Apr 6, 1972 - Commission under contract to Union Carbide Corp. We are indebted to Professors W. W. Anderson, Th. Dobzhansky, V. Grant, and R. K. ...
Proc. Nat. Acad. Sci. USA
Vol. 69, No. 6, pp. 1475-1477, June 1972
Protein Polymorphism and Genic Heterozygosity in a Population of the Permanent Translocation Heterozygote, Oenothera biennis (gene loci/alleles/primrose/starch-gel electrophoresis/enzymes)
DONALD A. LEVIN, GARY P.
HOWLAND*, AND ERICH STEINERt
Department of Biology, Yale University, New Haven, Connecticut 06520; and tDepartment of Botany, University of Michigan, Ann Arbor, Mich. 48105
Communicated by Verne Grant, April 6, 1972
Genic allozyme polymorphism and heteroABSTRACT zygosity was studied in a large population of the evening primrose, Oenothera biennis, growing in North Haven, Connecticut. This species is a permanent structural heterozygote for the entire complement of 14 chromosomes, and is thus capable of accumulating and maintaining genic heterozygosity. A total of 11 protein species were examined, and these were judged to be controlled by 19 loci. Polymorphism occurs at 26% of the loci, with only two alleles being present at each locus. The proportion of polymorphic loci per population in organisms with normal meiotic systems is no less than in Oe. biennis, indicating that structural heterozygosity is not necessarily accompanied by a greater proportion of polymorphic loci. On the other hand, the average Oe. biennis was heterozygous at 26% of its loci, an amount considerably greater than in most other organisms. With few exceptions, all plants in the North Haven population had the same genotype. Since the species is a colonizer, it is likely that the population was founded by a single individual and retained this genotype by virtue of the genetic system.
An abundance of allelic variation at structural loci that encode for polypeptides has been disclosed in a wide range of invertebrates, vertebrates, and plants by the technique of gel electrophoresis. In animals, the proportion of electrophoretically polymorphic loci per population averages about 30%/, and the proportion of heterozygous loci per individual averages about 10% (1-9). These parameters have been considered in too few natural populations of plants for the averages to be representative; however, considerable allelic variability has been documented within both outbreeding and predominantly self-fertilizing species (10, 11). In the organisms studied to date, heterozygosity is destroyed and reassembled each generation as a result of the normal meiotic process. These organisms have no mechanism for passing on specific heterozygous associations from one generation to another. There are, however, numerous species that do have special mechanisms that can preserve heterozygosity. In some species, apomictic or parthenogenic reproduction bypasses the sexual process, assuring transmission of the maternal genotype and the continuity of the genotype. In other species, the genome has survived a history of chromosomal translocations and/or inversions, that have led to a permanent state of partial or complete structural heterozygosity. Such chromosomal differentiation, accompanied by a
balanced lethal system, would preclude the formation of structural homozygotes. If the species were predominantly self-fertilizing, a large fraction of the progeny would be identical to their parents, barring mutation. Both apomictic and structurally heterozygous species are especially interesting subjects for genetic analyses, since they are uniquely equipped to accumulate and retain allelic heterozygosity. The structurally heterozygous species also may be expected to accumulate recessive deleterious or nonfunctional genes since both chromosomes, or fractions thereof, are sheltered from recombination. The species best known for permanent structural heterozygosity is the evening primrose Oenothera biennis (12). This species carries two genomes (Renner complexes), which involve seven chromosomes each. The genomes differ totally in the arrangement of their chromosome segments, so that a ring of 14 chromosomes is formed at diakinesis. Alternate disjunction occurs at anaphase I, assuring the migration of an entire chromosome complex to the same pole. Even though Oe. biennis is normally self-pollinated, the complexes fail to appear in a homozygous state because of a system of sporophytic and gametophytic lethals. Thus, linkage relationships and epistatic interactions are maintained over a large portion of the genome. We describe here the pattern of polymorphism and level of heterozygosity within a large population of Oe. biennis that occupies an ecologically heterogeneous locality isolated from conspecific populations, and compare these data with those obtained from organisms with normal meiotic systems. MATERUILS AND METHODS The study population was located in a 5-acre (2.0 hectare) empty lot in North Haven (New Haven County), Conn. The population contained several hundred large, vigorous plants, at a density averaging five plants per square meter. The site was bulldozed and scraped clean of most perennial plants several years ago, and has since undergone continued disturbance. The terrain is heterogeneous, containing one large hill of subsoil from an excavation nearby, depressions, ridges, and gentle slopes. Oenothera biennis was located throughout the site, whereas other weeds were distributed in accordance with some specific habitat features. Seeds from this population were obtained in April, 1971, having overwintered in the fruit. Tens of seed were collected from each of 200 plants sampled throughout the site, and were com-
Present address: Division of Biological Sciences, Oak Ridge National Laboratory, Oak Ridge, Tenn. *
1475
1476
Proc. Nat. Acad. Sci. USA 69 (1972)
Botany: Levin et al.
