The surprisingly mild phenotype of
rab-3 (
y251)homozygotes defective in the gene encoding a synaptic vesicle associated ras family GTP binding protein caused us to determine if additional genes encoding
rab3 -likeproducts are present in the C. elegans genome. The fact that multiple related
rab3 family members are found in bovine and other mammals suggested that this was a real possibility. It remains unclear if the mammalian
rab3 genes are functionally redundant. The rab family consists of a large number of small 200-250 amino acid GTP binding proteins hypothesized to regulate cellular membrane transport. A number of these proteins have been demonstrated to associate with small vesicles that shuttle between cellular compartments. For example,
rab1 and
rab2 are thought to be involved in ER to Golgi transport,
rab6 in intra-Golgi transport,
rab4 and
rab5 in plasmalemma-endosomal transport, and
rab3 in synaptic vesicle transport. Over a dozen additional rab family members have been identified, most of which have not been cellularly localized. We designed three PCR experiments capable of amplifying both
rab-3 and more divergent rab family members reasoning that if we were capable of isolating clones encoding less conserved members of the family we should also retrieve clones encoding novel conserved
rab3 family members. We performed the PCR reactions from reverse transcribed poly A+ mixed-staged RNA and analyzed products by sequencing large numbers of independent product clones. In the first experiment we designed oligos corresponding to a sequence found in all rab family members [#1 QIWDTAGQE] and a sequence found specifically in
rab3 [#2 NFDYMFK]. We isolated 98
rab-3 clones and 8 clones encoding
rab8 .In the second experiment we widened our search, using oligo #1 and an oligo corresponding to a sequence found in multiple rab family members including
rab3 [#3 GNKCDME]. From this PCR reaction we isolated clones encoding the following proteins: 10
rab1 A,2
rab1 B,2O
rab3 ,40
rab8 ,13
rab1 O,5
rab11 ,5
let-60 ras, and 7 rabQ (for question mark). In the final experiment, we used oligos to region #1 & the SL1 trans-spliced leader to isolate rab family members as non-specifically as possible from similar cDNA stocks. We isolated clones encoding
rab1 A,
rab1 B,
rab2 ,
rab3 ,
rab6 ,
rab6 B,
rab7 ,
rab8 ,
rab10 ,
rab11 ,
rab14 ,
rab19 A,
rab19 B,RAM, and two other less conserved rab family members using this strategy (we have not saturated this screen). In summary, in these experiments we isolated and sequenced cDNAs encoding from 60 -120 a.a. of 16 different GTP binding proteins including a dozen divergent members of the rab family, yet only isolated the one known
rab-3 gene. We analyzed these sequences, EST cDNA sequencing project data (
rab1 ,
rab3 ,
rab5 ,and
rab10 ),and genome sequencing project data (
rab6 )using the GCG program PILEUP to determine how these new gene products fit into the family tree of some of the known GTP binding proteins (Figure 1). It is clear that C. elegans expresses a large variety of GTP binding proteins homologous to the rab family members involved in membrane transport. If, as proposed, rab members mark different types of intracellular vesicles, even lowly C. elegans is prepared for serious vesicular traffic jams. The tree also confirms that none of the genes we have isolated encode products that are closely related to
rab3 .Previous experiments that isolated
rab-3 also failed to identify other genes encoding
rab3 -likeproteins in C. elegans. These included amplification of genomic DNA and cDNA using oligos corresponding to other sequences conserved among
rab3 family members and Southern blotting experiments that eliminated the possibility of a tandem duplication of
rab3 .Despite some caveats, we feel these experiments suggest that no other gene encoding a
rab3 member is present in the C. elegans genome.