Doctor of Philosphy
School of Biological Sciences
Roberts, David G., Assessment of the frequency and consequences of hybridisation between an obligately estuarine and a migratory marine fish, Doctor of Philosphy thesis, School of Biological Sciences, University of Wollongong, 2010. http://ro.uow.edu.au/theses/3626
Populations of obligately estuarine species (i.e., species that complete their entire life-cycle in estuaries) are potentially small and isolated and may lack genetic variation and display regional differentiation as a result of genetic drift and inbreeding. Low levels of hybridisation with a migratory marine congener should introduce genetic variation and reduce the effects of inbreeding depression and drift. However, high levels of hybridisation can have negative impacts including breakdown of localised adaptation, and demographic and genetic swamping and loss of species identity. Although hybridisation between marine and estuarine species has rarely been considered, estuaries may be hot spots of hybridisation and introgression may be exacerbated by anthropogenic disturbances, including fishing and mechanical modification of the frequency of estuary flushing. In south east Australia, the estuarine Black bream Acanthopagrus butcheri, and its migratory marine congener Yellowfin bream, A. australis, have overlapping distributions and spawning times and therefore potential to hybridise within estuaries. Here I use a combination of molecular population genetics and breeding experiments to determine population structure of A. butcheri and A. australis, and the frequency of occurrence and consequences of hybridisation for the two species but particularly the estuarine A. butcheri.
I used a broad scale survey of microsatellite and mitochondrial DNA (mtDNA) variation in 565 adult fish from 25 estuaries spanning the distributional range of the two species, to initially characterise the species and their putative hybrids, then to test the prediction that hybrids are restricted to the area of sympatry. Hybrids were widespread (68% of estuaries studied) and hybrid frequencies varied greatly among estuaries (0 to 58%). Most hybrids (88%) were classed as advanced generation backcrosses with A. butcheri and displayed A. butcheri mtDNA haplotypes. I found most hybrids (57%) in the three estuaries within the area of sympatry, though hybrids were detected throughout the ranges of the two species.
Acanthopagrus australis is considered highly dispersive and this may have major implications for A. butcheri, although A. australis is also known to spawn in close association with estuaries which suggests populations could be genetically subdivided. I investigated the genetic structure of A. australis using both nuclear (microsatellite data) and mtDNA. My data from surveys of allele frequencies at six microsatellite loci for 350 fish, revealed high levels of genetic diversity within all sites but no genetic differentiation of groups collected from sites separated by up to 50 km (FST = 0.002, P > 0.05) and no differentiation of adults and juveniles spread across the distributional range of the species (several 100’s of kms, FST = 0). I obtained similar results from analysis of sequence of the mtDNA control region for a subset of 47 fish (pairwsie FST = 0.000 – 0.036, P > 0.05). Acanthopagrus australis form a single panmictic population on the east coast with genetic homogeneity reflecting the predicted active northwards dispersal of adults to spawn and the southwards dispersal of larvae affected by the Eastern Australian Current.
The surprisingly high number of hybrid fish in estuaries in the area of sympatry could reflect ancient hybridisation events with little continuing introgression of the A. australis genome. Alternatively, Acanthopagrus hybrid zones may be dynamic and reflect contemporary processes. I used mtDNA sequence variation in a range wide phylogeographic study of A. butcheri, to test the prediction that populations within different estuaries and geographical regions are genetically differentiated and that eastern and western A. butcheri represent divergent genetic lineages. Unlike many species historically fragmented by the Bassian Isthmus land bridge that form deeply divergent eastern and western lineages or incipient species, A. butcheri form closely related eastern and western lineages that diverged no later and perhaps considerably earlier (owing to uncertainty surrounding the rate at which mitochondrial control region DNA mutate) than the middle Pleistocene, leaving open the possibility that Acanthopagrus hybrid zones are maintained by contemporary processes.
As for many pairs of obligately estuarine and migratory marine fish, the degree or extent of reproductive (gamete) compatibility is unknown for A. butcheri and A. australis, though this will influence the likelihood of initial and later hybridisation. I compared the fertilisation success achieved when ova of estuary caught A. butcheri females were crossed with the semen of either ocean-caught A. australis males or estuary-caught A. butcheri males, to test for early acting barriers to initial and later hybridisation. The experimental crosses, which by chance included the two pure species and hybrid fish, revealed no evidence that gametic incompatibility provides a barrier to fertilisation among both pure species and their hybrids. Sperm of both A. australis and A. butcheri, and hybrid males, were apparently equally compatible with ova of A. butcheri and hybrid females (~ 40% viable larvae), for the single sperm concentration that was employed in the experiment.
The coastal lakes and lagoons of southern Australian are dynamic in space and time and this will influence opportunity for hybridisation. I surveyed genetic variation at microsatellite loci and the mtDNA control region of juvenile fish (collected after a single recruitment event after the known spawning time of A. butcheri) from five coastal lagoons that vary in terms of their morphology and frequency of opening to the ocean (including temporal replication in two lagoons) (total n = 970) to determine the frequency and persistence of hybridisation in the area of sympatry and to test the prediction that hybrid fish occur in greater numbers in lagoons generally closed to the ocean, where the two species may be trapped together by lagoon entrance channel closure. Of 688 juvenile fish genotyped 95% were either A. australis (347) or hybrids (309); only 5% (32) were A. butcheri. Most hybrids were later generation hybrids or A. butcheri backcrosses that are likely multi-generational residents within lagoons. Far greater proportions of hybrid juveniles were found within two lagoons that are typically closed to the ocean (> 90% hybrid fish within closed lagoons vs. 12 – 27% in permanently or intermittently open lagoons). In both lagoons, this was consistent across multiple cohorts of fish (79 – 97% hybrid fish [n = 282]).
Although hybrid adults and juveniles are abundant, and pure A. butcheri extremely rare, within coastal lagoons at the approximate southern range limit of A. australis, my initial broad scale survey of estuarine populations uncovered hybrids within a small sample of fish from the Gippsland Lakes, a large estuary some 250 km further south of my intensive sampling of juveniles. However, the dynamic nature of both lakes and lagoons and ocean currents imply that hybridisation between migratory marine and estuarine species may vary greatly in space and time. I took advantage of the opportunity provided by the relative great stability of DNA in preserved fish within museums to describe the genotypic composition of the Gippsland Lakes population of Acanthopagrus spp. and test the stability of the hybrid zone over a 60 year period. I used a microsatellite-DNA survey of contemporary (n=114) and historical (n=133) samples. The genetic composition of the samples varied little over time and comprised complex hybrid swarms within which most fish were more similar to A. butcheri. Genetic diversity as measured by mean number of alleles/locus and He ranged from 8.2 to 9.2 and 0.66 to 0.70 respectively, and I detected little temporal allelic differentiation (FST = 0.003).
My study highlights the underemphasized importance of estuaries as sites of hybridisation and my data imply that Acanthopagrus hybrid zones are more widespread and persistent than previously understood. The great mobility of A. australis coupled with likely range expansion predicted under global warming and the complexity of introgression within lakes and lagoons on the south east coast of Australia implies that hybridisation may pose a threat to the genetic integrity of remaining A. butcheri populations and has yet to be investigated for large numbers of obligately estuarine species.