TABLE 1. Allelic frequencies and heterozygosity at 19 gene loci in Oenothera biennis
Gene PGI-1 PGI-2 PGM
G-6-PD-i G-6-PD-2 6-PGD GOT
LAP-i LAP-2 LGGP-1 LGGP-2 LTP-1 LTP-2 LTP-3 VLP LPP AcP-I AcP-2 AcP-3
Sample size 455 455 262 215 122 455 493 457 275 214 214 368 368 368 308 168 52 271 246
Allelic frequency ( %) Slow Fast 99.6 0.4 99.7 0.3 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100
Mean heterozygosity (%) 0.4
0.3 0 0 .0
0 100 0 0 0 0 100 100 0 100 0 100 0
0
bined prior to planting. As the data will show, a family-byfamily analysis was not necessary. Seeds were germinated on moist vermiculite in a laboratory at ambient temperature, or in a plant growth chamber at 25° and 60%o relative humidity, under 12 hr/day illumination (30-40 foot-candles of cool-white fluorescent light). There was no difference noted in the quality of the samples under these two different conditions. Individual (2-week-old seedlings (1-2 cm) were homogenized in 25 or 50 ,uA of extraction buffer (0.1 M tris(hydroxymethyl)aminomethane, pH 7.0-1 mM Na2 EDTA-25 mM 2-mercaptoethanol-50 ,uM NADP+) and applied to a single sample slot in a vertical-type starch gel. 12% Starch gels were prepared in a LiOH discontinuous buffer (5), or in a Tris-EDTA-borate continuous buffer (13). The gels with Tris-EDTA-borate also included 50 ,M NADP+, as did the cathode electrode buffer solution. Electrophoresis was performed at 1-2° for 18 hr at a potential of 280 V (LiOH) or 150 V (Tris-EDTA-borate). Three or four enzymes were tested on separate slices of each gel. About 30 different enzymes were tested; data from only those enzymes that could be reliably scored are presented here. Staining techniques for phosphoglucomutase (PGM), glucose-6-phosphate dehydrogenase (G-6-PD), acid phosphatase (AcP), and leucine amino peptidase (LAP) are from Shaw and Prasad (14); for phosphoglucose isomerase (PGI) and 6-phosphogluconate dehydrogenase (6-PGD) from Selander et al. (5); for glutamate-oxalacetate transaminase (GOT) from DeLorenzo and Ruddle (15); and for valylleucine peptidase (VLP), leucyl-tyrosine peptidase (LTP), leucyl-glycyl-glycine pepidase (LGGP), and leucyl-proline peptidase (LPP) from Ruddle and Nichols (13). Several hundred seedlings were analyzed. The number indicated in Table 1 for the various loci represent readily scorable assays. In the section that follows, the abbreviations used to designate each enzyme are written in italics to represent the
genes coding for the enzymes. When more than one form of the enzyme exists, each controlled by a separate locus, a hyphenated numeral is added to the symbol of the enzyme. The isozyme with the least anodal migration is called one, the next two, etc. When a locus is monomorphic, the allele is referred to as fast. When two alleles are present, the one with the greatest migration velocity is referred to as fast and the other one as slow. Multiple bands appeared upon staining of several of the enzymes. This pattern may represent homozygosity or heterozygosity at one or more loci, or some combination thereof. The distinction between these alternatives was not possible from the Oe. biennis gels, because the variation pattern afforded little of the necessary information. As a recourse, we analyzed protein profiles in several bivalent-forming taxa that are close relatives of Oe. biennis. The National Seed Storage Laboratory (U.S.D.A. Crops Research Division, Ft. Collins, Colo.) provided the following taxa: Oenothera argillicola race argillicola Bonn (OeA-1), Oe. hookeri ssp. hewettii race Albuquerque A (OeH-1), Oe. hookeri ssp. hertissima race San Pedro (OeH-7), Oe. hookeri ssp. montereyensis