Do Sleeper Viruses Awaken in ALS?

A hibernating retrovirus, prodded from slumber, can provoke neurodegeneration, according to a paper in the September 30 Science Translational Medicine. Human endogenous retrovirus-K pumps out viral RNA and proteins in a subset of amyotrophic lateral sclerosis cases, according to senior author Avindra Nath and colleagues at the National Institute of Neurological Disorders and Stroke in Bethesda, Maryland. Moreover, rousing HERV-K in cultured cells killed neurons, and doing the same in mice sickened them in an ALS-like manner. “We are pretty comfortable in saying that the virus can cause motor neuron damage,” Nath said, cautioning, “What we haven’t shown is that it is a cause of ALS.”

Insidious infection.

Once awakened in one neuron, HERV-K particles might induce neurodegeneration in its neighbors. If so, anti-retroviral therapy might help. [Courtesy of Avindra Nath.]

Nath studies a more famous retrovirus, the human immunodeficiency virus that gives rise to AIDS, and its neurological effects, such as dementia. In 2006, he treated a man who had both AIDS and an ALS-like syndrome, piquing his interest in the overlap between these two diseases. Once the patient started anti-retroviral therapy, his motor symptoms improved. As it turns out, scientists have reported a handful of such HIV cases over the years, and some regained their motor skills after anti-retroviral treatment. Scientists have also found reverse transcriptase, a marker for retrovirus activity, in the brain, serum, and cerebrospinal fluid of some people with sporadic ALS. However, they could not find evidence for infectious retroviruses associated with the disease, leading Nath to hypothesize that endogenous retroviruses might be involved (reviewed in Alfahad and Nath, 2013).

People possess scores of endogenous retroviral DNAs in their genomes, left over from retroviruses that copied themselves into human DNA at some point over the millennia and then fell dormant. These interloper sequences make up 8 to 10 percent of the human genome. In a 2011 paper, Nath and colleagues reported that DNA from one such intruder, HERV-K, was transcribed in the brain tissue of people who died of ALS, though they did not confirm it affected the disease process (see Dec 2010 news). The human genome houses dozens of copies of the HERV-K genome in various states of degradation compared with the active viral genome.

In the new study, first author Wenxue Li and colleagues followed up on that finding, asking if HERV-K contributes to neurodegeneration. Li examined brain tissue from 11 people who died of ALS and 16 controls. Nine of the cases had sporadic ALS, while one person carried a C9ORF72 expansion and another exhibited signs of both ALS and frontotemporal dementia. Li amplified RNA for HERV-K genes in those cases and controls. HERV-K expression varied from person to person, but overall was higher in ALS cases than in controls. Nath estimated that about 30 percent of the ALS cases possessed appreciable copy numbers of HERV-K RNA. The scientists found no correlation between clinical phenotypes and HERV-K expression levels.

The authors also stained postmortem tissue from 10 people who had died of ALS with antibodies for the HERV-K envelope protein and detected binding in cortical large pyramidal neurons and spinal cord anterior-horn neurons. HERV-K protein was absent from the brain tissue of 10 people who died of Alzheimer’s disease.

Could this reactivated, vintage virus do harm? Li transfected its genome, or just the envelope gene, into human neural cultures derived from pluripotent stem cells. The cells retracted their neurites and died. To discern whether triggering the endogenous HERV-K lurking in the genome could do the same, Li used a trick based on the CRISP/Cas9 gene editing system (see Sep 2014 series). Instead of altering the genome, he used the CRISPR/Cas9’s gene-homing ability to target transcription factors to an HERV-K gene. This turned the gene on, again causing the neurons to pull back projections and degenerate.

Next, co-author Myoung-Hwa Lee activated HERV-K in mice. As a quick test of its effects in animals, she electroporated the env gene into wild-type mouse brain before the pups were born. Two weeks after birth, the animals’ neurons appeared odd, with swollen beads along their neurites. Encouraged by this success, Lee generated a new line of transgenic mice, expressing the env gene under control of the Thy1 promoter in their neurons. They produced env RNA at levels about twice as high as seen in human brains. When Lee examined the tissues of 6-month-old mice, she observed their dendrites were shortened and less branched than normal, and the dendritic spines were fewer in number and abnormally shaped. Astrocytes ran amok, a common effect during neurodegeneration. As in human ALS, both upper and lower motor neurons degenerated. Compared with wild-type animals, the transgenic mice lacked about one-third of their corticospinal motor neurons and the volume of the motor cortex was reduced by about a fifth. Some parts of the spinal cord had barely any motor neurons.

These defects resulted in progressive motor symptoms. Compared with wild-type mice, the transgenics wandered less in an open field, and rested more often. They were quicker to tumble off a rotating rod, a common test of strength and coordination for ALS mice. Half were dead within 10 months. Defects were worse and progression faster than in the commonly used SOD1 model of ALS. “These env mice look like models that express genes associated with ALS,” said Josh Dubnau of Cold Spring Harbor Laboratory in New York.

Stanley Appel of the Houston Methodist Neurological Institute was not so sure. “I get the feeling that the pathology in this model is more in the brain than in the spinal cord,” he said, noting that lower-motor-neuron problems predominate in ALS. Neither Appel nor Dubnau were involved in the study.

