A glucose molecule can assume various shapes. If presented as a linear rather than a ring structure, one of the two end carbons will contain an aldehyde grouping. The C=O structure of the aldehyde can react chemically with an amino group on another organic molecule to give a Schiff base. This is a horribly complex structure where the carbon of the sugar is double bonded to the nitrogen of the amino group, which stays attached by its third bond to whatever structure it's part of. Classically this is a "free" amino group of lysine or arginine. Needless to say this structure is unstable and falls apart in various ways. The sugar can remain attached or break off taking the amino group with it. Either way there is damage to the protein and the potential for cross linkages to form. Lysine and arginine are, as I mentioned, the most susceptible amino acids. Glucose is one of the least problematic sugars, fructose is one of the worst.
The end result of this sugar driven reaction is the generation of Advanced Glycation Endproducts (AGEs). They're probably bad.
There is a section of the apolipoprotein B100 molecule (That's that sole protein on the surface of the LDL cholesterol particle) which is very prone to AGE formation. These people have looked at the process in detail. They have located, within the apoB100 protein, a distinct continuous sequence of 67 amino acids which are exquisitely prone to AGE formation. This section of apoB100 is probably not part of the binding domain for the LDL receptor, but formation of AGEs here does strongly influence the binding domain and effectively stops it working. So AGE formation in this area inhibits LDL particle attachment to its receptor, so reduces clearance from the plasma.
This is highly reminiscent of the situation in familial hypercholesterolaemia.
Why is apoB100 designed this way? Stuff doesn't just happen "accidentally" like this. Evolution has selected the sequence of apoB100 protein to provide a 67 animo acid section in which AGE formation inhibits uptake of the LDL particle in to cells which might want it.
Let's look at the logic.
During glycation conditions the LDL cholesterol particle stops sticking to its normal receptor. Why? My answer is that under these conditions it is more advantageous for the body to have the LDL cholesterol particle in the blood stream than it is to have it endocytosed by an endothelial cell.
Glycation is related to sugar concentration. Glucose is the sugar least prone to glycate anything to an AGE. Couple that with the tendency for humans, over an evolutionary time scale, to eat diets that rarely budge the blood glucose outside of a relatively narrow range, and I'm not sure hyperglycaemia is what the "switch" on LDL is evolved to look at.
No, my guess is that fructose was the original lever to develop a glycation based switch on the apoB100 molecule. Just about the only time a human ought to get a big enough load of sugar to risk any damage to themselves is during late summer or autumn in temperate regions, the source would be fruit. Possibly large amounts during a short period. If this happens in autumn it's a good source of calories to convert to fat and not to be wasted. But fructose is ten or seventeen times as good as glucose at AGE formation, depending on which AGE you look at and which model you use.
Fructose appears to be bad news, especially if any gets in to the systemic circulation. I can see some logic in taking the LDL particle away from arterial endothelial cells, which can make their own cholesterol anyway, and having it handy in the circulation for other purposes, like patching up fructose induced damage. Bear in mind that, while fructose is only present in the systemic circulation in trace amounts (which are probably bad for you) it will be present in copious amounts in the portal vein from gut to liver after each fruit meal. Lipoproteins are present throughout the circulation. Fructose will meet LDL particles with head on impact in the portal vein. AGEs on apoB100 suggest AGEs elsewhere, which mean repair is going to be needed. The LDL particle is diverted away from the endocytosis receptor. This is probably physiological.
But raising blood glucose from below 6mmol/l to above 30mmol/l would allow glucose to become the primary glycating agent. For a diabetic on the ADA diet the glycosylation of apoB100 is probably a fact of life. This is not our normal autumn carb loading pre winter. It's more of a pathological process.
Finding high levels of LDL cholesterol is one of the more logical aspects of the hyperglycaemia of type 2 diabetes. Putting these patients on to LC diets usually drops their calculated LDL cholesterol levels along with their blood glucose levels. It probably markedly reduces AGE formation throughout their physiology. LDL can then get back to supplying normal lipid to normal cells through the LDL receptor.