Project Work Plan
Department of Interior USGS GE PES
Fiscal Year 2014 Study Work Plan
Study Title: Development of Trojan Y technology to control invasive fishes in the Greater Everglades
Current Study Start Date: 1 Oct 2013
Current Study End Date: 30 Sept 2015 with possibility of future funding tied to progress
Location (Subregions, Counties, Park, or Refuge): The work for this project will take place in the lab in FY 2014; however, the potential ramifications of the research extends system-wide and include Everglades National Park, Big Cypress National Preserve, Biscayne National Park and various state and private lands. If we are successful in developing the Trojan Y technology in a live model, it could be extended to other species of invasives across the Nation.
Funding Source: GEPES
Funding History: PES since 2012
FY14 USGS Funding: FY14
Principal Investigator: Pamela J. Schofield, US Geological Survey, Gainesville, FL
USGS Project Officer:
USGS Technical Officer:
Supporting Organizations: none
Overview & Objective(s):
Statement of Problem: Within Florida, dozens of foreign non-native fishes have established self-sustaining populations. There is concern that these introduced species could negatively affect native communities by predation, competition, or by serving as vectors for disease. Once a non-native fish has become established, there are almost no methods available to eradicate it or control its population size. New technologies must be developed to control non-native species, and help stop their spread. In this project, we are investigating the applicability of a theoretical model of invasive species control using Trojan Y sex chromosomes.
The Trojan Y technique of population control for invasive species was originally described by Gutierrez and Teem (2006) and consists of adding sex-reversed females containing two Y chromosomes (YY females) to a wild population to skew the sex ratio of subsequent generations to contain an increasing number of males (i.e., fewer and fewer females in each generation). The gradual reduction in females may lead to eventual extinction of the population. The potential of this technique for eradicating or controlling non-native species has been much-discussed and promoted (e.g., Cotton and Wedekind 2007; Teem et al. 2013; Thresher et al. 2013). However, until this project, no one has attempted to move beyond theory into practical application. The development of this technique requires several years of experimentation and exploration. Our project is the first in the world to attempt to adapt Trojan Y technology to a living model.
Trojan Y is favored over other techniques (e.g., genetic engineering) for the following reasons:
- It is kinder on the environment than poisoning, which kills all (or most) species, including natives. With Trojan Y, only the target species is affected.
- It is safer than genetic modification, because no genetically engineered genes are being transferred to subsequent generations.
- It is reversible. If the influx of Trojan Y fish into the environment is stopped (before the population is extinct) the effect will be stopped in a few generations.
- It is flexible. The strength of the effect is based on the proportion of treated individuals added. So it can be used for anything from mild control to complete eradication.
- Does not require massive population influxes, as with sterile male techniques. A one-time introduction of a small number of YY females (ca. 3% of the population size) can skew the population towards males for several generations.
To date, this theoretical construct has never been demonstrated with a population of organisms. Our objective is to develop the utility of this approach and test its applicability in a real-world setting.
To do this, two broad phases must be accomplished: Phase 1) YY females must be developed in the laboratory, then Phase 2) their interaction with other individuals and the population consequences of these must be assessed. More information on these is below.
Specific Relevance to Major Unanswered Questions and Information Needs Identified: At present, the only management techniques available to control non-native fishes are physical removal (e.g., electroshocking), dewatering or ichthyocides. Unfortunately, all of these methods negatively impact native fauna as well as the targeted non-native fishes and require a great deal of effort (and therefore, funding). Herein, we propose development and testing of a genetic technique (Trojan Y) to control non-native fishes. Currently, the concept of using Trojan Y technology to control non-native species exists solely as a theoretical construct. Our project is the first to attempt application of this theory to the real world (in this case, the Greater Everglades). If we are successful in development of the technology, it could potentially be applied to a wide variety of species, including other fishes (e.g., brown hoplo Hoplosternum littorale), invasive applesnails (Pomacea spp.), the Australian red claw crayfish (Cherax spp.) and the green mussel (Perna viridis).
