No abstract

In this study, we describe remarkable intraspecific variation in sensitivity to the broadly distributed pollutants, polychlorinated biphenyls (PCBs), among wild populations of the nonmigratory estuarine Atlantic killifish (Fundulus heteroclitus). Variation among killifish populations was characterized in 28-day laboratory challenges using embryonic and larval life stages and the highly toxic, dioxin-like PCB congener, 3,3′4,4′,5-hexachlorobiphenyl (PCB126). In summarizing results for 24 populations, we show that killifish populations vary over four orders of magnitude in their sensitivity to PCB126 and that this variation is adaptive to the magnitude of contamination at their residence site. The four least-sensitive killifish populations reside in US Atlantic coast urban harbors >100 km apart from one another: New Bedford, MA, Bridgeport, CT, Newark, NJ, and Norfolk, VA, USA. Prior studies examining all but the CT population have shown that these killifish are relatively insensitive to local contaminants, with mixed evidence concerning the heritability of this trait. We show here that tolerance to PCB126 is extreme, with some mechanistic similarities among these four killifish populations. However, these populations do not respond identically to each other, and in at least one population, tolerance appears to degrade over the F1 and F2 generations tested. Complementary ongoing studies using molecular approaches provide opportunity to identify unique and shared mechanisms of tolerance in these independently evolving populations and explore the adaptive benefits and costs of contemporary evolutionary responses in the wild.

BackgroundWhen a large number of alleles are lost from a population, increases in individual homozygosity may reduce individual fitness through inbreeding depression. Modest losses of allelic diversity may also negatively impact long-term population viability by reducing the capacity of populations to adapt to altered environments. However, it is not clear how much genetic diversity within populations may be lost before populations are put at significant risk. Development of tools to evaluate this relationship would be a valuable contribution to conservation biology. To address these issues, we have created an experimental system that uses laboratory populations of an estuarine crustacean, Americamysis bahia with experimentally manipulated levels of genetic diversity. We created replicate cultures with five distinct levels of genetic diversity and monitored them for 16 weeks in both permissive (ambient seawater) and stressful conditions (diluted seawater). The relationship between molecular genetic diversity at presumptive neutral loci and population vulnerability was assessed by AFLP analysis.ResultsPopulations with very low genetic diversity demonstrated reduced fitness relative to high diversity populations even under permissive conditions. Population performance decreased in the stressful environment for all levels of genetic diversity relative to performance in the permissive environment. Twenty percent of the lowest diversity populations went extinct before the end of the study in permissive conditions, whereas 73% of the low diversity lines went extinct in the stressful environment. All high genetic diversity populations persisted for the duration of the study, although population sizes and reproduction were reduced under stressful environmental conditions. Levels of fitness varied more among replicate low diversity populations than among replicate populations with high genetic diversity. There was a significant correlation between AFLP diversity and population fitness overall; however, AFLP markers performed poorly at detecting modest but consequential losses of genetic diversity. High diversity lines in the stressful environment showed some evidence of relative improvement as the experiment progressed while the low diversity lines did not.ConclusionsThe combined effects of reduced average fitness and increased variability contributed to increased extinction rates for very low diversity populations. More modest losses of genetic diversity resulted in measurable decreases in population fitness; AFLP markers did not always detect these losses. However when AFLP markers indicated lost genetic diversity, these losses were associated with reduced population fitness.

Wild populations of the killifish Fundulus heteroclitus resident in heavily contaminated North American Atlantic coast estuaries have recently and independently evolved dramatic, heritable, and adaptive pollution tolerance. We compared physiological and transcriptome responses to embryonic polychlorinated biphenyl (PCB) exposures between one tolerant population and a nearby sensitive population to gain insight into genomic, physiological and biochemical mechanisms of evolved tolerance in killifish, which are currently unknown. The PCB exposure concentrations at which developmental toxicity emerged, the range of developmental abnormalities exhibited, and global as well as specific gene expression patterns were profoundly different between populations. In the sensitive population, PCB exposures produced dramatic, dose-dependent toxic effects, concurrent with the alterations in the expression of many genes. For example, PCB-mediated cardiovascular system failure was associated with the altered expression of cardiomyocyte genes, consistent with sarcomere mis-assembly. In contrast, genome-wide expression was comparatively refractory to PCB induction in the tolerant population. Tolerance was associated with the global blockade of the aryl hydrocarbon receptor (AHR) signalling pathway, the key mediator of PCB toxicity, in contrast to the strong dose-dependent up-regulation of AHR pathway elements observed in the sensitive population. Altered regulation of signalling pathways that cross-talk with AHR was implicated as one candidate mechanism for the adaptive AHR signalling repression and the pollution tolerance that it affords. In addition to revealing mechanisms of PCB toxicity and tolerance, this study demonstrates the value of comparative transcriptomics to explore molecular mechanisms of stress response and evolved adaptive differences among wild populations.

