This is the intra-peritoneal insulin tolerance test (ITT) result from the mice in
Caloric Restriction Paradoxically Increases Adiposity in Mice With Genetically Reduced Insulin
as mentioned a post or two ago and which needs some sort of an explanation:
The two asterisks denote that for both of the calorie restricted groups of mice there is an elevated glucose compared to the ad-lib groups in the late part of the ITT, irrespective of whether the insulin gene dose had been reduced by 50% or 75%. Obviously the effect is biologically trivial but the p value of less than 0.05 makes me think the effect is real.
I think to understand this we have to go back some time and look at the concept that metformin has no effect on blood glucose in the absence of insulin. This is the graph from here, discussed here:
My interpretation was/is that, on a background of no metformin (upper line) that insulin (given at 90 mins) generated insulin induced insulin resistance from about sixty minutes later (time 150 mins) and this become p less than 0.05 by 90 minutes after the insulin (time 180 mins), illustrated by the failure to generate insulin-induced insulin resistance in the metformin treated mice (the lower trace).
This is insulin-induced insulin resistance in type 1 diabetic mice revealed by metformin treatment.
Next we can look at type 1 diabetic mice chronically treated with long acting exogenous insulin for a few weeks before an ITT. These have pre existing insulin-induced insulin resistance before any intra-peritoneal short acting insulin is given, taken from here, previously discussed here:
In this case the mice with established insulin resistance simply developed hyperglycaemia when injected with intra-peritoneal short acting insulin. There are no p values but by eyeball the 120 minute value on the upper curve looks like it might be stastically significantly elevated compared to time zero. The message here is that insulin given to an insulin resistance patient can produce hyperglycaemia.
This is the Somogyi Effect. It is real.
In the real world insulin secretion and insulin sensitivity are carefully balanced. Anything which increases insulin secretion increases tissue exposure to insulin and down regulates insulin action. Insulin is the messenger between insulin secretion at the pancreas and insulin response/resistance at the tissues. This is in addition to the shared use of reactive oxygen species to generate both insulin secretion and insulin responsiveness.
We know that the mice with full Ins2 knockout and with or without Ins1 partial knockout are phenotypically normal and have fairly normal insulin levels in their blood.
Aside: To actually get reduced insulin levels Johnson's lab have more recently used a full Ins1 knockout with or without partial Ins2 knockout. The partial Ins2 knockouts do have lowered insulin, are slim, don't get fatty liver and live much longer than they should do. It's all in here. Nice. End aside.
So reduced insulin genotype mice should be more insulin sensitive than full insulin complement mice, though we didn't have a fully normal group in either of the papers from the Johnson lab.
Without calorie restriction (CR) Ins1 partial knockout have their insulin system in balance. With caloric restriction they are so insulin sensitive that when they have an insulogenic calorie restricted small meal they lose calories in to their adipocytes and enter torpor, ie insulin signalling is verging on pathological. So they get fat too. They are not insulin resistant.
But during an ITT they do not merely have the modest insulin levels they might produce in response to their normal small meals. They get 0.75iu/kg of insulin IP, probably more than they have ever seen before. It works. Insulin signalling drops plasma glucose for about 30 minutes. At this point the tissues realise that they are seeing more insulin than they have ever seen in their lives. Insulin-induced insulin resistance kicks in and with it the Somogyi Effect to give elevated glucose.
I think the graph at the top of this page shows an acute onset insulin-induced insulin resistance.
This insulin resistance effect appears to be releasing glucose from the liver, or failing to oppose glucagon action here. It might also have allowed a release of free fatty acids from those greedy adipocytes which precipitated daily torpor. It is just possible that the transient insulin resistant state during the brief ITT might be the only few hours of the entire life of the CR mice that they were not hungry...