Genetic variation inSymbiodinium(=Gymnodinium)microadriaticumFreudenthal, and specificity in its symbiosis with marine invertebrates. I. Isoenzyme and soluble protein patterns of axenic cultures ofSymbiodiniummicroadriaticum

Abstract

Axenically cultured Symbiodinium (= Gymnodinium) microadriaticum Freudenthal, isolated from 40 host individuals (or colonies) representing 17 species in two phyla were resolved into 12 strains, each strain having a unique combination of four isoenzyme patterns as revealed by starch gel electrophoresis. Each individual host appeared to contain an electrophoretically homogeneous population of algae. Each characteristic isoenzyme pattern was maintained after re-isolation of a cross-infecting strain (i.e. a strain that had been experimentally introduced into a host of a species other than that from which it was originally isolated), indicating a genetic basis for the protein patterns observed in each strain. The strains could be separated into three groups after constructing a dendrogram by the unweighted average linkage clustering method based on calculations of similarity coefficients. Soluble protein patterns obtained from the algae were also strainspecific. A dendrogram constructed from data obtained by analysis of protein patterns resolved by a polyacrylamide gel disc electrophoresis was similar to the dendrogram constructed from the data on isoenzyme patterns, but suggested that the `phenetic' distances between strains were larger than indicated by the isoenzyme patterns. 1. When axenic cultures of the `zooxanthella' Symbiodinium microadriaticum were analysed by starch gel electrophoresis for isoenzymes, the patterns obtained differed depending on the host from which the algae were originally isolated, but there was no 1:1 relation between algal strain and host species. 2. Patterns of soluble proteins on polyacrylamide gels corroborated the results obtained through isoenzyme studies. 3. Twelve `electrophoretic strains' of Symbiodinium microadriaticum were recognized. These strains separated into three groups after construction of dendrograms by the unweighted average linkage clustering method based on calculations of similarity coefficients, of `genetic identity' and of `genetic distance'. 4. Because so little is unambiguously known about the life history of Symbiodinium microadriaticum, and because our sample size was relatively small, it is not possible to discuss our data in the context of the population genetics of this symbiotic dinoflagellate. 5. The evidence presented indicates that the variation that we have observed in these algae is intrinsic, and is probably genetically based. Two possible mechanisms that may act as a selective force in promoting specificity between algal strains and their respective hosts are discussed.