The env expression was clearly toxic, but how? In the mice, the authors observed markers for double-stranded DNA breaks as well as disruption to the nucleolus. Nucleophosmin, normally a nucleolar membrane protein, had translocated to the cytoplasm. Via chromatin immunoprecipitation, Li found that HERV-K DNA bound directly to TDP-43, a protein implicated in ALS pathology. In the human neural cultures, Li determined that overexpression of TDP-43 activated HERV-K, while TDP-43 knockdown dampened its expression. In their previous study, TDP-43 expression correlated with that of the HERV-K reverse transcriptase, and the proteins co-localized, in the brain tissues of ALS cases. TDP-43 regulates transcription of numerous genes, including sequences in HIV and transposable elements, another kind of genomic stowaway (see Mar 2011 news; Nov 2010 news; Ou et al., 1995; Li et al., 2012). HERV-K expression depends on TDP-43, Nath concluded.

Dubnau noted that the idea of a retrovirus or retrotransposon-inducing chronic disease has precedent. For example, transposable Alu elements, when transcribed, induce a form of macular degeneration (Kaneko et al., 2011). Many cancer-causing viruses copy themselves into the host cell genome (Carrillo-Infante et al., 2007). HERV-W abets inflammation in multiple sclerosis, for which scientists are considering it as a treatment target or biomarker (Perron and Lang, 2010; Curtin et al., 2015). Herpes retrovirus for many years has been proposed as a pathogenic factor in sporadic Alzheimer’s, though the idea has not caught on (Harris and Harris, 2015; Feb 2011 webinar).

Dubnau speculated that HERV-K might be the first of many retroviruses or retrotransposons to be found stirring in ALS. He suggested it may be a “canary in the coal mine” signaling a general problem with silencing of toxic genetic elements. The DNA damage Li observed could result from the insertion of reverse-transcribed elements, he suggested. Dubnau and colleagues have discovered an endogenous retrovirus involved in neurodegeneration in fruit flies overexpressing TDP-43.

Scientists who spoke with Alzforum were enthusiastic about Nath’s study, and eager for more information about HERV-K and ALS. “It does explain why people found retroviral traces [in ALS] without any evidence of a real virus,” Appel said. “Nath’s data points out elegantly that it is an activation of latent viral genomes that sit in all of us.” He was curious about whether HERV-K also awakens in genetic cases of ALS, particularly TDP-43 cases. Jeremy Garson of University College London, who did not participate in the research, noted that it raises numerous questions about what subsets of ALS might express HERV-K, whether people with the disease have extra copies of the integrated viral genome or activate certain variants, and what other factors might spur HERV-K revival. “It will be important for the major findings of this work to be replicated independently by other research groups,” he added (see full comment below).

Robert Brown of the University of Massachusetts Medical School in Worcester and Ammar Al-Chalabi of King’s College London, who co-wrote an editorial commentary in the same issue of Science Translational Medicine, agree the study raises many new questions. “Perhaps most fundamental is the proverbial chicken-and-egg challenge: Is the activation of HERV-K a cause or consequence of motor neuron degeneration?” they wrote. In addition, “Can one now define a cocktail of anti-retroviral interventions that are beneficial, and will the activity of the retrovirus prove to be a quantitative biomarker of disease activity?”

Nath is working on answering these questions, as well as the possibility of treating retroviral activation in ALS. He recently secured approval for a Phase 1 trial of a four-part anti-retroviral cocktail. For now, he aims not to necessarily treat the disease, but just to find out if the medications curb HERV-K titers in the blood. Nath cautioned that people with ALS should not try this treatment outside of a trial. “These are not innocuous drugs, and we do not know if the cocktail will work,” he pointed out. Typically, such treatments affect disease course only if they almost completely eliminate the retrovirus, he added.

Dubnau wondered if any HERV-K retroviral DNA hiding in the human genome can produce a competent, infectious virus. If so, it might transmit ALS from cell to cell (see image above). Appel believes the viral spread would likely incorporate glia somehow, as well as neurons. Scientists have labelled HERV-K a “dead” virus, able to express toxic gene products but not assemble a complete infectious particle. However, they based that conclusion on the analysis of just a few reference DNA samples, which indicated the HERV-K sequences in the human genome were incomplete or contained point mutations. Might some people have a functional HERV-K genome, waiting to be revived? To determine that will be difficult, Dubnau noted, because there are many copies of repetitive sequences that are challenging to assemble by short-read sequencing. Nonetheless, Nath has started a project to examine HERV-K copy number and polymorphisms. He also plans to look for cell-to-cell spread of HERV-K in vitro, and research possible interventions in his animal model. Furthermore, Nath plans to explore HERV-K in tissue samples from people who had frontotemporal dementia, which includes TDP-43 proteinopathy in some cases.


1. Li W, Lee MH, Henderson L, Tyagi R, Bachani M, Steiner J, Campanac E, Hoffman DA, von Geldern G, Johnson K, Maric D, Morris HD, Lentz M, Pak K, Mammen A, Ostrow L, Rothstein J, Nath A. Human endogenous retrovirus-K contributes to motor neuron disease. Sci Transl Med. 2015 Sep 30;7(307):307ra153. [PubMed].

2. Brown RH Jr, Al-Chalabi A. Endogenous retroviruses in ALS: A reawakening?.Sci Transl Med. 2015 Sep 30;7(307):307fs40. [PubMed].

To view commentaries, primary articles and linked stories, go to the original posting on here.

Copyright © 1996–2018 Biomedical Research Forum, LLC. All Rights Reserved.

disease-als HERV-K topic-preclinical virus
Share this: