Is chronic fatigue syndrome caused by a rare brain infection of a common, normally benign virus? Bjorn Grinde(*) * National Institute of Public Health, P.O. Box 4404, Nydalen, 0403 Oslo, Norway Tel.: +47 22042420; fax: +47 22042447. Summary Chronic fatigue syndrome (CFS) is a disabling disease of unknown aetiology. A variety of factors have been suggested as possible causes. Although the symptoms and clinical findings are heterogeneous, the syndrome is sufficiently distinct, at least in relation to the more obvious cases, that a common explanation seems likely. In this paper, it is proposed that the disease is caused by a ubiquitous, but normally benign virus, e.g., one of the circoviruses. Circoviruses are chronically present in a majority of people, but are rarely tested for diagnostically. Normally these viruses do not penetrate the blood-brain barrier, but exceptions have been reported, and related viruses cause disease in the central nervous system of animals. The flu-like illness that often precedes the onset of CFS may either suppress immune function, causing an increased viremia, and/or lower the blood-brain barrier. In both cases the result may be that a virus already present in the blood enters the brain. It is well known that zoonotic viruses typically are more malignant than viruses with a long history of host-virus evolution. Similarly, a virus reaching an unfamiliar organ may cause particular problems. Introduction Chronic fatigue syndrome (CFS), also referred to as myalgic encephalomyelitis (ME), is characterised by a severe debilitating fatigue lasting for at least 6 months (reviewed in [1-4]). The worldwide prevalence may be as high as 0.5%. The pathological findings include immune system abnormalities such as chronic immune activation, dysfunction of the hypothalamic-pituitary-adrenal (HPA) axis, brain abnormalities, emotional stress, and evidence of infectious agents. The cause of CFS has been attributed to a variety of factors. Although an infection of the central nervous system (CNS) would seem likely, the lack of affirmative findings has led to a preference for other theories, including: psychological stress, disturbances of immunity, endocrine control, oxidative stress, cardiovascular homeostasis, and systemic infections. Not surprisingly, genetic constitution also appears to be important [5,6]. A majority of cases report that the onset of CFS was preceded by a 'flu-like' illness. A popular theory is that the infectious agent causing this illness triggers a chronic activation of the immune system, which subsequently leads to CFS. Thus, the condition may be caused by a virus in the absence of any actual infection of the CNS. Emotional stress may be relevant as it reduces the efficiency of the immune system, and because it leads to activation of the HPA-axis. In fact, some practitioners suggest that the condition is solely due to psychosomatic factors and thus should be labelled as a psychiatric disorder. Although mental problems are highly relevant in connection with fatigue, it seems likely that the extreme state ob- served in many CFS patients have a more distinct cause; particularly as many of the patients were highly active and well-adjusted prior to the onset of disease. With a condition such as CFS, it is inherently difficult to distinguish between cause and consequences. Yet, an actual infection of the CNS offers a parsimonious theory. The remaining pathology can convincingly be ascribed as secondary effects. For example, if the virus causes damage in the hypothalamus, an effect on the HPA-axis would be expected, as well as a subsequent impact on immune parameters. In order to support the present hypothesis one should ideally detect one (or a few) infectious agents in the spinal fluid in a substantial number of patients. So far this has not been possible. The present article suggests a theory that would explain the various observations relating to CFS, including the epidemiological data and the fact that no convincing agent has yet been described. Moreover, a strategy for testing the hypothesis is indicated. Evidence for CNS infection by a common virus A variety of microbes have been implicated in CFS, including Epstein-Barr virus, cytomegalovirus, enteroviruses, parvovirus B19, Coxiella burnetii and Chlamydia pneumoniae . Some of these agents are found in the spinal fluid, and could exert their effect there, while others are typically found elsewhere in the patients and might affect the CNS indirectly. Several outbreaks of CFS (or CFS-related disorders) have been described, and in some