In order to understand my theory, you will need to know more about glucose metabolism. Skeletal muscle cells and fat cells break down glucose in the presence of oxygen in their mitochondria, and in the process they produce ATP, the basic energy currency of all cells. A glucose transporter called GLUT4 is present in the cytoplasm of muscle cells, and it migrates to the cell membrane upon stimulation by insulin. GLUT4 essentially acts as a key that unlocks the door, letting glucose into the cell, but, like a key, it only works when it's inserted in the membrane. Both glucose and oxygen, unless they are carefully managed, can cause harm to the cell's proteins and fats. The glucose enters the cell within special cholesterol rich sites in the cell wall called lipid rafts [Inoue2006]. This is likely orchestrated to protect the cell wall from damage, because extra cholesterol allows the vulnerable lipoproteins in the cell wall to pack more tightly and reduce their risk of exposure. In muscle cells, myoglobin is able to store additional oxygen, bound to an iron molecule safely sequestered in an interior cavity within the myoglobin protein.
Sulfur is a very versatile molecule, because it can exist in several distinct oxidative states, ranging from +6 (in the sulfate radical) to -2 (in hydrogen sulfide). Glucose, as a powerful reducing agent, can cause significant glycation damage to exposed proteins, leading to the formation of Advanced Glycation End Products (AGE's) that are extremely destructive to health: they are believed to be a major contributor to heart disease risk [Brownlee1988]. So, I hypothesize that, if sulfur (+6) is made available to glucose as a decoy, the glucose will be diverted into reducing the sulfur rather than glycating some vulnerable protein such as myoglobin.
In searching the Web, I came across an article written in the 1930's about the striking ability of iron sulfate, in the presence of the oxidizing agent hydrogen peroxide, to break down starch into simple molecules, even in the absence of any enzymes to catalyze the reaction [Brown1936]. The article pointedly mentioned that iron works much better than other metals, and sulfate works much better than other anions. In the human body, starch is first converted to glucose in the digestive system. The muscle and fat cells only need to break down glucose. Thus, their task is easier, because the iron sulfate is now starting from an intermediate breakdown product of starch rather than from starch itself.
Where would the iron sulfate come from? It seems to me that the cholesterol sulfate, having hopped across the cell membrane, could transfer its sulfate radical to the myoglobin, whose iron molecule could provide the other half of the formula. In the process, the sulfur molecule's charge would be driven down from +6 to -2, releasing energy and absorbing the impact of the reducing effects of glucose, and therefore serving as a decoy to protect the proteins in the cell from glycation damage.
When the cell is exposed to insulin, its mitochondria are triggered to start pumping both hydrogen peroxide and hydrogen ions into the cytoplasm, essentially gearing up for the assault by glucose. If cholesterol sulfate enters the cell alongside the glucose, then all the players are available. I conjecture that cholesterol sulfate is the catalyst that seeds the lipid raft. Iron sulfate is then formed by bonding the iron in the heme unit in myoglobin to a sulfate ion provided by cholesterol sulfate. The cholesterol is left behind in the cell wall, thus enriching the newly forming lipid raft with cholesterol. The hydrogen peroxide, provided by the mitochondria upon insulin stimulation, catalyzes the dissolution of glucose by the iron sulfate. The pumped hydrogen can pair up with the reduced sulfur (S-2) to form hydrogen sulfide, a gas that can easily diffuse back across the membrane for a repeat cycle. The oxygen that is released from the sulfate radical is picked up by the myoglobin, sequestered inside the molecule for safe travel to the mitochondria. Glucose breakdown products and oxygen are then delivered to the mitochondria to complete the process that ends with water, carbon dioxide, and ATP -- all while keeping the cell's cytoplasmic proteins safe from glucose and oxygen exposure.
If I'm right about this role for cholesterol sulfate both in seeding the lipid raft and in providing the sulfate ion, then this process breaks down miserably when cholesterol sulfate is not available. First of all, the lipid raft is not formed. Without the lipid raft, the glucose can not enter the cell. Intense physical exercise can allow glucose to enter the muscle cells even in the absence of insulin [Ojuka2002]. However, this will lead to dangerous exposure of the cell's proteins to glycation (because there is no iron sulfate to degrade the glucose). Glycation interferes with the proteins' ability to perform their jobs, and leaves them more vulnerable to oxidation damage. One of the important affected proteins would be myoglobin: it would no longer be able to effectively carry oxygen to the mitochondria. Furthermore, oxidized myoglobin released into the blood stream by crippled muscle cells leads to painful and crippling rhabdomyolysis, and possible subsequent kidney failure. This explanation accounts for the observation that sulfur deficiency leads to muscle pain and inflammation.