race Mateo (OeH-8), Oe. hookeri ssp. venusta race Dalton (OeH-9). E.S. provided seed stocks of Oenothera hookeri ssp. venusta race Dalton (OeH-S1), Oe. hookeri ssp. franciscana (OeH-S2), and Oe. grandiflora (OeG-S3). From the protein segregation pattern observed in these materials, the following judgments were made: (a) single bands for G-6-PD-2, PGI-1 and 2, PGM, LAP-2, LGGP-1 and 2, LTP-, and LPP were controlled by single homozygous genes; (b) double or triple bands for GOT, LTP-1 and 2, VLP, and AcP-1 were controlled by single heterozygous genes; (c) multiple bands for G-6-PD-1, 6-PGD, LAP-1, and AcP-2 and 3 were controlled by single homozygous genes. RESULTS AND DISCUSSION The results of our survey of enzyme variation in the Oe. biennis North Haven population are summarized in Table 1. A total of 11 protein species was examined, and these were judged to be controlled by 19 loci. Polymorphism is detected at five (26%) of the loci; an additional two loci (PGI-1 and PGI-2) displayed a variant in less than 0.5% of the plants sampled. The mean number of alleles per locus is 1.37; no more than two alleles are detected at a single locus. The pro-
portion of polymorphic loci in the Oe. biennis population does not differ conspicuously from that described in other organisms, despite its permanent structural heterozygosity. The proportions of polymorphic loci per populations in several organisms with a normal genetic system are as follows: Limulus polyphemus, 0.25 (1), Mus musculus, 0.22-0.30 (3-4), Peromyscus polionotus, 0.23 (5), Acris crepitans, 0.14-0.23 (6), Homo sapiens, 0.30 (7), Avena barbata, 0.31 (10), Avena fatua, 0.54 (10), Drosophila persimilis, 0.25 (2), Drosophila pseudoobscura 0.42 (8), Drosophila willistoni 0.54 (9), and Drosophila paulistorum 0.55-0.67 (17). Thus, the presence of permanent structural heterozygosity does not of necessity mean that a species will have a greater proportion of polymorphic loci per population than one whose karyotype is completely or primarily homozygous. We may surmise that either the proportion of polymorphic loci, or the specific loci that are polymorphic, is controlled by selection. If selection were not involved, how can we explain a "normal" level of polymorphic loci associated with a genetic system that per-
Proc. Nat. Acad. Sci. USA 69
(1972)
mits the accumulation of these loci? Recall that the level of heterozygosity in Oe. biennis is not subject to stochastic processes and is not a function of population size. The genetically effective size (Ne) of a population approaches infinity, since the opportunity for the decay of variability is vanishingly small. Although Oe. biennis is not set apart from other species by the proportion of polymorphic loci per population, the proportion of heterozygous loci per individual is unusually high. 26% of the loci are heterozygous in Oe. biennis, as compared to 2-4% in Avena barbata (11), 6% in Peromyscus polionotus (5) and Limulus polyphemus (1), 7% in Homo sapiens (7), 8-1 1% in Mus musculus, and 10-21% in various species of Drosophila (2, 8, 9, 17). At the five polymorphic loci, the proportion of heterozygotes is 1.0, i.e., there is permanent genic heterozygosity (Table 1). We may infer that these loci reside in the interstitial chromosome segments, i.e., between the centromere and translocation break points, and are sheltered from recombination. The monomorphic loci also may reside in these segments, or on the distal regions of the chromosome arms where recombination is possible. Although the proportion of heterozygous loci per individual is high in Oe. biennis, the allelic array is meager. All