I hadn't thought about it too much. With regards to a polyamine related mechanism, I am picturing two types of cells under duress, those cells with an underexpression of polyamines (probably due to being short-changed when the cells split during mitosis and losing the active copy of important genes) and those cells that have overexpression of polyamine genes (perhaps gaining extra chromatin during mitosis or reactivating part of their inactive X chromosome). I know there are different paths of apoptosis but the first type of cell I think would go into a calcium/caspase apoptosis if the polyamines can't control calcium flow. But, I think there is a higher concentration of calcium in the mitochondria relative to the rest of the cytoplasm and nucleus, I might be wrong. But, if that is the case, damage to the mitochondrial membrane could release that calcium, perhaps from an overexpression of spermidine/spermine N1-acetyltransferase (SSAT) and polyamine oxidase (PAO). I believe that SSAT and PAO can create peroxides that can damage membranes but I am not familiar with that route. Anyway, calcium could come from two different directions, outside the cell and/or from the mitochondria. In the second type of distressed cell, if an increase in spermine in the cell starts disrupting chromatin in a cell that is still living, I believe some previously sequestered genes and pseudogenes could become active. I can envision ways in which autoantigens could be created from that, but it is not as simple as people might think, ie. not: previously sequestered gene is transcribed to a messenger RNA that is translated into a protein, a protein which the immune system attacks as a foreign protein. I don't think it's that straight forward.
Okay, now you've got me started. I should probably tackle how I believe there could be chromatin disruption that leads to further problems. This is probably the most complicated, wackiest part of the hypothesis. It is based a lot on lupus autoantigens but I think you will occasionally see carry over to MS and RA, especially when I get into reverse transcriptases.
I think I need to explain a little about the basic chromatin structure. Some of you may know this but please bear with me. I am trying to get some old pictures I made posted so I can refer to them. That would help. In the meantime...
DNA is normally double stranded and in the B-DNA form, which is a right-handed spiral (stick your right hand thumb parallel to the DNA chain and your fingers curl in the same direction as the strands curl around each other). The strands each have a backbone along the outside of the DNA chain made of alternating phosphate groups (phosphorus and oxygen) and ribose sugars (flat pentagon shaped carbon and oxygen structures). On the inside of the strands are the bases: guanosine (G), cytosine (C), adenosine (A), and thymidine (T). These are flat carbon, oxygen, nitrogen structures attached one base to one ribose. The two bases across from each other on the two strands are always paired as: G with C, or A with T. The pairs of bases are not fully connected to each other but are held by hydrogen bonds. They are attracted to the same hydrogen ion (a proton) but they do not share their electrons so it is weaker than the covalent bonds where electrons are shared. There are three hydrogen bonds between each G-C and two between each A-T, so that we can speak of stretches of DNA that are G-C rich as having greater attraction in the strands and so the strands can not be separated as easily. Due to the flat character of the bases, the base pairs can stack on top of each other. This stacking, the hydrogen bonding, and the phosphate backbones help maintain the integrity of the DNA chain. If a base is damaged, it can be replaced by matching the base in the opposite strand: G-C or A-T pairing. Strand separation is needed during transcription, repair, and replication when the DNA sequence (ex. GCATTTCCA....) is being read. During this though the phosphate backbones should maintain their integrity.
The two strands wrap around each other once every 10.5 base pairs on average. If they are overwound (ex. a turn in only 9.5 base pairs) that is said to be positive supercoiling stress. If they are underwound (ex. a turn in 12 base pairs) that is called negative supercoiling stress. The stress can be converted from overwinding or underwinding of the DNA spiral into what is actually supercoils, where the whole DNA chain warps into a loop. This is like twisting a stiff rope between your hands until it starts to bend with the stored up strain. You can see this conversion of supercoil stress if you take a rope and tape the ends together forming a circle. Then disconnect the rope ends and give it a bunch of twists and retape the ends. You will see the rope can remain strained as a circle or it can relax the strain by flipping into a figure 8 pattern. When this happens in DNA, it can reduce the linear storage requirements by 7x. This is part of how one meter of DNA can be stored in something as small as a cell. There is a lot of self-repulsion of the DNA in bending, however, because of the many negative charges along the phosphate backbones of the DNA. Histones are positively charged proteins that can form nucleosomes with DNA. Eight histones form a protein core and the DNA wraps 1 1/2 to 1 3/4 turns around this core in a left handed (negative supercoil) over the surface of the proteins. The histones neutralize about half of the DNA negative charges so the DNA is locally flexible enough to wrap around the proteins. About 145 base pairs are attached to the histones and another 55 (on average in humans) are in the linker region between nucleosomes. These nucleosomes appear as beads on a string when viewed under an electron microscope. The attachment of the DNA to the histones is not covalent, but is along the lines of hydrogen bonding between the negative phosphate groups of the DNA with the positively charged arginine and lysine residues (amino acids in a peptide/protein chain) in the histones. So these are not permanent connections. The histones' hold to DNA can be weaker if some of the lysine or arginine residues are methylated, which neutralizes the positive charge at that site, or acetylated, which converts the positive to a negative. So, if polyamines stimulate acetylation of histones, that could reduce the hold of histones to the DNA and open up the underlying gene. Also, the positively charged polyamines can bind to the DNA, competing with the histones or in other sites actually helping to hold the histone/DNA packaging. The nucleosomes can stack together, especially when histone H1 binds to the linker DNA further neutralizing the DNA's negative charges. When the nucleosomes are stacked, one could imagine polyamines binding to the DNA in two different but proximal nucleosomes acting as an additional piece of tape to hold that stacked structure. On the other hand, polyamines can stabilize Z-DNA, which is more rigid, less flexible than B-DNA. Z-DNA is a left handed coiling of the strands with 12 base pairs per turn. It can not bend over the surface of a histone so histones and Z-DNA are mutually exclusive. Z-DNA and nucleosomes both are forms of stored negative supercoiling so they compete for any negative supercoiling stress fluxing through the DNA locally. It takes about 5 Kcal/mol (ignore this, just an energy term) to initiate the flip from B-DNA to Z-DNA but then it only takes about 0.3 Kcal/mol for each additional base pair to flip. So once Z-DNA begins to form, it can grow and recede through a stretch of DNA, depending on the particular sequence (best is G-C rich, especially alternating methylated G-C base pairs, ex. GC,CG,GC). The moving section of DNA between Z-DNA and B-DNA is referred to as a B-Z junction and it can range from 3-12 base pairs. This is where there is unstacking and restacking of bases and is a site of disruption and vulnerability. I believe, but nobody has shown it, that this can be a site where enzymes might be able to access the bases more easily for such things as methylation or demethylation. Just one of my minor hypotheses. Anyway, polyamines (especially spermine) can stabilize Z-DNA and perhaps help encroach on a nucleosome's hold.