Planned Products for FY14: We are still in data-collection mode, but should have a few peer-review journal articles out next year.
Work Plan - African jewelfish
See FY 13 progress report for details of our species selection (guppy and African jewelfish) and a breeding chart showing the steps involved in breeding a YY female. The FY 13 report also provides detailed discussion of the problems we faced with the first batch of the F2 generation (sex ratios highly skewed towards males).
We continue to work on discriminating between the F1XX females (which are not useful to us) and the F1XY females (which we need to continue on the path of developing the YY female). The difficulty in discriminating F1XX from F1XY females is that they all are phenotypically female. Thus, we cannot tell the difference in the genetic makeup of the fish based on anatomical characters. We have been working on this problem using two different approaches (breeding confirmation and searching for a genetic marker):
We should to be able to determine the F1 mother's genetic makeup (XX or XY) by the % males she produces (F2 generation). F1XX females would produce 50% males, F1XY females would produce 75% males. Early breeding confirmation studies were stymied when we reared batches that did not conform to either of these expected sex ratios.
See FY 13 report for details of problems with the first round of F2s. In summary, most batches were highly skewed towards males and did not conform with expected proportions of males:females. Clearly, something was playing havoc with sex determination in our lab. To discern what that may be, we tested a few different hypotheses:
Differential survival favoring males: Over the years while working on this project we have continually refined our fish-husbandry/fish-rearing techniques. We continue to do so, increasing the number of fry that survive to adulthood. When rearing the wild batches for this task, we increased time and effort spent caring for the young and carefully followed survival by regularly assessing the number of fish in each batch. This allowed us to determine that very few fish were dying during the rearing process (survival is ca. 90%). This small amount of mortality cannot explain the prevalence of males in our batches. Furthermore, in FY 13 we split large broods of fry (ca. 2,000 individuals) into multiple tanks and reared them to adulthood in various tank configurations, both inside and outside in our mesocosm facility. Sex ratios differed between broods, but were the same across tanks that came from the same brood, regardless of rearing conditions. Thus, we are satisfied that there is no differential mortality between genders in our lab and do not need to examine this hypothesis further.
Temperature influence on sex determination: In FY 13 we began carefully-controlled breeding experiments, wherein we breed several batches of fish at a series of different (constant) temperatures. We had a hard time keeping the water temperature constant, especially during cold fronts. Regardless of equipment issues and interruptions due to the furlough, we were able to breed three batches at relatively constant temperatures and our results were surprising. It is really looking like temperature affects sex determination in African jewelfish, and we will continue on with experiments in this vein.
We consider these data as preliminary, as we would prefer to have more constant temperatures. This was a problem, because we need flow-through well-water to rear the fry. However, the temperature of the flow-through water in the lab is highly variable, and changes with the outside air temperature. However, our fabulous facility guys recently developed a system that will allow us to use flow-through well water while maintaining constant temperatures. Well water flows into a large, insulated sump, where it is heated to a constant temperature (adjustable via digital thermostat). Water is constantly pumped from this reservoir into breeding tanks.
In FY 14, we will use our new system to breed several batches of fish at different temperatures. Offspring will be old enough to sex in early FY 15. This information is vital to continued work with this species in our lab, as we must know what the limits are to determine the appropriate window of temperatures for breeding, so we do not unintentionally influence outcomes of sex ratios. Furthermore, we intend to publish these data as a peer-review journal article. Temperature-mediated sex determination is relatively rare in fishes, and especially rare in freshwater fishes.
Also in FY 14, we will re-breed F1 females that previously produced skewed sex ratios. We will use our new constant-temperature lab system and the knowledge that maintenance of temperatures between 26 and 28 should not skew sex ratios. We aim to re-breed at least six of the F1 females that we previously bred and rear their fry, which should be sexually mature in early FY 15. These offspring should be either 50% male (indicating the mother was XX) or 75% male (indicating the mother was XY).
