Drug cocktails have transformed HIV infection from a deadly to a chronic disease, giving HIV carriers a new lease on life. But new long-term consequences of harboring the virus and of taking powerful drugs indefinitely keep cropping up, and the prospect of an Alzheimer’s-like dementia is the latest one. At the 34th meeting of the Society for Neuroscience here in San Diego, Cristian Achim of the University of Pittsburgh Medical School, and Lynn Pulliam of UCSF, yesterday and today presented early data suggesting that middle-aged people with HIV who are on highly active antiretroviral therapy (HAART) develop Aβ deposition, both inside neurons and as extracellular plaques. The two labs are working independently of each other.

It’s long been known that HIV infection will affect the brain at some point, but this is different, the scientists report. The brain can become a reservoir for the virus even while it is undetectable in blood and the HAART-treated patient shows no outward signs of disease. Accompanying HIV-associated risk factors for dementia include damage to the blood-brain barrier, infiltration of monocytes/macrophages (which most likely carry in the virus), and high levels of inflammatory cytokines. On top of that, a more specific risk for Alzheimer’s pathology may arise from two related processes, the scientists proposed. For one, these patients tend to develop high levels of insulin and insulin resistance, and this may have to do with interactions between the HIV drugs and the Aβ-degrading enzyme IDE, Achim’s work suggests. For another, an HIV protein called stat inhibits neprilysin, another enzyme that keeps Aβ levels down, reported Pulliam.

Achim described early data of an ongoing study in which he assessed amyloid deposition in postmortem tissue of 162 cases aged 25 to 70, with an average of 40, from the University of California Los Angeles and San Diego. These people had not had HIV encephalitis. Sixty percent of them had intraneuronal Aβ deposition confirmed by immunoelectron microscopy, and half of those also had extracellular deposits, mostly diffuse plaques, but no tangles. Achim hopes to follow a group of patients on long-term HAART forward with the PET amyloid imaging agent PIB to test how amyloid deposition correlates with cognitive performance over time.

Pulliam took a different tack by studying the effect of tat, an HIV protein secreted from infected macrophages and microglia, on neprilysin. Tat inhibited neprilysin in membrane preparations from human brain cultures and led to an increase in Aβ when added to cultures directly. Then she looked for Aβ in autopsy brain sections and, like Achim, found that middle-aged people with HIV had more plaques, large diffuse ones, than did controls. Plaque number correlated with how many years the person had been infected, but not with age. Whether this AD pathology leads to clinical Alzheimer disease is unclear, as is the question of how ApoE genotype influences the outcome. Even so, independent scientists at the conference agreed that long-term HAART-treated HIV survivors might constitute a new natural model of a form of Alzheimer’s.

This study will encourage researchers who propose that bacterial or viral pathogens cause some cases of sporadic Alzheimer’s see (Alzforum Discussion). This view is not widely held in the field, but it gained support from a presentation from Pat McGeer’s lab at University of British Columbia in Vancouver, Canada, reporting the cultivation of Borrelia burgdorferi spirochetes (the bacteria that cause Lyme disease) from Alzheimer brains.—Gabrielle Strobel.


  1. The finding of accumulation of β amyloid in persons exposed to HAART appears to parallel the situation with Parkinson disease and MPP. In both cases we have a natural experiment, where unintentional exposure to a toxic agent replicates some features of a sporadic neurodegenerative disorder.

    In the case of PD, intake of the mitochondrial complex I inhibitor, MPTP, reproduces clinical and pathological features of PD, and provides a model for research on the disease. The MPTP finding led to the discovery of complex I dysfunction in PD (Parker, 1989), the rodent model of PD (Greenamyre, 2003), and eventually to the recent finding of microheteroplasmic mtDNA mutations which correlate with the phenotype (Smigrodzki, 2004).

    One might hope that the availability of a human model of AD, which (in contrast to existing models based on β amyloid) has both the signature pathology and dementia, will reinvigorate AD research. Mitochondrial toxicity of HAART is well-established, and given the other evidence in favor of mtDNA dysfunction in AD (complex IV dysfunction (Parker, 1994), cybrid data (Khan, 2000), a mitochondrial mechanism for the present observation appears likely, although the confounding factor of direct viral neurotoxicity has to be accounted for.

