key assumption in our work is that the myosin motor passes through a number of defined states or shapes, and that mutations and suppressors act by 'locking' the motor into or out of specific states. If this is true, then we would expect that most or all of our mutants should fall into a defined number of classes based on their behavior. For example, if the motor has 5 states, we might naively predict 10 kinds of mutants--ones that make each state 'too happy' or stable, and ones that make each state too difficult to achieve or unstable. While this is doubtless an oversimplification, it holds the essence of our basic assumptions.
f this is true, amongst the hundred or more suppressors we expect to observe that many of them share similar properties or similar relationships with the mutants they suppress. We should reach a point where there is "...no new thing under the sun", to quote Ecclesiastes. To my great excitement, it looks very much like this is the case, and mutants are indeed falling into a set of 'families', where each family shares the ability to suppress or be suppressed by a common set of mutants. Similarly, we would predict that the cold-sensitive mutants we have isolated, while being two dozen or so in number, should contain individuals that share family memberships. Below is a summary of several emerging families: cs mutants: Not as different as you might think
Pretty cool that out of two dozen mutations selected just on the property
of being sickly, 11 should end up acting as 'healers' for each other, eh?
But wait! There's more!
Parting the Red Sea: Two primary kinds of mutants?
here's the scoop: Arrows depict SUPPRESSION, i.e. the mutation at the base of the arrow can restore function to a myosin molecule defective because of the mutation at the head of the arrow. Two-headed arrows indicate cases in which mutants fix each other: either alone results in a defective myosin, both the presence of both yields a functional motor. The positions in the middle are a key aspect of the argument. For these amino acid positions, ONE KIND of mutation (shown in parentheses on the left, i.e. (F)L176) restores function to mutants in the LEFT column, while a different mutation (shown in parenthese on the right) restores function to mutants in the right column. Mutants in outline font are from the original cs mutant hunt; underlined mutants are known to disrupt motor function to an extent that can be manipulated genetically (i.e. confer significant plaque expansion rate defects). Pretty neat, huh? While this is terribly intriguing (particularly those 'men in the middle'), for the case to be compelling (and to reveal subtleties in the groupings) we need to know about the 'missing arrows'--at present, any arrows not shown represent LACK OF INFORMATION, not a failure of suppression to occur. This will be my summer '99 project: to fill in much of the rest of this chart! In the meantime, my money sits as follows: I believe that the mutants on the LEFT represent biases favoring strong actin-binding states of the motor, whereas those on the right are weak actin-binding. This is a sweeping generalization and ignores many important subleties, but it's a place to start. One fly in the ointment: Erik Misner found G691C (top, right column) as a suppressor of P536R (bottom, right column). This will not do! I anticipate that this is the first of many complications; the data above suggest only 2 states of the motor, whereas I would anticipate something more like 5; instead of either/or I look for the final data to show circular relationships; nonetheless, the crude assessment allowed by current data fit reasonably neatly into the two-state diagram above.
Love triangles: when two mutants share suppressors
second way in which we can uncover relationships between mutants is by observing that two mutants can be suppressed by the SAME suppressor. Unlike above, where cs mutant-cs mutant suppression implies that the mutants are somehow 'opposite' of one another, two mutants suppressed by the same suppressor are implied to be in some way SIMILAR to each other. At this point in the work, we have several emerging 'familes' of mutants. Birds of a Feather at the End of the CamShaft
hree cs mutants not mentioned above show a very interesting spatial relationship. Y494K, W501L and G691C, despite their different locations in the linear protein sequence, form a tight grouping in the Dicty and chicken crystal structures. In the image below, residues in question are shown as magenta, with a stripe color corresponding to the part of the protein chain they arise from. Thus Y494 is magenta with red stripe, W501 magenta with orange stripe, and G691 is magenta with blue stripe. 
Perhaps more exciting than their clustering is fact that the three also share suppressors--mutants isolated as a suppressor to one turn up as suppressors of another! We have extended these random findings by testing suppressors of one against the others, and now have the following relationships (N483S was not isolated in the cs hunt, but rather as a suppressor of G680V, and is in the image as green with red stripe):
Note: empty cells indicate untested pairings
The actin binding zone: more nepotism
rik Misner recently completed a master's degree in my laboratory. His extensive genetic analysis of suppressors of mutants E531Q, P536R and R562H can be summarized in concept as follows: the three mutants above share at least 6 suppressors in common! This site will be updated with that information when the data is submitted for publication; as an intriguing tidbit, however, most of the suppressors cluster around the C-terminus of the 'forehead helix' (amino acids 411-441, magenta in updated images shown on these pages). What do we intend to do about it?
here are two ways in which we are proceeding with this data. First, we are trying to complete the sets: to determine how many commons suppressors the 483-494-501-691 family members and the 531-536-562 family members have. Matt Scholz is now working on the first set, extending Leonardo Mendoza's initial work, and Meghan Kreeger on the second. They are using a combination of error-prone DNA replication (to generate large numbers of suppressor candidates) and selection for suppressors in Dictyostelium (see the Page of Mutant Hunting) to identify suppressors.
he second level of effort is the most exciting part of this work to date: trying to figure out why these suppressors all have the same 'power' to fix the mutants, and why the mutants behave similarly (at least in genetic terms--i.e. each can be suppressed by the same suppressors). We reason that mutants 'unbalance' the smooth cycling of the motor, getting stuck or skipping key steps. Suppressors act to 're-balance' the motor cycle, perhaps by introducing a compensatory imbalance! Overall, then, we anticipate that family members will be having the same detrimental effects on the same structures. So what ARE the structures? Can genetic data lead us to them? This is new ground, so we're making things up as we go. Briefly, we are trying to consider the following:
|
||||
