Worm Breeder's Gazette 12(5): 92 (February 1, 1993)
These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.
We recently reported a genetic relationship between dauer formation and the timing of expression of an L1 -specificsurface antigen (WBG Vol. 12 #4, p. 63). Briefly, we observed that some dauer-constitutive (daf-c) mutants ( daf-1 , daf-4 , daf-7 , daf-8 , daf-11 ,and daf-14 ,but not daf-2 )bound L1 specific monoclonal antibody M37 at larval stages L1 -L4but not as adults, as do mutants altered in gene srf-6 II (5 alleles). We call this phenotype the Srf heterochronic phenotype. Unlike the daf-c mutants, the srf-6 mutants do not exhibit a ts dauer-constitutive phenotype.
The Srf phenotype of daf-c mutants is shown by larval stages from stocks grown at 16° C, a temperature at which the dauer-constitutive phenotype is maximally suppressed. This, together with the fact that srf-6 mutants are not dauer-constitutive, suggested to us that the Srf phenotype might be independent of dauer formation. However, some dauer-constitutive strains (e.g., daf-4 )exhibit high percentages of dauer larvae under noninducing conditions, including growth at 16° [Vowells and Thomas, Genetics. 130, 105-123 (1992) and our own observations]. It is possible that the Srf phenotype of daf-c mutants requires some aspect of the dauer program to be executed, even if the "overt" formation of dauers is not observed at 16°.
To investigate this issue, we tested double mutants containing the dauer-defective mutation daf-12 ( m20 )and various daf-c mutations for their ability to bind M37 . daf-12 was chosen as the dauer-defective for this purpose because it is strongly epistatic to dauer-constitutive mutations, i.e., these double mutants do not form dauer larvae under any conditions. daf-12 has been placed near the end of the dauer formation pathway on this basis [Riddle, Swanson, and Albert, Nature 290, 268-271 (1981)]. daf- 1( m40 ); daf-12 ( m20 ), daf-4 ( e1364 ); daf-12 ( m20 ), daf-7 ( e1372 ); daf-12 ( m20 ),and daf-8 ( m85 ); daf-12 ( m20 )(grown at 16°) all showed the Srf heterochronic phenotype, that is, daf-12 ( m20 )did not prevent the late larval expression of the L1 -specificantigen. daf-12 ( m20 )as a single mutant did not show the Srf heterochronic phenotype. These results show that the ability to form dauers is not required for the Srf heterochronic phenotype caused by these daf-c mutations. They also are consistent with the idea that daf-12 acts specifically on dauer formation.
The story does not end here, however. daf-3 and daf-5 are dauer-defective genes that have been placed at the same step in the dauer formation pathway as daf-12 ,but some differences in their behavior compared to daf-12 have been reported (Vowells and Thomas 1992). We therefore also tested double mutants containing daf-c mutations and dauer-defective mutations in daf-3 or daf-5 . daf-1 ( m40 ); daf-3 ( e1376 ), daf-4 ( e1364 );daf-3( e1376 ),and daf-4 ( e1364 ); daf-5 ( e1386 )(grown at 16°) only showed binding of M37 at the L1 stage, that is, the presence of a daf-3 or daf-5 mutation prevented heterochronic expression of the L1 -specificantigen. Because the double mutants did bind antibody at the L1 stage, this result is not readily explained as a defect in antigen expression per se, but appears to be an effect on the surface antigen switching that is presumed to occur at the L l1 molt.
We conclude that daf-12 wild-type activity is dispensible for the heterochronic expression of the L1 -specificantigen caused by the dauer-constitutive mutations we tested In contrast, both daf-3 ant daf-5 wildtype activity are required for the Srf heterochronic phenotype
Our interpretation of these results is that the the daf-c gene products signal to both dauer formation and surface antigen switching, but through downstream steps that eventually split into procoss-spocific branches. Incorporating our results into the pathway hypothesized for dauer formation based on epistasis analysis, one such branch point would be downstream of the daf-c genes (with the exception of daf-2 )and daf-d genes daf-3 and daf-5 ,but upstream of daf-12 (see Figure)
We would also like to place srf-6 in this model by epistasis analysis. If daf-3 and daf-5 are epistatic to srf-6 for the Srf heterochronic phenotype, it would suggest that the srf-6 product acts upstream of these steps. On the other hand, if srf-6 mutants still express the Srf heterochronic phenotype in the presence of these daf-d mutations, it would suggest that srf-6 acts in a surface antigen-specific way, perhaps downstream of the branchpoint proposed above. We plan to construct the appropriate double mutants to test these hypotheses
To summarize, every daf mutation we have tested so far except for daf-2 and daf-12 mutations also affects the surface antigen switching detected with monoclonal antibody M37 .Because it involves the continued expression of a marker normally expressed on the first larval stage, the Srf heterochronic phenotype is reminiscent of the propensity of mammalian cancer cells to express so-called carcinoembryonic or ÒoncodevelopmentalÓ surface antigens characteristic of embryonic or fetal cells. It remains to be seen whether the effects of the daf genes on surface antigen switching observed in this formal analysis signify a molecular involvement of the gene products in signalling to two distinct processes
We thank Don Riddle and Patrice Albert for encouragement and for gifts of double mutant strains, and Victor Ambros, Gary Ruvkun, and Jim Thomas for discussion