Alzheimer's Part III: Antiretrovirals

Previously, we mentioned the relevance of antiviral therapy to Alzheimer’s disease. Specifically, we’re looking at a 2011 study, Significant Effects of Antiretroviral Therapy on Global Gene Expression in Brain Tissues of Patients with HIV-1-Associated Neurocognitive Disorders, in which post-mortem brains of HIV patients were compared according to several parameters, one of them being the usage of antiretrovirals. Utilizing our lists of transcripts that are canonically up/down-regulated in Alzheimer’s, one sees an impressive alteration of these transcripts on antiretroviral therapy:

The P-values work out to about 10^-4 and 10^-43. You can go through the exercise yourself: just choose the “Match Studies” tool, input the IDs for the two “canonical” Alzheimer’s lists (123049121 and 123050121), choose “drug” under “Experiment Type”, and choose “Inverse Correlations” (that way, you’ll be looking for datasets that counter Alzheimer’s). The ensuing “p_multiple” is simply the result of multiplying -4 * -43.

Here are transcripts at the intersections:

up-Alz/down antiretrovirals		down-Alz/up antiretrovirals
APLNR		ARPP21	ELAVL4	NEFL	SNAP25
CD163		B4GALT6	GABRA1	NEFM	SST
CD44		BEX5	GABRG2	NELL2	STMN2
CXCR4		CADPS	GAD1	NRN1	SV2B
NUPR1		CALB1	GLRB	NRXN1	TSPAN7
SERPINA3		CAP2	GNG3	NSG2	VSNL1
		COPG2IT1	HCN1	OLFM3
		CPNE4	HOPX	RGS4
		DCLK1	INA	RPH3A
		DIRAS2	KCNIP4	SCN3B

If you use our tool as above, you’ll note that there really are no other treatments with this profound effect. Next in line, somewhat surprisingly, would be a study in which neuroinflammation was induced in mice retinas!

Now, we cannot simply conclude that antiretrovirals would be effective against Alzheimer’s. We note that the single best drug that mimics the transcriptomic effects of Alzheimer’s is cocaine. A Google Scholar search, however, reveals no connections between cocaine use and Alzheimer’s. Drugs that cause Alzheimer’s –like transcriptomic effects don’t necessarily cause Alzheimer’s; drugs that reverse Alzheimer’s transcriptomic effects cannot be assumed to work as treatments.

Are there other studies in our database in which brain tissue was examined following the use of antiretrovirals? There’s one, an apparently unpublished rhesus study with GEO accession GSE117336. In this case, there aren’t any interesting correlations with our Alzheimer’s lists.

However, let’s take the human results at face value for the moment. The simplest conclusion would be that Alzheimer’s is caused by a viral infection. Such a view cannot be considered “fringe”, but neither is it a leading hypothesis. Naturally, there are results both supporting and disputing the Alzheimer’s/virus link. Most of the studies supporting the link refer to herpes virus, which is not a retrovirus and does not seem to be strongly inhibited by antiretrovirals. Alternatively, the antiretrovirals may be acting upon a host-intrinsic process. Retrotransposition? Here, one paper may be particularly relevant: Somatic APP gene recombination in Alzheimer’s disease and normal neurons. In addition to suggesting mechanisms by which retrotransposition may relate to Alzheimer’s, the paper also points out a low incidence of Alzheimer’s in HIV patients on antiretrovirals!

One should bear in mind that there are, in a sense, two variables being tweaked in the antiretroviral study; the application of a drug, and the load of a virus. Thus it is possible that the untreated patients are progressing to an Alzheimer's like-state with great rapidity, and the antiretroviral drugs are not actually "reversing" any neuropathology. In this case, the question is not "how are the antiretrovirals working?", but "how does HIV cause Alzheimer's-like transcriptomics?"

whatismygene.com 

Become a Patron!