cases particular infectious agents, such as C. burnetii and enteroviruses, were assumed to be responsible [1,7,8]. Yet, the idea of a CNS infection causing CFS is not advocate in recent reviews . The problem is lack of consistent finding, and lack of obvious markers, or symptoms, of infection in the brain. The present hypothesis suggests that CFS is caused by an infection of the CNS. The lack of substantial proof indicates that the putative agent is more likely to be a virus than a bacteria. The hypothesis furthermore suggests that the culprit is commonly present in humans, but only rarely penetrates into the brain. Below are listed observations that are congruent with this theory, and thus may be considered as evidence in favour. 1. The disease is associated with a disturbance of brain function. It is well known that virus infections of the brain can cause various behavioural changes. 2. Several viruses that normally infect other organs, occasionally penetrate into the brain. At least two, hepatitis C virus [9,10] and Nipah virus , have been reported as the causative agents in chronic or disabling fati- gue of select patients. It is possible that dif- ferent viruses can cause fatigue, however, it is also conceivable that the actual cause in these patients was an unknown virus that crossed the blood-brain barrier along with the observed virus. Hepatitis C is rare in young people except for drug abusers, and Nipah virus is a rare zoonotic virus. These viruses are therefore unlikely to be responsi- ble for more than a few CFS cases. 3. If CFS had been a normal consequence of the infection of one or a few rare viruses, one would expect a different epidemiological pat- tern. That is, the cases should cluster along possible infectious pathways. Most cases do not. The epidemiology fits better with the notion that the event leading to the disease is an unexpected crossing of the blood-brain barrier of a virus present in most people. 4. Several outbreaks of CFS-like disease have been described . These outbreaks could be explained by the introduction of a particularly pathogenic strain of an otherwise common and mostly benign virus. It is interesting to note that most of the reported epidemics date back 50 years or more. It seems likely that several viral strains of common viruses, which previ- ously were endemic to particular regions, at some time started to spread worldwide. In the case of circoviruses, for example, the genetic variability within a local population is close to the variability observed worldwide [12,13]. Presumably, the genetic diversity was originally due to independent viral prolif- eration in different regions. Thus the present local variability is best explained by a recent worldwide mix of viral strains. 5. If present in the majority of the population, it implies that the causative virus is normally benign. However, even if the virus does not cause overt symptoms when circulating in the blood, it may very well do so upon entering an organ for which it is less adapted. The situa- tion would be a parallel to the observation that the more severe viral diseases (e.g., HIV, Ebola, SARS and bird-flu) are caused by zoonotic viruses. Viruses with a long term relationship with a host, or in this case cer- tain organs, generally develop a more benev- olent profile. 6. A number of viruses fit the description of high prevalence, continuous replication and nor- mally benign nature; including members of the Anellovirus genus of circoviruses (TTV, TTMV and TTMLV) and the polyomaviruses (BK and JC). Circoviruses and polyomaviruses are normally associated with other organs, yet can penetrate the brain and replicate there [14,15]. Polyomaviruses are known to cause clinical symptoms in the brain . Cir- coviruses have no know pathogenicity in humans, but animal circoviruses can infect the brain and cause disease, e.g., post- weaning multisystemic wasting syndrome in pigs, a condition with certain similarities to CFS . 7. CFS is typically preceded by an acute systemic infection, or by severe stress. These condi- tions may cause a virus present in the blood to enter the CNS, either by suppressing immune function, and thus increasing the viral titre, or by weakening the blood-brain barrier. It has been shown that immune sup- pression causes an increase in the serum con- centration of circoviruses . 