    Analyzing the impact of HAART on mtDNA may eventually lead to progress in AD research, to parallel the developments in the PD field.


    . Abnormalities of the electron transport chain in idiopathic Parkinson's disease. Ann Neurol. 1989 Dec;26(6):719-23. PubMed.

    . The rotenone model of Parkinson's disease: genes, environment and mitochondria. Parkinsonism Relat Disord. 2003 Aug;9 Suppl 2:S59-64. PubMed.

    . High frequency of mitochondrial complex I mutations in Parkinson's disease and aging. Neurobiol Aging. 2004 Nov-Dec;25(10):1273-81. PubMed.

    . Electron transport chain defects in Alzheimer's disease brain. Neurology. 1994 Jun;44(6):1090-6. PubMed.

    . Alzheimer's disease cybrids replicate beta-amyloid abnormalities through cell death pathways. Ann Neurol. 2000 Aug;48(2):148-55. PubMed.

  2. The preliminary reports at the Society for Neuroscience meeting in San Diego that nucleoside anti-retroviral therapy can produce AD-like pathology in human beings is of great potential interest. HAART has been under scrutiny for some time, as it produces toxicities that include lipodystrophy, myopathy, and encephalopathy. The mechanism of this toxicity seems to center on the mitochondria, as nucleosides inhibit not only viral polymerases, but also the mitochondrial DNA polymerase, as well. This causes a drop in levels of mitochondrial DNA in a variety of tissues, an increase in ROS production, and an eventual drop in oxidative metabolism.

    The association between AD and mitochondrial dysfunction has also been the subject of much work (and is hypothesized in Swerdlow and Khan). That HAART initiates intraneuronal Aβ and diffuse plaque deposition further points to the mitochondria as the nexus for AD, emphasizing that amyloid is subsequent to mitochondrial dysfunction. How this mitochondrial dysfunction causes amyloid deposition is an area of active pursuit in our lab.

    Achim and Pulliam, however, attempt to identify extramitochondrial effects of the HIV drugs as inhibitors of amyloid degradation. Though a possibility, the vast majority of literature on HAART clearly shows mitochondria as the prime mediator of the toxicity of this class of drugs. Nevertheless, the findings of Achim and Pulliam may enable the first pharmacological models of AD that do not make a tautology out of Aβ.


    . A "mitochondrial cascade hypothesis" for sporadic Alzheimer's disease. Med Hypotheses. 2004;63(1):8-20. PubMed.

  3. The new studies by Pulliam and Achim provide an interesting parallel in HIV-infected humans to what has recently been reported by us in our studies of Chlamydophila (Chlamydia) pneumoniae infection of BALB/c mice (see Little et al., 2004). In essence, we observed plaques that labeled with antibodies to Aβ1-42 as well as intracellular neuronal Aβ1-42 immunoreactivity in some areas of the brain. The plaques developed over a one- to three-month period following intranasal inoculation of the organism. More plaques appeared after three months than after one, suggesting that longer infection times resulted in more plaques. More recent, but unpublished studies, show that more inoculations and longer postinfection times also result in larger and greater numbers of plaques in the brains. In the Achim study, plaque number correlated with length of time a person was infected with HIV, but not with the age of the person.

    One of the most important features of the two HIV studies is the precedence of infection "triggering" amyloid generation in the human brain. This provides evidence that pathogens can act as primary initiators of pathology consistent with Alzheimer disease. For a recent review, see Itzhaki et al., 2004.

    Given our limited understanding of how chronic infection results in chronic disease, we are surprised at how little credence is given to the research demonstrating connections between infection and AD and potentially many other neurodegenerative processes. Let us not forget how Warren and Marshall were initially vilified for their demonstration of the bacterial pathogen Helicobacter pylori connection with stomach ulcers, and now the wide acceptance of this pathology caused by the pathogen. This is but one example of many in which chronic diseases and infection are linked. Hopefully, the recent HIV studies and those mentioned in these comments will lead to expanded efforts in pathogen associations in neurodegenerative diseases.


