Severity of NRTI-mediated mitochondrial toxicity differs in blood and fat cells

This article is more than 18 years old. Click here for more recent articles on this topic

Damage to the mitochondria caused by the nucleoside reverse transcriptase inhibitors (NRTIs) varies according to tissue type, Australian researchers report in the 1st August edition of The Journal of Acquired Immune Deficiency Syndromes. While both ddI (didanosine, Videx / VidexEC) and d4T (stavudine, Zerit) cause mitochondrial toxicity in fat tissue, only ddI is linked to the side-effect in white blood cells.

Mitochondria are cellular components that produce energy for the cell by breaking down food molecules. Damage to the mitochondria is caused by some NRTIs, leading to symptoms such as fat loss, damage to the peripheral nerves and elevations of lactic acid levels in the blood. Studies have reached divergent conclusions regarding the relative risk of mitochondrial toxicity with different NRTIs. This may be due to their analysis of different tissue samples.

To clarify the relationship between NRTI use and damage to the mitochondria in different tissues, investigators took blood samples and fat specimens from the thighs of 61 adult HIV-positive patients attending a clinic in Melbourne. They then compared the levels of mitochondrial DNA, a measure of how many mitochondria are in each cell, between patients taking different combinations of NRTIs.


deoxyribonucleic acid (DNA)

The material in the nucleus of a cell where genetic information is stored.


Structures in cells that are the sites of the cell’s energy production.

peripheral blood mononuclear cells (PBMCs)

Any blood cell having a round nucleus (e.g., a lymphocyte, a monocyte or a macrophage). These blood cells are a critical component in the immune system. 

mitochondrial toxicity

Mitochondria are structures in human cells responsible for energy production. When damaged by anti-HIV drugs, this can cause a wide range of side-effects, including possibly fat loss (lipoatrophy).



Almost all of the study participants were white men. On average, they had been taking their current antiretroviral therapy treatment regimen for 16.5 months, although 14 of the participants were not taking any anti-HIV drugs at the start of the study.

There was no association between mitochondrial DNA levels and the patients’ age, CD4 percentage, use of protease inhibitors or duration of NRTI exposure in either the fat tissue, or in the peripheral blood mononuclear cells (PBMCs) extracted from the patients’ blood samples.

However, mitochondrial DNA levels were significantly reduced in patients taking either ddI or d4T (p Retrovir) without ddI or d4T had similar levels of mitochondrial DNA to the patients taking no anti-HIV therapy.

When they examined the mitochondrial levels in the patients’ PBMC’s, the investigators found a similar reduction in mitochondrial DNA with the use of ddI (p = 0.003). However, the patients taking d4T without ddI had similar levels of mitochondrial DNA to the patients taking no anti-HIV therapy (p = 0.5).

“Current NRTI exposure is the major determinant of mitochondrial DNA levels in fat and PBMCs, but the precise associations are different in each tissue,” the investigators conclude. “Both ddI and d4T exposure are associated with fat mitochondrial DNA depletion, whereas ddI exposure was the only observed association with mitochondrial DNA depletion in PBMCs.”

The investigators speculate that mitochondria in PBMCs are less susceptible to d4T, as these cells may lack the enzyme thymidine kinase-1, which is required to convert d4T into its active form.

Importantly, however, this suggests that analysing mitochondrial damage in cells from blood samples may not produce a representative picture of the effects occurring in other tissues in the body. “Although blood is easily sampled, it is unlikely to be an appropriate tissue for understanding the events occurring in tissues where problematic NRTI toxicities occur, including fat and neurones,” they warn. They also point out that AZT can cause damage to muscle cells, which may not be reflected in analysis of blood cells.

However, their results could explain why improvements in fat levels under the skin after switching from d4T to abacavir (Ziagen) were not mirrored by improvements in mitochondrial DNA levels in blood cells in clinical studies.

Although they could not detect a link between mitochondrial toxicity and symptoms in their relatively small sample, the investigators suggest that NRTI use could cause damage to the mitochondria soon after a patient starts to take the drugs, since they saw no association between time on therapy and mitochondrial toxicity.

Conversely, they saw no differences in mitochondrial levels between patients who had never taken NRTIs, and those who had stopped taking them an average of 21 months earlier. “This suggests that mitochondrial DNA depletion reverses when NRTIs are ceased, whereas clinical toxicities, such as lipoatrophy and neuropathy, do not reverse rapidly,” they write.


Cherry CL et al. Tissue-specific associations between mitochondrial DNA levels and current treatment status in HIV-infected individuals. J Acquir Immune Defic Syndr 42: 435-440, 2006.