Worm Breeder's Gazette 11(4): 40

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.

Molecular Characterization of mec-4 Dominant Alleles that Induce Neuronal Degeneration

Monica Driscoll, Maud Kinnell, Mary Sabolsice and Marty Chalfie

Fifty-six recessive mutations in the mec-4 gene disrupt the function 
of the six mechanosensory neurons of C.  elegans.  Three rare dominant 
alleles of the mec-4 gene (e1611, u214, and u231) induce the cell-
autonomous degeneration of the touch cells.  Degenerating touch cells 
have a swollen, vacuolar appearance, suggesting that osmotic balance 
or membrane integrity is disrupted by mec-4(d) gene products.
We have cloned the mec-4 gene by standard transposon-tagging 
techniques (WBG 11(3)-44).  DNA sequence analysis has demonstrated 
that mec-4 is highly homologous to deg-1.  (A rare dominant allele of 
deg-1 induces vacuolar degeneration of a group of neurons other than 
the touch receptor neurons (Chalfie and Wolinsky, Nature 345:410)).  
We have isolated cDNAs from at least two other C.  elegans genes that 
are members of the mec-4 ly and are 
currently sequencing these clones.
We have identified a 6 kb HindIII restriction fragment from cosmid 
T20B9 that harbors a functional mec-4 gene, since it can restore touch 
sensitivity to mec-4 recessive mutants in DNA microinjection 
experiments.  We also note that high copy number of the wild-type mec-
4 gene does not induce neurodegeneration in N2 animals.  This 
observation supports the idea that dominant mec-4 alleles encode a 
novel product that induces neurodegeneration rather than causing 
overexpression of a normal mec-4 protein.
In order to identify the mutations that induce neuronal degeneration,
we have cloned and sequenced the three dominant mec-4 alleles.  We 
find changes in the same amino acid, Ala 438, in all three 
independently isolated alleles.  In e1611 and u214, Thr is substituted 
for Ala and in u231 Val is substituted for Ala.  To confirm that a 
single amino acid substitution at position 438 is sufficient to cause 
degeneration, we substituted Val for Ala 438 in the wild.type 6 kb 
HindIII mec-4 clone through site-directed mutagenesis.  This in vitro-
synthesized mutation acted in a dominant fashion to induce 
degeneration of the touch cells in microinjection experiments.
It is striking that simply the addition of a methyl group on the  -
carbon of the amino acid at position 438 can make the mec-4 protein a 
toxic product.  We hypothesized that steric hinderance at position 438 
might be a factor in inducing the degenerative state.  We have 
constructed site-specific mutations at position 438 in which we 
substituted smaller or similar sized amino acids (Gly, Ser) or larger 
amino acids (Asp, Arg, Leu, Phe, Pro).  These constructs were tested 
in vivo for their ability to induce degeneration of the touch cells.  
We find that Gly or Ser substitutions do not induce degeneration.  
Rather, they are silent mutations since these constructs can rescue 
the touch insensitive phenotype of recessive mec-4 mutations.  Thus, 
the amino acid at position 438 need not be Ala for mec-4 function.  
All larger substitutions that we constructed act dominantly to induce 
degeneration in wild-type animals.  Thus, it appears that steric 
hinderance around Ala 438 plays a crucial role in the degenerative 
process.
What is the role of Ala 438 in the mec-4 protein? To speculate on 
this issue, we considered homology to deg-1, hydrophobicity analysis, 
and predictions of secondary structure.  The region in which Ala 438 
resides is very highly conserved between mec-4 and deg-1.  In a 43 
amino acid stretch including Ala 438, 33 of 43 amino acids are 
identical and only two substitutions are non-conservative.  Thus, this 
region is likely to be functionally important.  Hydrophobicity plots 
of the deduced mec-4 protein sequence show a hydrophobic domain from 
amino acids 433-468 that is somewhat interrupted by amino acids 439 to 
446.  Secondary structure predictions suggest that this region is 
likely to be involved in a turn in the protein; the regions more N-
terminal and more C-terminal to this turn are predicted to readily 
form alpha-helices.  We speculate that the region from amino acids 447 
to 468 (or so) forms a membrane spanning domain in the mec-4 protein.  
N-terminal to this transmembrane domain is a turn in the protein.  
This turn may be embedded in the membrane or may be located just 
outside the membrane.  Ala 438 is at the junction of this turn and 
another alpha-helical stretch.  Since mec-4 mutations seem to disrupt 
osmotic balance in the touch cells, we suggest that the mec-4 protein 
may be directly or indirectly be involved in transport.  In order for 
this transport function to be turned off, Ala 438 must come in close 
contact with another residue.  If this proximity is prevented by 
steric hinderance, neurons uptake excess solute, swell and eventually 
lyse.  The importance of other amino acids in this region can be 
tested by further in vitro mutagenesis.