heterozygotes have the same two alleles (Table 1). With a few exceptions at the PGI loci, all plants at the North Haven site have the same genotype (for 19 loci), in spite of the fact that they occupy conspicuously different microhabitats. The absence of genotypic variability is in marked contrast to the considerable store of variability in ecologically heterogenous populations of the predominantly self-fertilizing Avena barbata and A. fatua (10), and populations of the apomictic Taraxacum officinale (16). The ability of Oe. biennis to cope with the environmental heterogeneity at North Haven suggests that it has considerable developmental flexibility or a broad ecological amplitude. Perhaps this is related to the relatively high level of heterozygosity. A high level of flexibility would permit a population to exploit a range of environments with a minimum of genetic heterogeneity, and would indeed be an asset to a colonizing species such as Oe. biennis. Relevant to this argument is the inverse relationship between genetic variability and developmental flexibility that has been demonstrated in Avena (10, 18-20). Finally, we may consider why there is so little intergenotypic variation, and a maximum of two alleles per locus in the
Gene Heterozygosity in Oenothera biennis
1477
North Haven population. The most plausible explanation is that this population was founded by a single colonizer, and by virtue of the genetic system developed into a clone. Oenothera biennis is winter annual or biennial, is primarily self-pollinating and self-fertilizing, and produces thousands of small seeds, which are readily dispersed. Rapid population growth is facilitated by these features, as well as by the retention of the genotype of a successful invader. Moreover, the invasion of a site by Oe. biennis is contested by few other species. If more than one colonizer were involved, either they were all of the same type or all but one genotype was eliminated by selection. This work was supported in part by National Science Foundation Grant GB 17987 to D.A.L., and by the U.S. Atomic Energy Commission under contract to Union Carbide Corp. We are indebted to Professors W. W. Anderson, Th. Dobzhansky, V. Grant, and R. K. Selander for their critical reading of the manuscript. 1. Selander, R. K., Yang, H. Y., Lewontin, R. C. & Johnson, W. E. (1970) Evolution 24, 402-414. 2. Prakash, S. (1969) Proc. Nat. Acad. Sci. USA 62, 778-784. 3. Selander, R. K., Hunt, W. G. & Yang, S. Y. (1969) Evolution 23, 379-390. 4. Selander, R. K. & Yang, S. Y. (1969) Tex. Univ. Publ. 6918 271-338. 5. Selander, R. K., Smith, H. H., Yang, S. Y., & Johnson, W. E., (1971) Tex. Univ. Publ. 7103, 49-90. 6. Dessauer, H. C. & Nevo, E. (1969) Biochem. Genet. 3, 171188. 7. Harris, H. (1969) Brit. Med. Bull. 25, 5-13. 8. Prakash, S., Lewontin, R. C. & Hubby, J. L. (1969) Genetics 61, 841-858. 9. Ayala, F. J., Powell, J. R., Tracey, M. L., Mourdo, C. A. & P6rez-Salas, S. (1972) Genetics 70, 113-139. 10. Marshall, D. R. & Allard, R. W. (1970) Heredity 25, 373-382. 11. Allard, R. W. & Kahler, A. L. (1971) Stadler Symp. 3, 9-24. 12. Cleland, R. E. (1958) Planta 51, 378-389. 13. Ruddle, F. H. & Nichols, E. (1971) In Vitro 7, 120-132. 14. Shaw, C. R. & Prasad, R. (1970) Biochem. Genet. 4, 297-320. 15. DeLorenzo, R. J. & Ruddle, F. H. (1970) Biochem. Genet. 4, 259-273. 16. Solbrig, 0. (1971) Amer. Sci. 59, 686-696. 17. Richmond, R. C. (1972) Genetics 70, 87-112. 18. Jain, S. K. & Marshall, D. R. (1967) Amer. Natur. 101, 1933. 19. Jain, S. K. & Marshall, D. R. (1970) Theor. Appl. Genet. 40, 73-75. 20. Marshall, D. R. & Jain, S. K. (1968) Amer. Natur. 102, 457-467.
