The hump-killer and mate-killer traits of Paramecium aurelia are known to be due to the presence of symbionts-kappa 51 and mu, respectively-in the cytoplasm of the killer paramecia. The presence of each of these symbionts also determines specific resistance to the killing action determined by that symbiont. Persistence and multiplication of mu and of kappa 51 occur only in paramecia possessing gene M or K, respectively. After these genes are replaced by their recessive alleles (m or k), the symbionts are actively destroyed; but destruction is delayed and reproduction continues for seven cell generations before the symbionts are suddenly destroyed after varying numbers of further cell generations. The delayed but sudden destruction and the kinetics of origin of mu-less cells in the course of cell generations 8 to 15 led Gibson and Beale (1962) to postulate the existence of a stable but non-replicative intermediate, the metagon, between the dominant gene (M) and its effect on mu, one metagon being held to suffice for the maintenance of a full complement (more than 100) of mu. The metagon was assumed to be formed in cells of genotype M-, not in the cells of genotype mm. In the latter, the initial metagons (about 103) inherited from an Mm parent were passively distributed to progeny, mu being promptly destroyed in any cell that failed to receive at least one metagon from its parent cell at fission. Much support was obtained for this hypothesis from a variety of experiments, of which the major kinds were recounted. The results also led to characterizing the metagon as a stable, infective, non-replicative RNA. It was suspected of being the mRNA of gene M because it was formed only in cells possessing this gene and was reformed in them after RNase had destroyed or inactivated the metagons present at the time of enzyme treatment. After I had found that kappa 51 persists and multiplies in certain stocks of Didinium and Dileptus, ciliates that prey upon Paramecium, the question arose as to whether metagons occurred also in these predators and protected the kappa from destruction. The existence in K paramecia of a comparable metagon for kappa 51 was suggested because the manner of loss of this kappa in kk paramecia was long ago reported by Chao (1953) and the details were strikingly like those later found for mu by Gibson and Beale. Because full evidence for this metagon was lacking, Gibson and Sonneborn (1964) explored, in Didinium, the situation as to mu, specifically mu 540, the one on which metagon studies had been made. They found that Didinium maintained mu 540 after its introduction (via feeding) and that its maintenance depended on the presence of the metagon. They reported, however, that Didinium possesses the metagon only after acquiring it from paramecia; otherwise it is absent. Yet, once acquired, it persists in the progeny for as long as followed. That this persistence was not due to the presence of an M-like gene in didinia was indicated by the failure of the metagon to "hybridize" with DNA from didinia although it does hybridize with DNA from paramecia, especially from those bearing gene M. Apparently, therefore, maintenance of mu 540 in didinia for hundreds of cell generations is due to its replication, although in M- paramecia it arises as a complement of gene M. In agreement, Gibson (unpubl.) has evidence that actinomycin D inhibits its production in paramecia, not in didinia; while the reverse is true for the action of guanidine hydrochloride. Although the evidence for the existence of a metagon and for its replication in Didinium is much more extensive and direct in the case of mu 540 than in any other, some evidences suggest the existence of metagons in the cases of at least two other killer symbionts, kappa 51 and mu 138. These two symbionts also replicate in certain strains of Didinium and/or Dileptus. Yet the situation is not pat, simple, or free from perplexity, even in the case of the metagon for mu 540. First, one of the test systems is not regularly reliable: the kinetics of loss of the symbionts from homozygous recessive (mm or kk) paramecia varies for unknown reasons and cannot yet be controlled. Second, the question of the source of the RNA-dependent RNA polymerase, which one would suppose to operate in metagon replication, is not yet firmly answered. Third, the replication of the metagon in mm paramecia at a low rate is in accord with certain observations and is at least not yet excluded; in any case, the control of the occurrence and rate of its replication is at present totally obscure. Fourth, no clue has yet been noted as to the explanation of the puzzling two-generation lag in the production of the first metagon, both when gene M is introduced into a cell and after M- cells have been exposed to RNase. Much remains to be learned and understood before an entirely clear and firm picture of the metagon emerges. Aside from the few possible cases mentioned in P. aurelia, no other examples of metagons appear to have been reported. Yet a few possible parallels in bacteria and higher animals suggest the desirability of further examination from this point of view. In regard to cytoplasmic inheritance, the metagon of Paramecium appears to be a relatively simple, macromolecular unit, unlike the DNA-containing organellar units mainly dealt with in this symposium, and to be an effective unit of cytoplasmic inheritance only in certain foreign organisms where it seems to present the character of an RNA gene or plasmagene. If, as now appears, the metagon originates as an RNA transcript of a DNA gene and partakes of the nature of a defective virus in foreign cells, this further suggests an important step in the origin of RNA viruses. It obviously becomes important now to explore whether, to what extent, and under what conditions, the RNA transcripts of other genes can acquire metagon-like properties of stability and replication.
No takes yet. Share an insight, caveat, or question.
T. M. Sonneborn (1965) studied this question.