    . Chlamydia pneumoniae induces Alzheimer-like amyloid plaques in brains of BALB/c mice. Neurobiol Aging. 2004 Apr;25(4):419-29. PubMed.

    . Infiltration of the brain by pathogens causes Alzheimer's disease. Neurobiol Aging. 2004 May-Jun;25(5):619-27. PubMed.

  4. It was good to read the report on the AIDS study presented at San Diego by Lynn Pulliam and colleagues, because her poster on this topic at the International AD conference in Philadelphia in July might not have been seen by many. However, it is worth mentioning that back in 1998, Margaret Esiri and colleagues reported detection of argyrophilic amyloid plaques in the cerebral cortex (temporal and frontal lobes) of 97 AIDS patients; in the controls and the AIDS patients, the plaque prevalence increased with age and was significantly greater in the AIDS group as a whole. (Incidentally, in another virus-induced brain disease, sub-acute sclerosing panencephalitis, the agent—measles virus—has long been known to induce the formation of NFT, the other hallmark of AD brain). These studies, and that of Achim et al., add to the diverse body of work linking β-amyloid or APP with viruses, such as the detection of a sequence homology between the glycoprotein gB of HSV1 and β-amyloid (Cribbs et al., 2000), and evidence of the association between APP and HSV1 during axonal transport of the virus (Satpute-Krishnan et al., 2003)—as well as our work here showing the effect of HSV1 on APP degradation in cultured cells (Dobson et al., 2002, and submitted). Further, Corder et al. (1998) found that the ApoE4 genotype increased susceptibility to dementia and peripheral neuropathy in pre-AIDS HIV-infected individuals—strikingly parallel to our findings (Itzhaki et al., 1997) that herpes simplex virus type 1 in brain of ApoE4 carriers is a risk for AD, and ApoE4 is a risk for cold sores (which are caused by HSV1). Thus, HIV infection can lead to a dementing illness that involves amyloid deposition and is influenced by ApoE, though its primary cause is, indisputably, viral infection. This sets a clear precedent for viral causation of a neurodegenerative condition that has appreciable overlap with AD.

    Brian Balin has mentioned his group’s important animal studies linking another pathogen, Chlamydia pneumoniae, to AD, as well as the striking difficulties in even initiating discussions on these topics, let alone serious attempts to replicate them. Brian reminds us of the hostility to the discovery of H. pylori in stomach ulcers, but much earlier, Pasteur’s theory of germs was ridiculed, just as later, a viral involvement in certain types of cancer was derided. Why are new concepts treated thus? Surely, it is not just lack of knowledge of virology or bacteriology. Let’s hear what the scientific objections are.


    . Prevalence of Alzheimer plaques in AIDS. J Neurol Neurosurg Psychiatry. 1998 Jul;65(1):29-33. PubMed.

    . Fibril formation and neurotoxicity by a herpes simplex virus glycoprotein B fragment with homology to the Alzheimer's A beta peptide. Biochemistry. 2000 May 23;39(20):5988-94. PubMed.

    . Fast anterograde transport of herpes simplex virus: role for the amyloid precursor protein of alzheimer's disease. Aging Cell. 2003 Dec;2(6):305-18. PubMed.

    . Abnormal processing of amyloid precursor protein after acute infection of human neuroblastoma cells with herpes simplex virus type 1. Neurobiol Aging. 2002 Jul-Aug; 23(Suppl 1):394.

    . HIV-infected subjects with the E4 allele for APOE have excess dementia and peripheral neuropathy. Nat Med. 1998 Oct;4(10):1182-4. PubMed.

    . Herpes simplex virus type 1 in brain and risk of Alzheimer's disease. Lancet. 1997 Jan 25;349(9047):241-4. PubMed.

Make a Comment

To make a comment you must login or register.


Other Citations

  1. Alzforum Discussion

Further Reading