Effects of FFA and Ketone Levels on Basal Insulin Secretion

From Guenther Boden's group at Temple  

Acute Lowering of Plasma Fatty Acids Lowers Basal Insulin Secretion in Diabetic and Nondiabetic Subjects

In this study, these researchers used nicotinic acid (NA, aka niacin) to lower fasting/basal free fatty acid (FFA aka NEFA) levels and looked at basal/fasting insulin secretion rates (ISR's).   They also measured beta-hydroxybutyrate (they use β-OHB abbreviation in the manuscript), otherwise known as ketones.   The NA was administered orally every 30 min for the four hour duration of the test for subjects in the fasted state.   They essentially compared two groups of diabetics with non-diabetic controls.  Euglycemic clamps were used in one group of diabetics and controls -- blood glucose held constant at normal levels -- mean glucose 5.3 and 5.1 mmol/l  respectively.  They also tested a group of diabetics at elevated fasting glucose with an isoglycemic clamp -- again blood glucose held constant but this time at 10 mmol/l.  To rule out any direct effect of NA on insulin, a control was done with a lipid infusion to prevent FFA levels from changing and it was determined that NA did not alter insulin.  I've summarized the data in the table below:


From the discussion:

  • Lowering FFA levels from ~600 to ~100 µmol/l was associated with a decrease of 25–35% in ISRs and ~30% in peripheral insulin concentrations in normal subjects and diabetic patients. 
  • Ketone bodies may have contributed to this effect; in the normal subjects, their contribution, if any, was probably small.  β-OHB decreased by only ~60 µmol/l (from 110 ± 20 to 50 ± 10 µmol/l). 
  • Acetoacetate, [another ketone] ... may have added another 20 µmol/l, fora total decrease in ketone bodies of ~80 µmol/l compared with a fall in plasma FFAs of ~500 µmol/l.
  • In type 2 diabetic patients (during isoglycemic [elevated glucose] clamping), however, the fall in β-OHB was larger (from 252 ± 88 to 38 ± 8 µmol/l). Hence, it is likely that ketone bodies supported a larger part of basal ISRs in type 2 diabetic patients than in normal subjects.
Translation: Since the absolute amount of ketones was small for the non-diabetics and the euglycemic (normal glucose levels) diabetics, the change in ketones is not likely to be a significant factor in the resulting change in insulin secretion/levels. However the hyperglycemic diabetics had higher levels of ketones to start that were reduced to the same, insignificant, level by the nicotinic acid. Therefore the ketones may have a more significant effect in these subjects.  The researchers conclude:  
... under these study conditions, basal plasma FFA levels were responsible for 25–33% of basal ISRs—that is, they exerted long-term potentiating effects on glucose-stimulated ISRs. 
Addressing the relationship between obesity and basal hyperinsulinemia:
Lowering plasma FFA levels in our mildly obese nondiabetic and diabetic subjects normalized those subjects’ elevated basal insulin levels. ...  elevated plasma FFA levels may be responsible for at least some of the hyperinsulinemia in normoglycemic obese subjects. They may not always account for all of it, as insulin levels in some cases of normoglycemic obesity are higher than can be accounted for from the degree of FFA elevation.
One of the common mantras in LC circles is the misconception that the obese have too much of their fat "locked away" because of their hyperinsulinemia.  This study demonstrates that the failure of adipocytes to adequately lock away the fatty acids, leading to elevated FFA, that are, at least in part, responsible for the increase in basal insulin secretion and levels.

In summary:
... the results of this study demonstrated that basal plasma FFA levels supported 25–33% of postabsorptive insulin secretion in obese nondiabetic and diabetic subjects and, hence, may be physiologically important, long-acting potentiators of glucose-stimulated insulin secretion. Moreover, FFAs were probably responsible for some of the hyperinsulinemia in normoglycemic obese subjects.
I'm slightly confused about the part in italics above.  What they're saying is that FFA's are responsible for a certain portion of basal insulin secretion and therefore may act in the long term to enhance GSIS -- e.g. contribute to post-prandial hyperinsulinemia.  Firstly, I don't see where there's anything they looked at in this study that would lead to extrapolation to make such a statement.  Perhaps it is just the background knowledge of the researchers "speaking", but this seems contradictory to some of what I'm reading.  Elevated FFA's do enhance GSIS (indeed are required for it) in some contexts, but more elevated FFA's have a detrimental effect and suppress the GSIS response.  Still, this postulates a potential mechanism by which obesity, through the elevated FFA's can set a hyperinsulinemic snowball in motion down the hill.  I see it as something like this:

Diabetics are characterized by a diminished GSIS, however.  It could be that in these people, increased FFA magnify insulin secretion they don't cause proinsulin formation or may even suppress it.  By concurrently increasing basal insulin secretion, they further deplete insulin from the ß-cells leading, ultimately, to impaired GSIS.  Sounds plausible to me.


MM said…
Is it a well known fact that niacin lowers FFA? How does that work? Does it just shunt the FFA back into storage, or somehow encourage them to be burned? If it's the latter, then I would think high doses of niacin would be an effective weight loss supplement. Maybe it is and I'm just out of the loop.
CarbSane said…
I had never heard of it until I came across this paper. I did try to find any info on weight loss or gain with niacin and came up empty (albeit not an exhaustive search). I think it inhibits lipolysis which should lead to gains if there's anything to that whole theory.

I do wonder if this isn't somehow linked to the flushing thing. Sort of a "hypolipemic" attack if you will. The heart does run quite a bit on FA's and sudden lowering may take levels too low for a spell.
Unknown said…
Here's a quote from my pharmacotherapy txt on the MOA of niacin (nicotinic acid):

"...reduces the hepatic synthesis of VLDL, which in turn leads to the reduction in the synthesis of LDL. Niacin also increases HDL by decreasing its catabolism."

Regarding the flushing:

"...cutaneous flushing and itching appear to be prostaglandin mediated and can be reduced by taking aspirin....shortly before niacin ingestion."

From what I remember from school lectures, I thought that niacin would decrease fat loss because it lowers lipids by inhibiting HSL in the liver.
CarbSane said…
Thanks for the info Akim. Anecdotally I've not heard weight changes (either way) for those taking niacin.