Friday, 7 December 2012

Another look at why LIRKO cant get fat

I have to string together a bit of a loose argument here, I may be wrong, but I want to put this post up anyway.

CarbSane smugly made this post today, although in the original paper the researchers report that the LIRKO mouse has "intact" insulin signalling in adipose tissue, im slightly suspicious of this.....

The MIRKO ( muscle insulin receptor knockout mouse ) is infact obese, here it is easy for me to speculate whats happening, a lack of glucose disposal in muscle is resulting in excess glucose disposal into adipose tissue and thus causing it to grow. A study from 1993  looked at over-expressing GLUT4 in adipocytes, which resulted in increased glucose uptake into adipose tissue.

The outcome was an obese mouse but the surprising finding was the main form of fat mass increase was adipocyte hyperplasia , and not adipocyte hypertrophy.  Scary thoughts...... Adipocyte hyperplasia ( fat tissue growth ) is caused by fat mass increasing its glucose uptake.

In 2004 Khan's research group ( who is responsible for developing the insulin-receptor knockout line of mice ) did another experiment with the LIRKO mouse where they administered the dreaded PPARγ agonist rosiglitazone to the LIRKO. 

Now here's where I get speculative.....

In the abstract, Khan reports that rosiglitazone caused an increase in adipocyte size in LIRKO mice, but a decrease in adipocyte size in normal mice.  The primary way that rosiglitazone improves glucose tolerance is by stimulating the production of new, small insulin-sensitive adipocytes, i.e. fat tissue growth. This is why in the normal mice it produced a decrease in adipocyte size,  there were born more new adipocytes, and the fat mass re-distributed over the higher number of fat cells.

But not in the LIRKO, no, rosiglitazone  just made them gain adipocyte volume, the last line of the abstract is where its at....

 but both metformin and TZDs require an operating insulin signalling system in the liver for their effects in glucose homeostasis.

Remember what we just said above, TZD's improve glucose homeostasis by making fat cells multiply, if TZD's failed to improve glucose homeostasis in the LIRKO mouse then this implies that TZD's failed to cause fat cell multiplication in LIRKO. The conclusion I am drawing is as quoted, an "intact" insulin signalling system in the liver is required for fat cell multiplication and thus the development of obesity.

So the hyperinsulinemia in LIRKO probably fails to result in obesity because for some reason the PPARγ adipogenesis pathway in adipocytes is not working.


  1. OK after Sidereal sent me the full text I see that what really happened was rosiglitazone caused apoptosis in fat cells in control mice but only increased fat cell size in LIRKO mice, although I was wrong in my initial speculation as to what was happening, what is clear however is that functional insulin signalling in the liver is indeed required in order for PPARγ in adipose tissue to work correctly.

    So again this points out that using the LIRKO mouse model with hyperinsulinemia to somehow "prove" that insulin doesnt cause obesity, is also a flawed proposal.

  2. The senior author (or someone taking the piss?) showed up on Carbsane's blog.

  3. Im inclined to think its someone fooling around but the comments were written with a sensible tone.