HIV-Associated Neurocognitive Disorders and the Gut Microbiome

Although antiretroviral therapy helps individuals living with HIV keep their viral loads suppressed, we're still seeing the rise of other comorbidities, such as HIV-associated neurocognitive disorders (HAND). One study set out to evaluate the role that the microbiome plays in HIV-associated neurocognitive disorders. Terri Wilder, M.S.W., spoke with study author Josue Perez-Santiago, Ph.D., at CROI 2017, in Seattle.

Terri Wilder: Can you take our readers through your study, from background and significance all the way to your conclusions?

Josue Perez-Santiago, Ph.D.: Sure. Even in the antiretroviral therapy era, people still experience HIV-associated neurocognitive disorders. The mechanisms regarding HAND, or HIV-associated neurocognitive disorders, remain unclear. This limits the development of interventions that actually can treat neurocognitive disorders.

On the other hand, the gut microbiome has been implicated in the development and the maturation of, basically, the wiring in our brain that deals with cognition and emotions. So, what we wanted to do here is to actually see, evaluate, the role of the microbiome in terms of HIV-associated neurocognitive disorders.

We had 126 patients that were recruited in the HIV Neurobehavioral Research Program at UC San Diego. From those 126, 96 patients were HIV infected, and 30 were uninfected individuals. We collected stool, clinical data and sociodemographic data.

Also, a subset of these participants underwent neuropsychological testing. [Of these,] there were 66 HIV-infected individuals and 24 uninfected individuals. The HIV-infected individuals were put in four different categories according to their cognitive status and [an assessment of] their daily activities. They were either cognitive normal or diagnosed with HIV-associated dementia, mild neurocognitive disorders or asymptomatic neurocognitive impairment.

The uninfected individuals, since they didn't have HIV -- even though the algorithm for this diagnosis was the same -- they were only classified as either normal or NPIO, which means neuropsychological impairment due to other factors that are unrelated to HIV. In addition, these participants also were screened for depression based on the self-report.

All these participants, we collected stool and then we extracted DNA. We used ultra-deep sequencing to do a 16S ribosomal profiling to see the populations of bacteria that were in these participants. Basically, in the demographics there was no significant difference between the HIV positive and the negative other than in gender -- which might be just a sampling bias because in San Diego we mainly have men that are infected. But there was no significant difference between all the other clinical variables.

In the subset [of 66] that had neuropsychological assessment, there were 31 that were normal, [and] 23 that had asymptomatic neurocognitive impairment -- which, I have to mention; this just means that they were impaired in, if I remember properly, in two domains. But they did not have any complaints of their daily activities. They could do everything without any problem.

There were eight participants that were diagnosed with mild neurocognitive disorder, which -- even though they had the same impairment, they did complain about having problems with daily activities. There were four people that had HIV-associated dementia.

In the group of HIV negative (we have 24), 13 of them were classified as cognitive normal [and] 1l of them as impaired.

In the first analysis that we did, we looked at the phylum level of the microbiome. And we looked firstly to the Firmicutes and Bacteroidetes ratio, which is a measurement of dysbiosis. And people look at it also as a marker of aging.

In HIV-negative individuals, we did not see any difference between the groups in the Firmicutes to Bacteroidetes ratio. When we looked at it in the HIV population, in the different diagnoses of HAND, we did see a reduction in the Firmicutes/Bacteroidetes ratio in the people that had associated dementia compared to people with mild neurocognitive disorders or asymptomatic neurocognitive impairment.

TW: For our audience, can you define what this is?

JP-S: People are used to seeing species. But according to the genetic diversity of the bacteria that classifies in different groups, right? So the Firmicutes are one of the big -- the phylum level is the biggest group that bacteria can be grouped, based on their genetic diversity.

In the phylum level, I don't remember how many phyla there are, but Firmicutes and Bacteroidetes are basically the most prevalent, the most dominant that we have in our gut. And normally we have like a one-to-one ratio, more or less.

