Proc. Natl. Acad. Sci. USA Vol. 81, pp. 1258-1259, February 1984 Population Biology

Mercury selection of allozymes in marine organisms: Prediction and verification in nature (allozyme polymorphism/natural selection/mercury pollution) EVIATAR NEVO, RACHEL BEN-SHLOMO, AND BATIA LAVIE Institute of Evolution, University of Haifa, Mount Carmel, Haifa, Israel Communicated by Hampton L. Carson, October 31, 1983

ABSTRACT The geographic distributions of mercury-tol- MATERIALS AND METHODS erant allozyme genotypes of the enzyme phosphoglucomutase We tested electrophoretically PGM variation in 341 individ- in the shrimp Palaemon elegans and the enzyme phosphoglu- uals from seven populations of the shrimp P. elegans and cose isomerase in the marine gastropod Monodonta turbinata PGI variation in 192 individuals from three populations of were compared in a mercury-polluted site versus several un- the marine gastropod M. turbinata. We compared in both polluted sites on the Israeli coast of the Mediterranean sea. We species allozyme frequencies in the nonpolluted versus the conclude that in both phosphoglucomutase and phosphoglu- polluted Akko site. The electrophoretic procedures and the cose isomerase, the level of the mercury-tolerant allozyme ge- allozymic patterns have been described (4). The mercury notypes was higher in the polluted as compared with the un- concentrations at Akko and in two sites north and south of polluted sites. These results suggest that mercury selection is Akko are given in Table 1. operating in nature on allozyme genotypes of these marine or- ganisms along patterns comparable with those found previous- RESULTS ly in laboratory experiments. We suggest that the enzymes studied here display an adaptive pattern in polluted environ- The data presented in Tables 1 and 2 indicate that the hetero- ments. Therefore, they may be used as potential indicators and zygote MS of PGM in P. elegans and the homozygote MM of monitors of marine pollution. PGI in M. turbinata, previously shown to be mercury toler- ant (3, 5) in laboratory experiments, displayed their highest The evolutionary significance of allozyme polymorphisms frequencies at the polluted Akko site. We analyzed the data was tested by gene-frequency analyses in barnacles under by focusing on the major comparison between frequencies of thermal (1) and chemical (2) pollution in the sea. Later, we tolerant allozyme genotypes at the Akko site as opposed to tested by controlled laboratory experiments the effects of all other nonpolluted sites. The two independent-signifi- heavy metal pollutants (Hg, Zn, Cd, Pb, Cu) on allozymic cance Bailey d tests (7) gave P = 0.089 for Palaemon and P frequencies of 15 phosphoglucomutase, PGM, genotypes in = 0.090 for Monodonta. Each test alone is approximating the shrimp Palaemon elegans (3, 4) and on 5 phosphoglucose significance. Pairwise comparisons between sites were insig- isomerase, PGI, genotypes in two closely related species of nificant. marine gastropods, Monodonta turbinata and Monodonta We conclude that allozyme frequency distributions found turbiformis (4, 5). All these studies, but most importantly the in nature were consistent with the expectations of laboratory controlled laboratory tests, indicated significant differential tests. We suggest that mercury selection operates in nature survivorship among allozyme genotypes. The differential vi- on allozyme genotypes of these marine organisms along pat- ability of allozyme genotypes is probably associated with the terns similar to those found in the laboratory. different degree of heavy metal inhibition, uniquely related to each specific pollutant (4). Thus, we conclude that these DISCUSSION results reflect an adaptive nature of the allozyme polymor- Mercurials are generally the most potent inhibitors among phisms tested and suggest that their differential sensitivity to the sulfhydryl reagents (8). Furthermore, PGM activation is the quality and quantity of specific pollutants can be used as highly increased by formation of the PGM-Mg complex. a potential genetic indicator and monitor of pollution (3-5). Several heavy metals inhibit PGM activity through competi- We missed, however, the crucial link between laboratory tion with Mg (9), as has also been reported for Be, in vitro and nature. Can the laboratory results be confirmed in the and in vivo (10). For PGM, activation and inhibition by met- sea? als are explained by the same competition model (9). Pre- We focus our testing on mercury pollution because along sumably, strong competitive inhibition of PGM by heavy the Israeli Mediterranean Coast, the Akko site is highly pol- metals replacing Mg in the enzyme complex is the major luted due to industrial discharge, whereas elsewhere along mechanism involved in pollution poisoning. No parallel bio- the coast, waters are relatively mercury free. Here we pre- chemical study on PGI activity is known to us. However, as sent evidence derived from nature, verifying our laboratory for PGM, the staining reaction of PGI needs the addition of predictions. The most mercury-tolerant allozyme genotypes, MgCl2. Furthermore, for both enzymes, no enzymatic activi- the MS heterozygote in Palaemon PGM and the MM homo- ty was detected on our gels when HgCl2 was added to the zygote in Monodonta PGI, were found at highest frequency staining mixture in similar concentration to that of MgCI2, near the mercury-polluted Akko site, as predicted from labo- with or without MgCI2. Only further biochemical research ratory experimentation. will be able to provide the direct mechanism(s) involved in the differential tolerance of allozyme genotypes to heavy The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" Abbreviations: PGM, phosphoglucomutase; PGI, phosphoglucose in accordance with 18 U.S.C. §1734 solely to indicate this fact. isomerase.

1258 Downloaded by guest on September 23, 2021 Population Biology: Nevo et aL Proc. Natl. Acad. Sci. USA 81 (1984) 1259

Table 1. Frequencies of PGM genotypes of the shrimp P. elegans at the polluted Akko site compared with unpolluted sites Genotype frequency by location Akhziv Akko Shikmona Dor Caesarea Tel Aviv Ashdod Individuals, no. 41 42 72 59 49 53 25 Allozyme genotype FF 0.024 0.097 0.017 0.041 0.056 FM 0.098 0.143 0.208 0.220 0.122 0.226 0.080 FS 0.122 0.119 0.083 0.220 0.102 0.151 0.200 MM 0.415 0.238 0.250 0.237 0.347 0.226 0.280 MS 0.292 0.357 0.236 0.254 0.265 0.189 0.240 SS 0.049 0.095 0.111 0.052 0.082 0.076 0.160 Other 0.024 0.024 0.015 - 0.047 0.076 0.040 Hg in water, ppb 0.04 0.10 0.03 Hg in sediments* 0.05 0.51 *Reported as ppm per dry weight. Values are from Roth and Hornung (6). Table 2. Frequencies of PGI genotypes in the marine gastropod on the manuscript. This research was supported by grants from the M. turbinata at the polluted Akko site and two unpolluted United States-Israel Binational Science Foundation, Jerusalem, Is- neighboring sites rael, and the MED Trust Fund of FAO. Genotype frequency at location 1. Nevo, E., Shimony, T. & Libni, M. (1977) Nature (London) Akhziv Akko Shikmona 276, 699-701. Individuals, no. 36 96 60 2. Nevo, E., Shimony, T. & Libni, M. (1978) Experientia 34, 1562-1564. Allozyme genotype 3. Nevo, E., Pearl, T., Beiles, A. & Wool, D. (1981) Experientia MM 0.50 0.57 0.48 37, 1152-1154. MS 0.42 0.34 0.45 4. Nevo, E., Lavie, B. & Ben-Shlomo, R. (1983) in Isozymes: SS 0.08 0.09 0.07 Current Topics in Biological and Medical Research, eds. Rat- tazi, M. C., Scandalios, J. & Whitt, G. S. (Liss, New York), pp. 69-92. metal pollution. The above considerations suggest that mer- 5. Lavie, B. & Nevo, E. (1982) Mar. Biol. 71, 17-22. cury selection operates directly on the PGM and PGI loci 6. Roth, I. & Hornung, H. (1977) Environ. Sci. Technol. 11, 265- rather than on linked genes. 269. Recent reviews of allozyme variation in nature (11-13) in- 7. Bailey, N. J. J. (1959) Statistical Methods in Biology (English dicate that the distribution of allozymic variation is nonran- University Press, London), pp. 38-39. dom and varies with spatiotemporal environmental varia- 8. Webb, J. L. (1966) Enzyme and Metabolic Inhibitors (Academ- tion. Our controlled laboratory experiments (3, 4) and the ic, New York), Vol. 2. current verification of the results in nature indicate differen- 9. Milstein, C. (1961) Biochem. J. 79, 591-596. tial tolerance to heavy metals or differential viability fitness 10. Hashimoto, T., Jayant, G., Del Rio, J. C. & Handler, P. (1967) of specific allozymes caused by a specific pollutant. J. Biol. Chem. 242, 1671-1679. We conclude that our results are inconsistent with the neu- 11. Nevo, E. (1983) in Evolution ofMolecules and Men, ed. Ben- tral theory of allozyme polymorphisms and appear to reflect dall, D. S. (Cambridge University Press, Cambridge, En- the adaptive nature of the allozyme polymorphisms studied. gland), pp. 287-321. Furthermore, allozyme polymorphisms may provide poten- 12. Nevo, E. (1983) in Adaptive and Taxonomic Significance of tial genetic indicators of pollution for the short- and the long- Protein Variation, eds. Oxford, G. & Rollinson, D. (Academic, term changes that populations undergo due to pollution. London), pp. 239-282. 13. Nevo, E., Beiles, A. & Ben-Shlomo, R. (1984) in Evolutionary We thank S. Samson, D. Kaplan, and Y. Fisher for field assist- Dynamics ofGenetic Diversity: Lecture Notes in Biomathema- ance and A. Beiles, E. Golenberg, and M. Feldman for commenting tics, ed. Mani, G. S. (Springer, Berlin), in press. Downloaded by guest on September 23, 2021