So the question becomes, if my lab is interested in pH, iron, and host responses, how in the blue hell did we end up with a paper on histone dosage? Its a good question and one I thin sheds some light on how science is actually done.
Some background first (be forewarned, this is going to be painful). In our studies on the way C. albicans responds to extracellular pH, we identified a gene called MDS3 (the name is historical drivel for all intents and purposes). In our follow up work on MDS3, we found some interesting phenotypes that got me thinking that maybe the Mds3 protein has a role in epigenetics. (Epigenetics is an inherited phenotype that is not due to a change in the DNA sequence ie a genetic change.) There are numerous examples of epigenetics, but the one most in play in the press is post-translational modification of histone proteins. Histones form a complex that the DNA is wound around this helps compact the DNA and, more importantly, gives it some organization. (Like a skein of yarn compared to that same yarn after the cat has been at it.) Since DNA is negatively charged, due to all the phosphates in the backbone, the histone complex is positively charged. This allows for a tight interaction because like the + and - ends of a magnet they stick together. Histones are positively charged because some of the histone proteins contain numerous lysine amino acids, which has a positive charge. Now cells need to access the DNA to make mRNA, and thus protein, which is difficult to do if the DNA is wrapped tightly around the histones. So cells modify the lysines by adding acetyl groups, which changes the positive charge to a negative charge. Going back to our magnet metaphor, what happens when you put a + end near another + end? They want nothing to do with each other. The same thing happens to the histone-DNA interaction, it gets a lot weaker. Now whether the histones in a particular area of the genome are acetylated or not can be inherited and this leads to a type of epigenetics. This is because the genes in an acetylated area are more likely to be expressed than those in a deacetylated area. These changes in gene expression between cells can confer inherited phenotypic differences without a change in the DNA sequence. In summary, we had an idea that MDS3, which is involved in pH responses, played a role in epigenetics and one type of epigenetics is through histone acetylation state. Ok, go get yourself another drink and try and recover.
Now researchers in Saccharomyces cerevisiae found that you can trick a cell into thinking histones are acetylated or not simply by changing the amino acid sequence. Normally, lysine is negatively charged but can be modified to be positively charged (acetylated). However, we can easily mutate that lysine to an glutamic acid, which is also negatively charged, but cannot be acetylated. This effectively locks the histones in a negatively charged state (for that amino acid). We can also mutate that lysine to an arginine, which is positively charged. This effectively locks the histones in a positively charged state (again for that amino acid). Finally we can mutate that lysine to alanine, which has no charge. This effectively removes that amino acid from any form of regulation. For those Im not losing in all the biology, you may be wondering if these mutations just kill the cell. They don't, generally, because we are dealing with a single lysine amino acid and there are numerous ones within the histone so the histone should function normally. So we thought that we could make similar lysine mutations in C. albicans and then use these mutants to learn something about MDS3.
This may sound like a ton of information (and it is), but this really came from a basic undergraduate and graduate school education in molecular biology as well as reading some papers in the literature. So despite the density of the above two paragraphs, it was pretty simply to see how we get from MDS3 to epigenetics!
Now comes the practical difficulties (all the above was intellectual not practical). A histone complex is made up of 4 distinct proteins Histone H2A, Histone H2B, Histone H3, and Histone H4 (Yes, Im ignoring Histone H1 for those in the know). Of these, Histone H3 and to a lesser degree Histone H4 have been studied most for their role in epigenetics. So we decided to focus on the C. albicans genes that encode Histone H3 and H4. Problem #1: many organisms including C. albicans have multiple copies of the genes encoding Histone H3 and H4. C. albicans has 3 genes for Histone H3 and 2 genes for Histone H4. This makes genetic analyses more difficult because C. albicans is a diploid (just like you and me). That means that there are two alleles of each Histone H4 gene or 4 copies total. So if we mutant a single lysine to glutamic acid of one Histone H4 gene, there are still 3 copies of the normal Histone H4 being expressed. Needless to say, that could confound our analyses.
So what can we do? One idea is to say screw it and focus on other things. We have frequently taken that course, but this project was TOO FUCKING INTERESTING to throw in the towel. So, we decided to simply delete the other alleles encoding Histone H4 and then mutate the remaining one. Problem #2: C. albicans is NOT S. cerevisiae. So all we have to do, is delete 3 copies of Histone H4, no biggie we have deleted 8 copies of genes in other contexts before.
So we delete 1 copy, 3 left!!
We delete 2 copies, 2 left!!
We delete 3 copies, 2 left!!......?!?!FUCK!!!!
This was the problem, we couldn't delete 3 copies, at least not initially. It turned out that C. albicans does not, I mean, does NOT deal well with having a single Histone H4 gene.
The short term outcome is that there is no way we can use this genetic approach to do the experiments we wanted to do. Bill Paxton had one viewpoint, but my group's more Leeroy Jenkins!!!
From the hell of a failed approach, we found that if we looked carefully we could find cells that were deleted for 3 copies, 1 copy left!! They were just sick as shit and had some cool characteristics. So as opposed to trash the whole damn thing, we invested a fraction more time and turned out a little pub that you will hopefully take a look at. Regardless, the paper is better than chicken.
So while we couldn't get there from here, we actually did get somewhere albeit not Kansas, and we still learned some really cool shit about MDS3!
BTW, if you do happen to look over our, or anyone's, PLoS ONE paper. Login (it's painless) and leave a comment. Help get a discourse on the research you're already reading about going.