Some pre-conceptions put forward by this paper are startling, even if you have zero insulin signalling in skeletal muscle AND adipose tissue ( i.e. no insulin receptors ) you can still be protected from diabetes provided the insulin signalling in brain, liver, and pancreas is working properly. Although your glucose tolerance is worsened, mainly because skeletal muscle is the primary site of glucose disposal in the human body ( 1 )
So to repeat that, aslong as you have functional insulin signalling in brain, liver, pancreas, you should be protected from diabetes, if you mess with insulin signalling in any of those organs, you get diabetes.
These findings are surprising as they ostensibly indicate that insulin action in tissues with insulin-dependent glucose uptake, such as muscle and fat, is dispensable for metabolic control.
This helps confirm one of my long-time held beliefs, that anaerobic exercise can really help with glucose tolerance ( because it helps with skeletal muscle insulin sensitivity ), however, it will not go anywhere towards helping diabetes if you have screwed insulin signalling somewhere in either brain/liver/pancreas.
There is also this paper from 1998 which came to the same conclusion, i.e. that disruption of insulin signalling in both skeletal muscle and adipose tissue is not sufficient to cause diabetes, although again, it does worsen glucose tolerance.
Lastly, it seems this research group have also caught onto the fact that fatty liver, i.e. NAFLD may just be an affect of insulin resistance, and not the CAUSE. This is the same conclusion that one is lead too after reading Lustigs paper on selective hepatic IR.