Genetics

A major thrust in our laboratory is the use of forward genetics to dissect microbial channel genes. 

Early on, we used Paramecium to show that elements underlying an action potential can be dissected by genetics.  Pawn mutations block its initiation, paranoiacs block its termination, etc. (1).  The development of molecular genetics was slow for Paramecium, because of the research community is small.  Nonetheless, our laboratory has worked out a method and cloned the genes for pawnA (2), pawnB (3), pnt (4), xntA (5).  This method has continued to be used by Japanese and French investigators to clone other Paramecium genes from mutations.  The arrival of the complete genome sequence and gene silencing has now brought Paramecium genetics to a new stage dealing with new questions (6).

Two generations of researchers advanced the genetics of S. cerevisiae and E. coli.  However, genetic dissections of physiology of these microbes revealed no ion-channel involvement.  It therefore came as a surprise when our patch-clamp survey revealed channel activities on their membranes in the late 80’s (7,8).  Channel genes, revealed in microbial genomes in the 90’s, contributed greatly through crystallography.  However, except for MscL and MscS ,what channels do for these microbes remain a mystery.

Starting from a mutant phenotype, forward genetics finds and clones channel genes (e.g. Shaker), making possible site-directed mutageneses of homologs (reverse genetics).  We have taken forward genetics to the next level – to find the crucial parts of the channel.  Specifically, we pioneered the use of “gain-of-function” mutations to find these parts.  We randomly mutagenize a channel gene and search for mutations that stop growth because they make “loose-cannon” channels (Fig. 1).  We then correlate the amino-acid changes with the electrophysiolgical changes, without preconceived hypothesis or bias.

Fig. 1  An example of the use of forward genetics to dissect the structure/function relationships of a channel (10).  A. The yeast K+ channel gene (TOK1, formerly YKC1) in a plasmid is randomly mutagenized before transforming into tok1Δ yeast.  Transformants are plated in recplicas comparing TOK1-repressed vs. induced state.  Colonies in the former but missing in the latter are picked.  The sites of mutations are then determined by sequencing.  The electrophysiological phenotypes of the mutant channels, examined in Xenopus oocyte, showed failure in proper closure of the channels, confirming their being “loose-cannons”. B. Examples of selected mutants, which grow in a repressing (glucose) but not in an inducing (galactose) medium.  In this experiment, the “loose-cannon” mutations pinpointed a “post-pore” region at the end of the helices that trailed the selective filters,.  The “post-pore” region coincide with the gates later visualized in the crystal structures of KcsA and MthK.  (See Fig. 3 under “Yeast K+ channel”)

“Loose-cannon” mutations pin-pointed the hydrophobic amino acids that line the very constriction of the mechanosensitive MscL in E. coli, later confirmed by crystallography (9).  Such mutations defined the “post-pore” region of Tok1, the yeast K⁺ channel (10) that anticipated the peptide below the gating hinge in the crystallized KcsA and MthK.  This mutations add weight in the regulatory role of the RCK domain of Kch, the K⁺ channel of E. coli (11).  They are currently being used to study TRPY1, the yeast vacuolar TRP channel.

1. Kung et al. (1975) Science 188: 898.
2. Haynes et al. (1998) Genetics 149: 947.
3. Haynes et al. (2000) Genetics 155: 1105.
4. Kink et al .(1990) Cell 62: 165.
5. Haynes et al. (2002) PNAS 99: 15717.
6. Haynes et al. (2003) Eukaryot. Cell  2: 737.
7. Martinac et al. (1987 ) PNAS 84: 2297.
8. Gustin et al. (1986) Science 233: 1195.
9. Ou et al. (1998) PNAS 95:11471.
10. Loukin et al. (1997) EMBO J. 16: 4817.
11. Kou et al. (2003) EMBO J. 22: 4149.

.