Human alterations to the environment can exert strong evolutionary pressures, yet contemporary adaptation to human-mediated stressors is rarely documented in wildlife populations. A common-garden experimental design was coupled with comparative transcriptomics to discover evolved mechanisms enabling three populations of killifish resident in urban estuaries to survive normally lethal pollution exposure during development, and to test whether mechanisms are unique or common across populations. We show that killifish populations from these polluted sites have independently converged on a common adaptive mechanism, despite variation in contaminant profiles among sites. These populations are united by a similarly profound desensitization of aryl-hydrocarbon receptor-mediated transcriptional activation, which is associated with extreme tolerance to the lethal effects of toxic dioxin-like pollutants. The rapid, repeated, heritable and convergent nature of evolved tolerance suggests that ancestral killifish populations harboured genotypes that enabled adaptation to twentieth-century industrial pollutants.

A population of the nonmigratory estuarine fish species Fundulus heteroclitus (mummichog) indigenous to a polychlorinated biphenyl (PCB)-contaminated Superfund site (New Bedford Harbor, MA, USA) demonstrated an inherited tolerance to local, dioxinlike contaminants (DLCs). These findings suggest that DLCs have acted as selective agents, allowing the survival of only the most tolerant individuals, forming DLC-adapted populations. We hypothesized that DLC-tolerant mummichog populations would reside where local conditions are toxic to sensitive individuals, and that toxic environmental conditions could be predicted based on responses of sensitive early life stages to laboratory exposures of DLCs. As a measure of DLC tolerance, progeny of field-collected fish were tested in the laboratory with a dioxinlike PCB congener, 3,3',4,4',5-pentachlorobiphenyl (PCB 126). Mummichog populations were collected from sites with sediment PCB concentrations predicted to range from nontoxic to toxic. Consistent with predictions, tolerant populations were indigenous to sites with elevated sediment PCB concentrations. Also, as predicted, DLC-tolerant populations were resident to sites far less contaminated than the Superfund site. These results suggest that exposures to persistent, bioaccumulative, and toxic contaminants may produce evolutionary effects on a geographic scale larger than previously envisioned. This study presents an approach and describes a model system that may improve understanding of the scale of occurrence for these potentially irreversible ecological effects.

No abstract

BackgroundThe most toxic aromatic hydrocarbon pollutants are categorized as dioxin-like compounds (DLCs) to which extreme tolerance has evolved independently and contemporaneously in (at least) four populations of Atlantic killifish (Fundulus heteroclitus). Surprisingly, the magnitude and phenotype of DLC tolerance is similar among these killifish populations that have adapted to varied, but highly aromatic hydrocarbon-contaminated urban/industrialized estuaries of the US Atlantic coast. Multiple tolerant and neighboring sensitive killifish populations were compared with the expectation that genetic loci associated with DLC tolerance would be revealed.ResultsSince the aryl hydrocarbon receptor (AHR) pathway partly or fully mediates DLC toxicity in vertebrates, single nucleotide polymorphisms (SNPs) from 42 genes associated with the AHR pathway were identified to serve as targeted markers. Wild fish (N = 36/37) from four highly tolerant killifish populations and four nearby sensitive populations were genotyped using 59 SNP markers. Similar to other killifish population genetic analyses, strong genetic differentiation among populations was detected, consistent with isolation by distance models. When DLC-sensitive populations were pooled and compared to pooled DLC-tolerant populations, multi-locus analyses did not distinguish the two groups. However, pairwise comparisons of nearby tolerant and sensitive populations revealed high differentiation among sensitive and tolerant populations at these specific loci: AHR 1 and 2, cathepsin Z, the cytochrome P450s (CYP1A and 3A30), and the NADH dehydrogenase subunits. In addition, significant shifts in minor allele frequency were observed at AHR2 and CYP1A loci across most sensitive/tolerant pairs, but only AHR2 exhibited shifts in the same direction across all pairs.ConclusionsThe observed differences in allelic composition at the AHR2 and CYP1A SNP loci were identified as significant among paired sensitive/tolerant populations of Atlantic killifish with multiple statistical tests. The genetic patterns reported here lend support to the argument that AHR2 and CYP1A play a role in the adaptive response to extreme DLC contamination. Additional functional assays are required to isolate the exact mechanism of DLC tolerance.