. Human endogenous retrovirus-K contributes to motor neuron disease. Sci Transl Med. 2015 Sep 30;7(307):307ra153. PubMed.


Please login to recommend the paper.


  1. Several prior strands of evidence (mainly indirect) pointed to the possibility of retroviral involvement in ALS. These can be summarized as follows:

    • Since the 1980s, certain murine retroviruses have been known to cause a motor-neuron disease in mice.
    • The human retroviruses HIV and HTLV both can cause neurological disease with some similarities to ALS.
    • Symptoms of the ALS-like syndrome associated with HIV infection in some cases may be completely reversed by anti-retroviral treatment.
    • Patients with ALS frequently have circulating antibodies that cross-react with various retroviral proteins.
    • Our research group demonstrated in three separate studies that the retroviral enzyme reverse transcriptase is found in the serum of patients with sporadic ALS significantly more frequently than in controls (McCormick et al., 2008; Steele et al., 2005; Andrews et al, 2000). Known human exogenous retroviruses were excluded as potential sources for this reverse transcriptase activity and an endogenous retroviral origin was therefore proposed.

    The new study expands on earlier work by the same authors (Jeffrey Rothstein and Avindra Nath), which suggested, on the basis of preliminary findings, that the source of the reverse transcriptase in ALS patients may be the human endogenous retrovirus HERV-K (Douville et al., 2011). The new study is substantial and impressive in that it has employed a wide range of powerful in vivo and in vitro molecular biological, immunological, genetic, and other techniques to demonstrate that (a) HERV-K is expressed in the brain of ALS patients at higher levels than in controls, (b) HERV-K expression in human neurons causes toxicity, (c) HERV-K expression in transgenic mice causes a form of motor neuron disease with some similarities to ALS, and (d) HERV-K expression may be regulated by the DNA-binding protein, TDP-43. 

    A key question is whether the increased expression of HERV-K in the brains of ALS patients is a cause or a consequence of the disease. In other words, is it an essential part of the pathogenic process or simply an epiphenomenon? In their new study, Li and colleagues have gone some way to answering this by demonstrating not only that elevated HERV-K expression can cause neurotoxicity but also that the elevated expression is not just a non-specific consequence of neuronal injury. The new data presented in the paper represent a significant contribution to our understanding of the potential role of endogenous retroviruses in neurological disease.

    In general, the methodology employed seems appropriate, although there are a few minor concerns. First, the RT-PCR assay used to measure HERV-K RNA expression in brain and neuronal cultures employed GAPDH RNA as a reference for normalization without demonstrating that GAPDH expression levels were stable. For accurate measurement of HERV-K RNA by this relative quantification method it is essential to know that the GAPDH RNA levels were the same in the ALS and control brains. The importance of validating reference genes in RT-PCR is emphasized in the MIQE Guidelines (Bustin et al., 2009) as follows: “Normalization against a single reference gene is not acceptable unless the investigators present clear evidence for the reviewers that confirms its invariant expression under the experimental conditions described.” Ideally, full MIQE (Minimum Information for Publication of Quantitative Real-Time PCR Experiments) details should have been included within the supplementary materials.

    Second, the demonstration of HERV-K RNA expression in HERV-K transfected human neurons in culture was not accompanied by proof of concomitant protein expression.

    Third, the attempt to demonstrate the specificity of HERV-K elevated expression in ALS by measuring the expression of other HERV families was not very comprehensive, and importantly failed to measure HERV-W expression. HERV-W expression levels would have been of particular interest here because HERV-W has been implicated in other human neurological/psychiatric disorders, including multiple sclerosis and schizophrenia.

