Worm Breeder's Gazette 1(2): 27

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.

The Genetics of Dauer Larva Formation

D. Riddle

Figure 1

Figure 2

Figure 3

The work on the genetics of dauer larva formation is aimed at 
identifying all the genes involved, and establishing the specific 
function of each of these genes.  Two basic classes of mutants have 
been isolated: (1) mutants which enter the dauer larva state even in 
the presence of food (dauer-constitutive), and (2) mutants which fail 
to form dauer larvae when starved (dauer-defective).  As reported in 
the previous issue of WBG, about half of the eleven dauer-defectives 
initially isolated seem to be sensory mutants.  Therefore, it is 
expected that at least some mutants isolated as non-chemotactic, will 
also be found to be dauer-defective.  Even before this work was begun, 
Jim Lewis and Jonathan Hodgkin observed that starved cultures of their 
non-chemotactic mutant, E1033, failed to accumulate dauers.
Genetic mapping of the dauer-defective genes is in progress (fig. 1).
So far, none of the dauer-defectives (daf 3, 5, 6, 10) has been 
found to be closely linked to a dauer-constitutive mutation; The 20 
dauer-constitutive mutants characterized thus far fall into 6 
complementation groups.  Each of the 6 genes (daf 1, 2, 4, 7, 8, and 9)
has been mapped (fig. 1).  Fourteen of the 20 mutants fall into a 
single complementation group, daf 2.  While working at Cambridge, I 
obtained many dauer-constitutive mutants from Drs.  Sydney Brenner and 
Bob Edgar.  Bob Edgar isolated several mutants using SDS resistance (
Cassada and Russell, Devel.  Biol.  46, 326; 1975) as a positive 
selection.
[See Figure 1]
A surprisingly large fraction of dauer-constitutive mutants (12/20) 
are ts.  This may reflect a bias toward selection of leaky mutants 
which is inherent in the SDS selection.  However, 2 out of 5 dauer-
constitutives isolated by visual inspection of F1 clones are ts.  The 
remaining 3 are absolute lethals in the homozygous state because 
dauers are formed which cannot recover.  These latter mutants (2 
alleles of daf 2 and 1 allele of daf 9) must be maintained as 
heterozygotes.  The most extreme ts mutants are ts lethals.  At 25 C, 
100% of the population forms dauer larvae which are unable to recover 
until the incubation temperature is lowered to 15 C.  Growth at 15 C, 
on the other hand, is normal; no dauers are formed in the presence of 
food.
The ts lethal dauer-constitutive mutants are quite suitable for 
reversion experiments.  Selection for revertants (which grow at 25 C) 
provided the first indication of the relationship between constitutive 
and defective mutants.  Revertants of daf 2, daf 4 and daf 7 alleles 
were selected, and nearly all revertants were found to be dauer-
defective.  That is, the revertants not only escape dauer formation at 
25 C, but they fail to form dauers even when starved.  Genetic 
analysis of the revertants revealed that they are, in fact, double 
mutants which not only carry the parental constitutive mutation, but 
are also homozygous for an epistatic dauer-defective mutation.  Thus, 
some dauer-defectives are suppressors of dauer-constitutives.  
Suppression is generally recessive.  Many revertants carried alleles 
of dauer-defective genes which had been already isolated and mapped.  
The spectrum of revertants quickly approached saturation and many 
alleles of a few dauer-defective genes were obtained.  Thus it was 
clear that some, but not all, dauer-defective genes could suppress the 
constitutives.
In order to determine which dauer-constitutive mutants could be 
suppressed by various defectives, a series of multiple mutants was 
constructed by genetic crosses.  The data from these 'cross-
suppression' tests is summarized in fig. 2.  Suppressors fell into 4 
classes based on their pattern of suppression of 4 dauer-constitutives 
tested.
[See Figure 2]
These data have been considered in the following way.  The dauer-
defective mutants are blocked in the natural pathway of dauer 
formation.  The constitutives on the other hand, generate a false, 
internal signal which causes the mutant to form a dauer even in the 
absence of the natural signal (which accompanies starvation).  If the 
pathway is blocked after the false signal, the double-mutant will be 
dauer-defective.  However, if the false signal is generated after the 
block, the double-mutant is dauer-constitutive.  Thus, the data in 
figure 2 can be organized in the form of a genetic pathway for dauer 
larva formation (fig. 3).  The pathway in the figure is undoubtedly 
incomplete and may become more complex as more mutants are 
characterized.  As of now, the order of the constitutive genes, daf 8, 
7 and 4, in pathway I is unambiguous.  The daf 2 gene presents a 
problem since it seems to be on the partially distinct pathway II.  
Pathway II shares at least one step in common with pathway I.  However,
pathway II must not be functional in wild-type animals since any 
block in pathway I produces the dauer-defective phenotype.  In other 
words, pathway II must not respond to the natural signal for dauer 
larva formation, but only functions in daf 2 mutants.
[See Figure 3]
A block in pathway I is both necessary and sufficient to produce a 
dauer-defective phenotype.  In contrast to blocks in pathway I, the 
M26 mutation (which blocks only pathway II) is not dauer-defective.  
The M26 mutation was selected as a suppressor of daf 2 and that is its 
only obvious phenotype.
Over half (7/11) of the original collection of dauer-defectives are 
sensory mutants, at least some of which have defects in amphid 
structure.  Such sensory mutants are not found among revertants of 
dauer constitutives.  Furthermore, the sensory mutants which have been 
combined with dauer-constitutives, fail to suppress constitutive dauer 
formation at 25 C.  Thus, it is concluded that these genes function 
early in the pathway, prior to daf 8.  This is consistent with the 
hypothesis that the amphids mediate the primary sensory signal for 
dauer larva formation, while the dauer-constitutive mutations generate 
a false, internal signal.

Figure 1

Figure 2

Figure 3