Maggie Hunter (USGS-Gainesville) has been assisting us by looking for a genetic marker to discriminate between XX and XY F1 females. Dr. Hunter has focused on developing a marker that indicates whether or not a particular individual contains a Y chromosome.
In FY13, approximately 100 RAPD primers were tested through 50 PCRs and gel electrophoresis on pools of jewelfish DNA from 30 individuals. All putative sex-specific bands were tested on eight individuals to determine sex-determination consistency. Although no sex-specific marker has been identified to date, testing of additional primers will continue. To date, over 170 RAPD primers have been tested. In other species, more than 200 primers were tested before deterministic primers are identified, so we are hopeful in our efforts.
To increase the types of markers being used, microsatellites and sex-determining loci were also tested. A total of five microsatellites were tested, however, due to the markers being developed in tilapia, they had poor cross-species amplification and no clearly definable sex-specific pattern. The sex-determining genes were also successfully PCR amplified and no difference was found between males and females at three genes.
IN FY 14, we will continue testing RAPD primers and microsatellites. We will also develop and sequence Restriction site Associated DNA (RAD) markers, using next generation sequencing. These markers provide genome-wide coverage at tens of thousands of genomic loci and may identify loci that show differences between the two sexes.
We need to determine whether African jewelfish uses a XY or a ZW sex-determination system. In species with an XY sex determination system, the female is homogametic (XX), while the male is heterogametic (XY). Thus, it is the male who determines gender of the offspring. In ZW systems, this is reversed - the female is heterogametic (ZW) and the male is homogametic (ZZ). The prevalence of ZW sex-determination systems is not well-understood, and, curiously, even closely related species in the same genus can have different sex-determination systems. For example, the Nile tilapia (Oreochromis niloticus) is XY, but Mozambique tilapia (Oreochromis mossambicus) is ZW. For the Trojan Y theory to work, the candidate species must be XY.
One way to determine whether African jewelfish is XY or ZW is to do a gynogenesis experiment, wherein haploid eggs are made into diploids. The process makes an exact copy of the haploid genome and doubles it to create a diploid zygote. These zygotes can then be reared to adulthood and sexed. The genders of the gynogens (i.e., offspring) reveal important information about the parents' sex-determination system. Females from an XY sex-determination system are XX and can only produce X eggs. When these are diploidized, they result in 100% female (XX) offspring. Females from a ZW sex-determination system are ZW, and can produce either Z or W haploid eggs. When diploidized, these produce approximately equal numbers of males (ZZ) and females (WW). The process of gynogenesis is complex, highly technical, and can be somewhat unreliable. It will require us to develop new techniques for expressing eggs from jewelfish and rearing non-adhesive (e.g., free-floating) eggs in our laboratory. Furthermore, previous researchers that have been successful with gynogenesis have reported overall low survival for gynogens (up to 8%). This is because any recessive deleterious genes in the haploid egg are copied and expressed in the diploid gynogen. This is a challenging, yet exciting sub-task that - if successful - will provide valuable information about our model species. Additionally, if our gynogenesis experiment is successful, we will be able to publish the results as a journal article as no one has ever attempted gynogenesis with this species.
Already in FY 14 we have made big strides in our ability to strip eggs and sperm from African jewelfish in the lab, and rear the resulting progeny. We hope to be able to complete the entire gynogenesis experiment in FY 14.
Work plan - Guppies
Progress on developing the African jewelfish has been slower than expected due to the aforementioned unforeseen complications. We are not daunted by these, because they are all items that we can overcome with more research. Furthermore, one of the primary goals of this project was to determine the applicability of the theoretical construct of using Trojan Y fish. The impediments and complications we have experienced provide important information not just for our own laboratory progress, but to other researchers who may want to apply Trojan Y technology in the future.