    It will be important for the major findings of this work to be replicated independently by other research groups. In addition, future questions to be addressed should include the following:

    • Is elevated HERV-K expression found in familial ALS cases as well as sporadic?
    • Is there a particular phenotype or subset of ALS associated with elevated HERV-K expression?
    • Is there evidence of elevated HERV-K expression found in the blood and/or cerebrospinal fluid (CSF) of ALS patients or is it restricted to the brain and spinal cord?
    • Can HERV-K expression levels in brain, blood, or CSF be used as a diagnostic marker for ALS or be used to monitor disease progression and response to therapy?
    • Is there any evidence for an increase in HERV-K DNA copy number in ALS?
    • Are any particular sequence variants of HERV-K specifically associated with ALS? Might these represent novel genetic factors predisposing to ALS?

    The regulation of endogenous retroviral latency is complex and a variety of agents, including physical, chemical, infectious, and immunologic, have been shown to be able to induce their activation. What infectious or other environmental factors might induce abnormal expression of HERV-K in ALS patients?

    The work described in this paper is unlikely to have immediate practical implications for patients currently suffering from ALS, although it is quite possible that the findings may initiate the search for novel approaches to therapy, directly or indirectly targeting retroviral nucleic acids, proteins, or associated pathways. It may be appropriate to consider establishing formal clinical trials of existing anti-retroviral drugs.

    For researchers in the field, these findings add further support for the theory of retroviral involvement in ALS. The HERV-K transgenic mouse may have potential as an animal model to facilitate the development and testing of existing and novel therapeutic agents for the treatment of patients with ALS.


    . Quantification of reverse transcriptase in ALS and elimination of a novel retroviral candidate. Neurology. 2008 Jan 22;70(4):278-83. PubMed.

    . Detection of serum reverse transcriptase activity in patients with ALS and unaffected blood relatives. Neurology. 2005 Feb 8;64(3):454-8. PubMed.

    . Detection of reverse transcriptase activity in the serum of patients with motor neurone disease. J Med Virol. 2000 Aug;61(4):527-32. PubMed.

    . Identification of active loci of a human endogenous retrovirus in neurons of patients with amyotrophic lateral sclerosis. Ann Neurol. 2011 Jan;69(1):141-51. PubMed.

    . The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin Chem. 2009 Apr;55(4):611-22. Epub 2009 Feb 26 PubMed.

    View all comments by Jeremy Garson
  2. This interesting and provocative paper carries on a theme that has been pursued for a number of years by Nath and colleagues. The investigators find that a subset of ALS patients has increased expression of HERV-K genes and increased immunoreactivity compared to controls of the HERV-K envelope (env) protein in motor neurons in autopsy tissue. In vitro studies show that overexpression of env leads to degeneration of IPSC-derived human neurons. Furthermore, transgenic mice that overexpress env under the thy-1 promoter have motor deficits, loss of motor neurons, denervation of muscle, and early death. The investigators show that TDP-43, which is mislocalized in most cases of ALS, binds to the HERV-K LTR, thereby leading to transactivation of HERV-K.

    It is instructive to review motor neuron diseases associated with mouse retroviruses. Infection of mice with a temperature-sensitive mutant of the Moloney murine leukemia virus, called ts1, causes progressive non-inflammatory hind-limb paralysis that is thought to result from apoptosis of glial cells leading to ER stress in neurons and culminating in spongiform changes in the brain stem and ventral horn of the spinal cord (Kim et al., 2004); transgenic mouse studies indicated that expression of env led to disease. Another related murine leukemia virus, Cas-Br-E-MuLV, also produces paralysis with similar spongiform changes concentrated in the lumbar spinal cord anterior horn (Kay et al., 1993). As with ts1, infection of non-neuronal cells and a key role of env (as suggested by chimeric virus and transgenic mouse studies) have been implicated in disease pathogenesis. More complicated is age-dependent poliomyelitis, a murine disease that requires multiple proviral copies of an endogenous murine leukemia virus, infection with lactic dehydrogenase virus, a permissive allele in the mouse, and an immunocompromised state (Contag et al., 1992). Of note, the immunocompromised state could result from age, which, interestingly, is the most significant risk factor in developing sporadic ALS. 

    Several questions arise from the paper by Nath and colleagues. The paper’s focus on the effect of env makes one wonder about the effect of gag and pol, the two other major genes encoded by retroviral genomes. How efficiently is HERV-K transactivated when TDP-43 is mislocalized to the cytoplasm, as is true in most cases of ALS? Do cases of mutant SOD1-induced familial ALS have increased expression of HERV-K despite the reported absence of TDP-43 abnormalities? Is HERV-K expression increased in IPSC-derived motor neurons from sporadic cases of ALS or from familial ALS patients with mutations of C9ORF72?

    The authors are appropriately cautious about the role of HERV-K in ALS pathogenesis since the mislocalization of TDP-43 causes varied splicing abnormalities with abnormal expression of multiple genes, including, as demonstrated in the present paper, those of an endogenous retrovirus.  It is likely that there are a number of bullets that contribute to the death of motor neurons in ALS.


    . Neurological disease induced in transgenic mice expressing the env gene of the Cas-Br-E murine retrovirus. Proc Natl Acad Sci U S A. 1993 May 15;90(10):4538-42. PubMed.

    . Activation of endoplasmic reticulum stress signaling pathway is associated with neuronal degeneration in MoMuLV-ts1-induced spongiform encephalomyelopathy. Lab Invest. 2004 Jul;84(7):816-27. PubMed.

    . Pathogenesis of age-dependent poliomyelitis of mice. In: Molecular Neurovirology Pathogenesis of Viral CNS Infections. Ed. R.P. Roos. Humana Press, 1992.

    View all comments by Raymond Roos
  3. This paper by Li and colleagues on the contribution of endogenous retroviruses (HERV) of the K family to the pathogenesis of ALS is exciting. Currently, ALS is a disease without effective treatments. Therefore, this insight into its possible pathogenesis opens new horizons and, hopefully, new room to manoeuvre.

    For more than 16 years I have studied another HERV family member, HERV-W, in the onset and neuro-pathogenesis of multiple sclerosis (MS). I knew of the existence of HERV-K expression in ALS from a poster by the Dr. Nath’s group at the 2013 meeting of the International Society of Neurovirology. What made me literally jump then was the sight of a nice, clear, electrophoretic band of what looked as a retro-transcribed cDNA from a full-length HERV genome, extracted from a brain sample of an ALS patient. Since then, every time I met Dr. Nath, I expressed my deep interest about their findings and the role of HERV-K in ALS neurodegeneration

    MS and ALS are quite different: Primarily, ALS affects neurons in the gray matter of the brain, while MS affects oligodendrocytes in the white matter. The behavior and progression of the two diseases have completely different patterns. However, both diseases start focally, and then the pathology spreads. The etiology of both diseases is uncertain, complex, and multifactorial. A genetic component is postulated for MS and, at least partly, for ALS, while some interaction(s) between genetic and environmental factors are believed to influence both.

    For MS, several independent studies, including ours, have shown a direct correlation between disease onset/ progression and expression of HERV-W/MSRV/Syncytin-1. The HERV-Wenv protein has neuropathogenic and immunopathogenic properties, both in vitro and in animal models, that can account for MS neurodegeneration. HERV-Wenv protein has been detected in glial cells surrounding the degenerating nerve sheaths. Currently, a Phase 2 clinical trial of an antibody to the HERV-Wenv protein is in progress for MS patients and seems promising.

    In the ALS cases studied by Li and colleagues, the HERV-Kenv protein is detected within cells relevant to the ALS disease and it is activated by the TDP-43 protein, a hallmark of ALS. Moreover, transgenic mice expressing the HERV-Kenv protein developed progressive motor dysfunctions, reminiscent of human ALS.

    It is not surprising that HERV-Kenv has these effects. The env proteins that make up the external envelope, or pericapsid, of retroviruses, including the gp120 env of the HIV virus and the aforementioned HERV-Wenv, have several neuropathogenic properties.

    In my opinion, the reason why this finding of HERV-K in an ALS subset is important is that it raises the possibility, so far for some patients only, of changing ALS by targeting the virus.  

    If further studies confirm HERV-Kenv expression in other ALS cohorts, and if its levels turn out to correlate with disease progression, then we can be allowed to hope that future innovative treatments aimed at HERV-K inhibition may attenuate ALS pathology.

    I am acutely aware that such optimism may be easily dashed, but this is a research path that deserves deeper consideration.

    View all comments by Antonina Dolei
  4. Even though I like the study, which is solid and agreeable in part, I have some philosophical points to make that may help unveil more concrete associations between ERV and human disease. I believe that it is too early to conclude that HERV-K expression within neurons of patients with ALS contributes to neurodegeneration and disease pathogenesis. This is because most HERVs are harmless and all human cells and all individuals have essentially the same set of HERVs. Thus, the negative correlations shown by Nath’s group show only indirectly how an HERV can be tied to a disease in humans—in this case ALS.  The proof that HERV is the etiological agent of ALS may never materialize, as the Koch postulates or similar rules adapted to present-day techniques mainly pertain only to exogenous agents.

    Eight percent of the human genome consists of human endogenous retroviruses, or HERVs, and, if we extend this to HERV fragments and derivatives, the retroviral legacy amounts to roughly half our DNA (Ryan, 2004). Interestingly, both human class I and class II MHC genes carry a high density of HERV elements compared to other multi-locus-gene families (Bartholomew and Ihle, 1991). Striking to note is that HERVs have contributed to the formation of extensively duplicated duplicon blocks that make up the HLA class I family of genes inherited as immuno-haplotypes (Bartholomew and Ihle, 1991). In this context, it is important to emphasize that the majority of ERVs found in vertebrate genomes are of ancient nature, inactivated by mutation, and have reached genetic fixation in the host species they have evolved in. Therefore, they are extremely unlikely to have negative effects on their hosts except under very unusual circumstances.

    Although not applicable to the present study, where only ALS patients were tested, it is important to note that Nath’s group had previously reported HIV-positive ALS patients testing positive for HERV antibodies. In this context it is important to iterate that viruses can interfere with the same pathways and anatomical structures in the brain that are involved in neurologic diseases, such as Alzheimer’s, Parkinson’s, ALS, and other dementias. Support for this argument comes from antiretroviral agents reverting dementia in HIV patients (Zhou et al., 2013). Thus, taking this and the previous study, the striking observations made by Nath’s group warrant a detailed investigation into the possible genetic interactions between HERV and other retroviruses and their role in neurodegeneration as a consequence of their weakening the immune system due to age.

    We have learned a lesson from an evolutionarily conserved ERVWE1 or "syncytin" gene—a derivative of ERV insertion that is vital to the creation of human progeny. Its tight conservation across primate species in the same genomic locus speaks volumes for its functional significance. As we understand now that microbes are integral to our very survival by guiding and managing our immune system against foreign attack, it is time to rethink how the genomic remnants of microbial elements, with which we have co-evolved, interact with other exogenous agents in vivo.

    Although the study by Nath’s group is exciting and the investigators were cautious not to over-interpret that HERVs cause ALS, it is important to consider the evolutionary and philosophical parts of the debate on ERVs. Extrapolating from in vitro findings may not be meaningful simply because they may not jibe when compared with the in vivo functionality of ERV elements. The inheritance of ERV elements between hosts and their genetic and functional conservation in humans and mammals are some of the points that can shed more light on which direction to pursue to determine whether ERVS are pathogenic.


    . Human endogenous retroviruses in health and disease: a symbiotic perspective. J R Soc Med. 2004 Dec;97(12):560-5. PubMed.

    . Retroviral insertions 90 kilobases proximal to the Evi-1 myeloid transforming gene activate transcription from the normal promoter. Mol Cell Biol. 1991 Apr;11(4):1820-8. PubMed.

    . Viruses and neurodegeneration. Virol J. 2013 May 31;10:172. PubMed.

    View all comments by Nitin Saksena

Make a Comment

To make a comment you must login or register.

This paper appears in the following:


  1. Do Sleeper Viruses Awaken in ALS?