8. Viruses affecting the CNS are reputedly diffi- cult to diagnose. Even in cases of encephalitis or meningitis where there is reasonable evi- dence to suggest a viral aetiology, positive findings are obtained only in a fraction of cases. The problem may be partly related to not testing for the right virus, but most likely it also reflects that the relevant viruses may cause clinical symptoms in the brain even if their replication is low, and thus less likely to be detected. In the case of CFS, spinal fluid is not routinely tested for any virus; and even in cases where standard viral diagnostics are attempted, it is unlikely to yield positive results. 9. Circoviruses or polyomaviruses have, to the author's knowledge, never been tested in connection with CFS. These viruses are tuned to keep a low profile towards the immune sys- tem. If present in the brain, they are likely to have a low copy number and not cause overt signs of infection. Moreover, the brain is less likely to initiate an obvious immunological response than are other organs . Yet, when carefully examined, some markers sug- gestive of infection are found in the spinal fluid of CFS patients . 10. Mothers of adolescents with CFS more often display related symptoms than do the fathers . This could be explained by the transfer of certain strains of virus during pregnancy or birth, which presumably happens in the case of circoviruses and polyomaviruses. 11. Treatment with the broad-spectrum antiviral agents interferon and Ribavirin have been reported to improve the condition of CFS patients . 12. Although full recovery from CFS is rare, improvement can be sufficient to suggest an average duration of 57 years . The per- sistent nature of the disease would be in line with the observation that viruses, such as cir- coviruses and polyomaviruses, typically are chronically replicating in the host. Yet, these viruses are partly controlled by the immune system, thus a reduction or clearance of virus in the brain is possible. Relevant viruses Circoviruses and polyomaviruses fit with the above observations, but the condition may equally well be caused by a range of other viruses. Picornaviridae, which includes enteroviruses, is a large family of viruses commonly present in the healthy population ; enteroviruses have been implicated in CFS , and several subtypes are known to occasionally infect the brain. Members of the herpes family and the parvovirus B19 have a similar profile [1,23]. Over the last few years, several novel viruses have been discovered. It seems likely that additional unknown viruses are commonly present in humans. In other words, the fact that no common viral cause has so far been described for CFS may very well be due to a lack of testing. Most of the observations listed are also compatible with alternative explanations as to the pathology of CFS. Taken together, however, the author considers the evidence sufficiently indicative to warrant further examination. Testing the hypothesis An obvious first possibility is to look for the specifically mentioned viruses in spinal fluid using a sensitive PCR. A blood sample alone have minimal value if, as hypothesized, one is dealing with a common virus. Ideally a blood sample taken at the same time should be available in order to evaluate whether a positive finding in the spinal fluid could reflect contamination from blood. Preferably one should analyse samples taken soon after the initiation of disease. It is conceivable that the virus cause damage, and is subsequently cleared from the brain. The damage may heal only slowly. It is, however, likely that the virus is continuously present. In the latter case, there should be hope of finding the virus regardless of when the sample is obtained. The next step, as to testing the hypothesis, would be to broaden the scope of virus search. Recent developments with microarrays allows for the screening of all known vertebrate-related viruses, as well as other microbes, simultaneously in clinical samples . The method relies on unspecific amplification of RNA/DNA and subsequent probing of the obtained nuclei acids onto microarrays containing single stranded DNA probes that capture viral sequences. In order to stand an even better chance of finding novel viruses of limited homology to known viruses, it is possible to run the amplified nucleic acids on a pyrosequencing machine. Pyrosequencing allows for up to 400 000 sequence fragments of some 300 basepairs to be obtained from one sample in a single operation. The obtained sequences would subsequently be examined by a computer to look for possible homologies with known viral sequences. Even viruses that have never been described are expected to have some genomic regions with homology to known viral sequences. Hopefully, it shall be possible to test