Nonetheless, we have spent all our energy to date on Phase 1 of the project (development of the YY female). We are eager to work on Phase 2, wherein we will assess applicability of Trojan Y in a model population. Once the YY female has been developed in the lab, it is important to determine its social compatibility. The Trojan Y model assumes that, other than genetics, all aspects of life-history and behavior are equivalent between wild and YY females. It is important to evaluate whether these model assumptions are correct, and if not to adjust the model accordingly.
We aim to answer the following questions:
- Are YY females equally as attractive to males as wild females? Or is there differential mating?
- Do life-history factors for YY females differ from wild females? Are they as fecund? As long-lived?
- Do the offspring of YY females have equivalent survivability when compared to wild offspring?
Furthermore, we are eager to begin some small-scale experiments, where YY females are added to a wild population. How will the overall population size change over several generations? How will the ratio of males to females change through time? Again, these are questions that have been modeled statistically but no application has ever been developed in a living model species.
To get to Phase 2 faster, we have added a new species to our project: the guppy (Poecilia reticulata). Our reasons for choosing this species include:
- The species is known to be XY (not ZW).
- YY males and females have previously been made in the laboratory for this species.
- There is published genetic information available.
- It is a small species with a short time to maturity (30 days). The small size allows us to hold many individuals in the laboratory.
- Temperature effects on sex ratios of offspring have already been documented for this species.
In FY 14, we will continue to refine our breeding protocols with guppies. We are hopeful that by the end of FY 14 we will have the F1 and F2 generations in hand (i.e., steps 1-4 below).
Steps to make a YY female guppy:
1) Test published genetic markers to determine whether or not they work for discrimination of Y marker for our guppies. To do this, we will simply compare XY wild males and XX wild females.
2) Breed the F1 population. The parents will consist of wild females mated to wild males. Guppies are live-bearers, and sexual differentiation occurs while the larvae are still inside the mother's womb. Therefore, to feminize the offspring, the pregnant females will be fed with estrogen-treated food.
3) Rear F1 generation to maturity. Confirm 100% phenotypic females.
4a) If the genetic marker works (we will have previously tested it on known XY males), we will be able to discriminate XX from XY F1 females. In this case, we will breed F1 XY females to wild males to produce the F2 generation.
4b) If the genetic marker is not ready, we will simply select several F1 females and breed them to wild males. The offspring will be reared to maturity and sexed. We should be able to determine genetic composition of the mothers by evaluating the ratio of males to females.
5) F2 batches that are from F1 XY mothers (e.g., batches that are 75% male), will contain 25% females (which will be discarded), 50% XY males and 25% YY males. Our next task is to discern the YY males from the XY males. To do this, we will breed a selection of the males to wild females and rear the batches of young to maturity. Offspring of YY males will only be male (e.g., 100% XY).
6) Once YY males have been identified, they can be back-crossed to F1 XY females and the resulting batches feminized. The batches will be 50% YY female and 50% XY female.
7) Subsequent progeny testing will discriminate YY females from XY females.
8) Once a few YY females have been created, they can be back-crossed to YY males and the subsequent progeny feminized. This will produce batches of 100% YY females.
Cotton, S. and C. Wedekind. 2007. Control of introduced species using Trojan sex chromosomes. TRENDS in Ecology and Evolution 22: 441-443.
Gutierrez, J. B. and J. L. Teem. 2006. A model describing the effect of sex-reversed YY fish in an established wild population: The use of a Trojan Y chromosome to cause extinction of an introduced exotic species. Journal of Theoretical Biology 241: 333-341.
Teem, J. L., J. B. Gutierrez and R. D. Parshad. 2013. A comparison of the Trojan Y chromosome and daughterless carp eradication strategies. Biological Invasions 10.1007/s10530-013-0475-2.
Thresher, R. E., K. Hayes, N. J. Bax, J. Teem, T. J. Benfey, F. Gould. 2013. Genetic control of invasive fish: technological options and its role in integrated pest management. Biological Invasions 10.1007/s10530-013-0477-0.