
Young, B.A., Bryan, M.B., S.A. Sower, Li, W.  2004.  The effect of 
chemosterilization on sex steroid production in male sea lampreys.  
Transactions of the American Fisheries Society.  133, 1270-1276.  
 
Zbikowska, H.M.  2003.  Fish can be first ? advances in fish transgenesis for 
commercial applications.  Transgenic Research.  12, 379-389.   
 
Zhu, Z.  1992.  Generation of fast growing transgenic fish:  methods and 
mechanisms. In:  Hew, C.L., Fletcher, G.L. (Eds.), Transgenic Fish.  World 
Scientific Publishing, Singapore, pp. 92-119.  
 
Zohar, Y., Mylonas, C.C.  2001.  Endocrine manipulations of spawning in 
cultured fish: from hormones to genes.  Aquaculture.  197, 99-136. 
 
 
 81
APPENDIX A 
 
Holzfretter?s Solution
 
3.5 g NaCl 
 
0.05g KCl 
 
0.1g CaCl 
 
0.2g NaHCO
3 
 
Dissolve ingredients in 1L dd H
2
O 
 82
 
APPENDIX B 
 
PAB construct 
This construct employs antisense mRNA salmon type GnRH to disrupt the 
production of the neuromediator, GnRH, which is essential to proper gonadal 
development  (Zohar and Mylonas, 2001) (Figure 23).  The first section of this construct 
(figure 3) is the salmon sGnRH Pab promoter.  The next sequence is the SD/SA, which is 
an acceptor signal fused to the promoter.  This section is followed by the GnRH antisense 
gene which is ultimately responsible for the disruption of GnRH production and/or 
utilization (Uzbekova et al., 2000).  The final section of this construct is the polyA SV40 
which terminates transcription.  The total size of this construct is 3,066 bp. 
PH3 Construct 
This construct employs antisense mRNA salmon type GnRH in order to disrupt 
the production of the neuromediator, GnRH, which is essential to proper gonadal 
development  (Zohar and Mylonas, 2001).
Figure 23.  Pab-sGNRH-antisense construct.  (taken from Uzbekova et al., 2000). 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 83
Table 17.  Mean mortality of channel catfish, Ictalurus punctatus, embryos electroporated 
with Sterile Feral (SF3 and SF4) gene constructs, PH3, PAB and a electroporated control 
in experiment one.  Dead eggs and embryos were identified and removed beginning from 
14 hours post fertilization up to 153 hours post fertilization. 
 
 84
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a,b,c
 - mean percentages with different letters within columns are different (P < 0.05), chi-
square. 
Mean Mortality (%) 
 
Treatment 
 
 
Experiment 
 1
 
SF3 
 
79.2
a
SF4 
 
70.6
b
Control 
 
28.1
c
PAB 
 
31.9
c
PH3 
 
31.2
c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 24.  Mean mortality of channel catfish, Ictalurus punctatus, in experiment one. 
Embryos were electroporated with the Sterile Feral (SF3 & SF4) gene constructs, PAB, 
PH3 and a electroporated blank control.  Dead eggs and embryos were identified and 
removed beginning from 14 hours post fertilization up to 153 hours post fertilization. 
0
10
20
30
40
50
60
70
80
90
Me
a
n
 Mo
r
t
a
l
i
t
y
 P
e
r
cen
t
SF3 SF4 Control PAB PH3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 85
Table 18.  Mean mortality of channel catfish, Ictalurus punctatus, embryos electroporated 
with the Sterile Feral (SF3 & SF4) gene constructs treated with doxycycline (100 ppm) 
from 4 hours post fertilization to first hatch, PAB, PH3 and a electroporated control in 
experiment one.  Dead eggs and embryos were identified and removed beginning from 14 
hours post fertilization up to 153 hours post fertilization.  
 
Mean Mortality (%) 
 
Treatment 
 
 
Experiment 
 1
 
SF3 Dox 100 
 
47.3
b
SF4 Dox 100 
 
63.0
a
Control 
 
28.1
c
PAB 
 
31.9
c
PH3 
 
31.2
c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a,b,c
 - mean percentages with different letters within columns are different (P < 0.05), chi-
square. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 86
 
Figure 25.  Mean mortality of channel catfish, Ictalurus punctatus, embryos 
electroporated with the Sterile Feral (SF3 & SF4) gene constructs treated with 
doxycycline (100 ppm) from 4 hours post fertilization to first hatch, PAB, PH3 and a 
electroporated control in experiment one.  Dead eggs and embryos were identified and 
removed beginning from 14 hours post fertilization up to 153 hours post fertilization. 
0
10
20
30
40
50
60
70
Me
a
n
 Mo
r
t
a
l
i
t
y
 P
e
r
cen
t
SF3 Dox SF4 Dox Control PAB PH3
 
 
 87