the hypothesis in the near future. The difficult part appears to be obtaining relevant samples. This is partly because spinal punctuation is not carried out routinely in connection with CFS patients, partly because of the bureaucratic difficulties in obtain- ing older samples from the odd patient that at some time had spinal fluid tapped. Discussion The presence of certain virus(es) within the spinal fluid of CFS patients is not sufficient to prove causality. If, however, the relevant virus is rare or absent in samples from other patient groups, it would appear to be a likely trigger. A more direct proof would be to develop antiviral treatment that both remove the virus and improves the is sufficiently common for the pharmaceutical industry to invest in developing novel therapy, or possibly a vaccine. Even in the absence of treatment, it is of great interest to establish a causality. Besides the comfort for the patient of knowing what the problem is, a viral connection would offer improved diagnostic opportunities. References 1 Devanur LD, Kerr JR. Chronic fatigue syndrome. J Clin Virol 2006;37:139-50. 2 Prins JB, van der Meer JWM, Bleijenberg G. Chronic fatigue syndrome. Lancet 2006;367:346-55. 3 Sanders P, Korf J. Neuroaetiology of chronic fatigue syndrome: an overview. World J Biol Psychiatry 2007;8:1-7. 4 Wyller VB. The chronic fatigue syndrome - an update. Acta Neurol Scand Suppl 2007;187:7-14. 5 Crawley E, Smith GD. Is chronic fatigue syndrome (CFS/ME) heritable in children, and if so, why does it matter? Arch Dis Child 2007;92:1058-61. 6 Goertzel BN, Pennachin C, de Souza Coehlo L, Gurbaxani B, Maloney EM, Jones JF. Combinations of single nucleotide polymorphisms in neuroendocrine effector and receptor genes predict chronic fatigue syndrome. Pharmacogenom- ics 2006;7:475. 7 Chia JKS. The role of enterovirus in chronic fatigue syndrome. J Clin Pathol 2005;58:1126-32. 8 Jenkins R. Epidemiology: lessons from the past. Brit Med Bull 1991;47:952-65. 9 Tillmann HL. Hepatitis C virus infection and the brain. Metab Brain Dis 2004;19:351-6. 10 Forton DM, Thomas HC, Taylor-Robinson SD. Central ner- vous system involvement in hepatitis C virus infection. Metab Brain Dis 2004;19:383-91. 11 Ng BY, Lim CC, Yeoh A, Lee WL. Neuropsychiatric sequelae of Nipah virus encephalitis. J Neuropsychiat Clin Neurosci 2004;16:500-4. 12 Huang L-Y, Jonassen TØ, Hungnes O, Grinde B. High prevalence of TTV viremia (90%) and diverse viral geno- types in Norwegian blood donors. J Med Virol 2001;64:381-6. 13 Moen EM, Huang L-Y, Grinde B. Molecular epidemiology of TTV-like mini viruses (TLMV) in Norway. Arch Virol 2002;147:181-5. 14 Maggi F, Fornai C, Vatteroni ML, Siciliano G, Menichetti F, Tascini C, et al. Low prevalence of TT virus in the cerebrospinal fluid of viremic patients with central nervous system disorders. J Med Virol 2001;65:418-22. 15 Doerries K. Human polyomavirus JC and Bk persistent infection. Adv Exp Med Biol 2006;577:102-16. 16 Seeliger FA, Brugmann ML, Kruger L, Greiser-Wilke I, Verspohl J, Segales J, et al. Porcine circovirus type 2- associated cerebellar vasculitis in postweaning multisys- temic wasting syndrome (PMWS)-affected pigs. Vet Pathol 2007;44:621-34. 17 Moen EM, Sagedal S, Bjoro K, Degre M, Opstad PK, Grinde B. Effect of immune modulation on TT virus (TTV) and TTV- like-mini-virus (TLMV) viremia. J Med Virol 2003;70: 177-82. 18 Galea I, Bechmann I, Perry VH. What is immune privilege (not)? Trends Immunol 2007;28:12-8. 19 Natelson BH, Weaver SA, Tseng C-L, Ottenweller JE. Spinal fluid abnormalities in patients with chronic fatigue syn- drome. Clin Diagn Lab Immunol 2005;12:52-5. 20 Van de Putte EM, van Doornen LJ, Engelbert RH, Kuis W, Kimpen JL, Uiterwaal CS. Mirrored symptoms in mother and child with chronic fatigue syndrome. Pediatrics 2006;117: 2074-9. 21 Reyes M, Nisenbaum R, Hoaglin DC, Unger ER, Emmons C, Randall B, et al. Prevalence and incidence of chronic fatigue syndrome in Wichita, Kansas. Arch Intern Med 2003;163:1530-6. 22 Witso E, Palacios G, Cinek O, Stene LC, Grinde B, Janowicz D, et al. Natural circulation of human enteroviruses: high prevalence of human enterovirus A infections. J Clin Microbiol 2006;44:4095-100. 23 Kerr JR. Pathogenesis of parvovirus B19 infection: host gene variability, and possible means and effects of virus persistence. J Vet Med 2005;B52:335-9. 24 Palacios G, Quan PL, Jabado OJ, Conlan S, Hirschberg DL, Liu Y, et al. Panmicrobial oligonucleotide array for diag- nosis of infectious diseases. Emerg Infect Dis 2007;13:73-81. -------- (c) 2008 Elsevier / ScienceDirect B.V.