Protease inhibitors may partially contribute to the development of body fat changes by blocking the uptake of glucose, due to inhibition of the glucose transporter protein glut4, according a report due to be published soon in the Journal of Biological Chemistry (Murata). Protease inhibitors are also responsible for a loss of insulin sensitivity in people without the body fat changes, the authors believe, following test tube experiments.
However, some experts in the field are sceptical about the latest theory.
Glut 4 is responsible for transporting glucose into skeletal and cardiac muscle and fat. Mice that lack glut4 are almost fat-free, suggesting that this protein plays an essential role in laying down fat stores. If peripheral fat is largely synthesised from glucose, as the authors speculate, this would explain why people on HAART often lose fat from their limbs and faces. Changes in insulin sensitivity usually precede body fat changes, and have been noted in one study after an average of just 3.4 months on therapy (Mulligan).
A weakness of this study is that only indinavir, ritonavir and amprenavir were used in the test; saquinavir and nelfinavir were not used. Indinavir reduced glut4 activity by 45%, amprenavir by 54% and ritonavir by 42%. The study did not look at effects of nucleoside analogues on glut4 activity either, yet insulin resistance was reported in 27% of people taking nucleoside analogues in one German cohort (Goebel).
"It would be surprising if nucleoside analogues caused insulin resistance given that they are structurally different from protease inhibitors" said Dr Graeme Moyle of the Chelsea and Westminster Hospital in London.
"Insulin resistance is also a feature of some non-HIV lipodystrophy syndromes", Dr Andrew Carr of St Vincent's Hospital, Sydney, told aidsmap, "and insulin is important in promoting adipocyte function and growth, but it's certainly not the only factor". Dr Carr's research group in Sydney has reported peripheral fat wasting in people who have taken nucleoside analogues but no protease inhibitors, yet they found no significant difference in levels of insulin resistance between people with or without lipodystrophy, regardless of their treatment history (Carr).
There are also question marks over the relationship between insulin resistance and elevated lipid levels, which tend to accompany body fat changes.
"We saw improvements in glucose tolerance in studies where people switched from PIs to efavirenz or nevirapine" Dr Graeme Moyle points out, "but this doesn't necessarily translate into improved fat handling". Similarly, German researchers have reported that insulin resistance improves when a PI is replaced by abacavir in the majority of cases.
What do the findings mean in practice?
The authors suggest that current methods for testing glucose tolerance in people on PIs may not detect many emerging cases of insulin resistance, because they are conducted whilst fasting. However, protease inhibitors must be taken with food (except in the case of indinavir as sole PI), and the researchers observed very rapid effects of PIs on glucose uptake. Total inhibition was achieved within six minutes of adding indinavir to a test tube containing cells expressing Glut 4. This suggests that insulin resistance may be at its least pronounced during the fasting period when drug levels will also be lowest.
However, Dr Graeme Moyle points out that drugs generally take several hours to be fully absorbed, and a further period to reach the tissues where protease inhibitors might be exerting a negative effect on glucose uptake. Thus, the current practice of measuring fasting glucose and then two hours after a 75g dose of glucose is still likely to capture this trough.
Do the results call into question the increasingly common practice of combining protease inhibitors in order to achieve substantially higher blood levels of one PI?
"If this does translate into clinical practice, it reinforces what we already know, which is that all drugs have a therapeutic window and that higher levels often produce a greater desired effect but also more side effects" says Andrew Carr. However, Dr Graeme Moyle is not so sure.
"There may be a window of inhibition of glut4 beyond which higher dosing makes no difference" he suggests, pointing out that we know very little as yet about the effects of PIs on insulin sensitivity and should not jump to conclusions.
For now, the impact of protease inhibitors on insulin sensitivity will join the ranks of mechanisms proposed to cause body fat changes on HAART. Further data is likely to be presented at a special session on insulin resistance during the Second International Workshop on Adverse Drug Reactions and Lipodystrophy in HIV, which takes place in Toronto in September.
References
Carr A et al. A syndrome of lipoatrophy, lactic acidaemia and liver dysfunction associated with nucleoside analogue therapy: contribution to protease inhibitor-related lipodystrophy syndrome. AIDS 14 (3): F25-F32, 2000.
Goebel FD et al. ART-associated insulin resistance: frequency, potential causes and possible therapeutic interventions. First International Workshop on Adverse Drug Reactions and Lipodystrophy, San Diego, 1999.
Murata H et al. The mechanism of insulin resistance caused by HIV protease inhibitor therapy. Journal of Biological Chemistry 2000 (in press, available now online at http://www.jbc.org/cgi/reprint/C000228200v1.pdf