One of the areas of debate in research is what happens to the histones when a polymerase is reading through the gene. Somehow the DNA strands have to be unwound and separated, and at least several bases of one strand have to be released from the histone connections. This is where it starts to get complicated (ha). The polymerase is separating the strands (negative stress) as it reads the gene but in doing so it generates an equal amount of positive stress ahead of it in the DNA. This positive stress can be prevented from building up too strong by its encounter and neutralization by the negative stress stored in a nucleosome. This linearizes the DNA in the nucleosome, which weakens the overall nucleosome structure. Do the histones come off as one group, do they come off individually or two at a time, or does the polymerase somehow negotiate its way through the nucleosome one or two histones at a time without them coming completely off of the DNA? That is the problem being researched still. The nucleosome can reform behind the polymerase, however, because of the negative stress that is still available after the strand separation. This movement through a nucleosome is rapid, so the histone/DNA interactions are only temporarily disrupted. However, Alexander Rich at MIT has shown that, if polymerases are going through a gene very frequently, Z-DNA may actually form behind the polymerase. This can slow the following polymerases so that there is a modulating effect attenuating the pace of transcription for that gene. We are also finding that there are some proteins that bind Z-DNA, perhaps giving it longevity. So between these proteins, polyamines, histones, and a lot of other factors, there is a dynamic character to the stresses and accessibility of DNA at different sequences. The human genome project of sequencing the DNA was child's play compared to what it will take to determine an epigenetic map and a transcriptional activity map of the genome. This is too dynamic probably due to histone, DNA modifications, the local ionic milieu (calcium, magnesium, polyamines), the local supercoiling stresses, and other factors bound in the chromatin. Each protein that binds DNA can impart some negative or positive stress into the DNA and the local stress situation can have a lot to do with how tightly a protein holds on, and it can affect the integrity and continuity of protein complexes assembled on DNA. There are enzymes called topoisomerases that are continually active cutting a strand or both strands, twisting the DNA, and then religating the strands in order to adjust the supercoiling stress. Overall the genome is held in a slightly negative supercoiled state since each nucleosome holds a net of one negative supercoil and nucleosomes occur on average every 200 base pairs. Nucleosomes have some preferences as to the sequences that they will bind but it is not real specific. So we sometimes speak of nucleosomes as positioned, especially near the beginning of genes. With the modifications to histones (acetylation) and the competition for stress storage, sometimes a nucleosome can slide a little, instead of completely being removed. Even if it is a displacement of just half the 145 base pairs, sliding 70 base pairs may be enough to open up the initiation site of a previously sequestered gene or pseudogene. So think about how an increase in available free spermine and stimulation of histone acetylation could affect this. This will be important in the biggest, wackiest part of the hypothesis that leads to autoantigen generation.
Let's revisit the genetic versus epigenetic definitions. A example of a genetic change would be changing the DNA sequence at one site from a G-C base pair to an A-T base pair, or vice versa. There could also be insertions or deletions of base pairs at a particular site. These change the DNA sequence and permanently change the code so that it will mean (in many cases) a permanent change in the resulting protein. An epigenetic change could be acetylation of histones at a site that loosens the nucleosome's hold and allows transcription of the underlying gene. If the histones are deacetylated and the nucleosome grabs back on tightly, the gene can be shut off. It could be turned on again later. So no permanent change has occurred to the DNA sequence and the protein is still the same when translated.
If you can get to some pictures in a molecular biology, biochemistry, or general biology textbook, you might see some of this better. I will keep trying to post some of my old seminar pictures that are geared towards this type of discussion.
So histones help package DNA to reduce its storage requirements and its accessibility. When nucleosomes form, it is like beads on a string and those can be stacked together so that it looks like clumps of beads, further reducing the storage area required and further reducing the accessibility. This is how heterochromatin like much of the inactive X chromosome may appear. I haven't mentioned about the heteronuclear RNA that appears to play a role in chromatin structure and dynamics. That's because not a lot is known about how the RNA figures in. XIST RNA is an example, but it may be somewhat unique. It helps in X inactivation but exactly how is unclear. One thing may be that it helps recruit methyltransferase enzymes that methylate DNA and histones. Also, macroH2A has RNA binding sites in its structure. Perhaps macroH2A provides anchor points for XIST RNA and other controlling RNAs along inactive chromatin. RNA may wind up being an important part of chromatin in an epigenetic sense. And with that, polyamine/RNA interactions may be an important epigenetic control mechanism but there is little knowledge so far on polyamine/RNA/chromatin relations.
I'll explain in the next post how I think strand breaks could be involved in the chromatin problems I believe are occurring in autoimmune diseases.