In HIV, in people who had some sort of neurocognitive impairment in HIV, we did see a reduction of the Firmicutes/Bacteroidetes ratio in people with dementia. But not only did we see that ratio actually was going down; when we actually saw the groups separately, we saw a significant increase of Bacteroidetes in people with dementia and a significant reduction of Firmicutes in people with dementia.

Like I mentioned before, the lower of the Firmicutes-to-Bacteroidetes ratio is associated with aging. So, as people age, you expect that go to down, which sort of makes sense in this context.

TW: Since your cohort was a little older, looks like.

JP-S: And associated dementia is the most severe diagnosis of HAND; so, in terms of phenotype, it's something that you would expect in very old populations. So, it's consistent with that idea that people who have dementia are the ones who have the lowest Firmicutes-to-Bacteroidetes ratio.

TW: And then, the cohort that you looked at, the average age was about 55, which, in the world of HIV, is considered an older adult.

JP-S: Because they're already old. Yes. After 50, it's already an older adult.

So, the differences in the species level, we found -- this particular species that I mentioned here -- all of them were reduced in HIV infection. And the interesting thing about all these species is that they can be put into a mechanism that reduces this metabolite that they produce in the gut called butyrate. So, to put you in the context of the mechanism, Bifidobacterium and Bacteroides, what they do in the gut, part of their function, is to break down the complex sugars into simple sugars.

The Faecalibacterium here, what it does is take those simple sugars, and by fermentation it produces Butyrate and hydrogen gas as a byproduct.

If you see here, we have Bacteroides, Bacteroides, Bifidobacteria and Bacteroides; you can think that this mechanism here that converts the complex sugars into simple sugars might be affected by the reduction of this bacteria. So, this might not be happening. But even if it was, we also see a reduction of this Faecalibacterium, which will inhibit the production of butyrate.

On the other hand, we have this Blautia obeum. And the function of Blautia obeum is to take hydrogen gas and basically produce acetate. We also see a reduction of Blautia obeum. What that means is that if there's any reaction that produces hydrogen gas as a product, it basically will accumulate, because we have a reduction of this bacteria. And accumulation of hydrogen gas inhibits fermentation. Inhibiting fermentation also means reduction in the butyrate levels.

I also have to mention that lower amounts of Faecalibacterium were associated with higher depression.

Also, higher levels of Bacteroides caccae were associated with better cognition. But I have to mention that that cognition was not clinically based. It was in self-reports. They will also give a score based on what the patients actually put in the forms. It was basically that -- which suggests that if participants are complaining about some sort of problems in cognition, it might be that they have some sort of symptom going on that might not be captured by this yet.

In conclusion, despite the sample size, we saw a reduction of the Firmicutes/Bacteroidetes ratios in people that have associated dementia. And, as I mentioned, that is also associated with aging.

All this bacteria that we saw reduced in HIV infection are all involved in butyrate production. Lower butyrate production has been associated with higher microbial translocation and higher immune activation, which are both associated also with incidence of neurocognitive impairment.

These bacteria, in particular here, Bifidobacterium longum, have been used as a probiotic to treat psychiatric disorders, so they call it a psychobiotic. And butyrate has been used as an experimental drug in mouse models for neurocognitive disorders, as well.

In conclusion, we actually think that there might be a microbial signature that probably can be used in the future, maybe to have an intervention and possibly alleviate the problem of impairment in HIV infection.

TW: Great. Basically, you're looking at: Is there this connection between the gut and neurocognitive [function]? And if there is, is there a way to heal the gut?

JP-S: Yeah. Can we change the gut in our advantage so that we can actually make people better, in terms of cognition, and even in the pathogenesis of HIV? All the same variables that are associated here are associated with general HIV progression. If we actually alleviate here, probably we will also make these participants' clinical outcomes better.

TW: Great. Thank you so much.

JP-S: Thank you.

This transcript has been lightly edited for clarity.

Terri L. Wilder, M.S.W., is a director of HIV/AIDS education and training in New York City.