The Origin of Hepatitis C Virus
Peter Simmonds
Abstract The origin of hepatitis C virus (HCV) can be conceptualised at several
levels. Firstly, origins might refer to its dramatic spread throughout the Western
world and developing countries throughout the twentieth century. As a blood-
borne virus, this epidemic was fuelled by new parenteral transmission routes
associated with medical treatments, immunisation, blood transfusion and more
recently injecting drug use. At another level, however, origins might refer to the
immediate sources of HCV associated with its pandemic spread, now identified
as areas in Central and West sub-Saharan Africa and South and South East Asia
where genetically diverse variants of HCV appear to have circulated for hun-
dreds of years. Going back a final step to the actual source of HCV infection in
these endemic areas, non-human primates have been long suspected as harbour-
ing viruses related to HCV with potential cross-species transmission of variants
corresponding to the 7 main genotypes into humans. Although there is tempting
analogy between this and the clearly zoonotic origin of HIV-1 from chimpanzees
in Central Africa, no published evidence to date has been obtained for infection of
HCV-like viruses in either apes or Old World monkey species. Indeed, a radical re-
think of both the host range and host-specificity of hepaciviruses is now required
following the very recent findings of a non-primate hepacivirus (NPHV) in horses
Contents
1 HCV Genetic Diversity and Genotype Classification .......................................................... 2
2 The Recent Spread of HCV ................................................................................................. 5
3 Endemic Circulation of HCV .............................................................................................. 6
4 Origins of Human Infections and HCV Homologues
in Other Mammals ............................................................................................................... 8
5 Concluding Thoughts ........................................................................................................... 11
References .................................................................................................................................. 12
P. Simmonds (*)
Infection and Immunity Division, Roslin Institute, University of Edinburgh,
Easter Bush, Edinburgh EH25 9RG, UK
e-mail: peter.simmonds@ed.ac.uk
R. Bartenschlager (ed.), Hepatitis C Virus: From Molecular Virology 1
to Antiviral Therapy, Current Topics in Microbiology and Immunology 369,
DOI: 10.1007/978-3-642-27340-7_1, © Springer-Verlag Berlin Heidelberg 2013
and potentially in dogs. Further research on a much wider range of mammals is
needed to better understand the true genetic diversity of HCV-like viruses and
their host ranges in the search for the ultimate origin of HCV in humans.
This review is written at a highly significant time in evolutionary studies of
HCV and its origins. The discovery of closely related viruses to human HCV in
horses and possibly dogs termed non-primate hepacivirus (NPHV) (Burbelo
et al. 2012; Kapoor et al. 2011) throws an entirely new light on the species dis-
tribution of hepaciviruses and their host range. Despite the significance of these
very recent discoveries, however, in many ways it is a particularly difficult time
to write a review of HCV origins and evolutionary history. Frequent infection of
horses worldwide with a virus reasonably similar to HCV breaks a key assump-
tion of much previous research that the closest relative of HCV would be found in
non-human primates. In its place we now have total uncertainty; domestic horses
seem an incongruous host species and the suspicion must be that hepaciviruses are
much more widely distributed in other mammals. At present, however, we simply
do not know what these are. More importantly, we do not know whether viruses
more similar to human HCV than NPHV exist and what species these may infect.
Discovering a zoonotic source for the epidemic of HCV infection that has swept
through the human population in the last century would be a truly important step
in our understanding of host relationships, adaptation and pathogenicity.
This review of HCV origins therefore concentrates initially on the better character-
ised recent epidemic transmission of HCV in the twentieth century and the existence
of suspected source areas for infection in sub-Saharan Africa and South-East Area.
Some aspects of the much less well understood history of HCV before this recent
spread will be speculatively discussed, as will the existence of HCV-like viruses in
non-human species. Inevitably any comments made about the latter will, through fur-
ther research, be revealed as either hopelessly cautious or naively overstated in a very
short space of time, errors for which I apologise in advance. However, the findings
cannot be simply omitted from a review with this title and I hope that they spur a
greater interest in the wider group of hepaciviruses and whether the attributes of HCV
that make it such an important human pathogen (persistence and hepatotropism) are
shared with other members of the genus.
1 HCV Genetic Diversity and Genotype Classification
HCV is classified as the type member of the genus Hepacivirus within the virus
family Flaviviridae (Fig. 1) (Bukh 1995; Simmonds et al. 1993, 2005). Although
variants of HCV show substantial genetic diversity from each other, the 7 cur-
rently classified genotypes are all classified as one species under current ICTV
rules notwithstanding their considerably antigenic variability and geographical
differences in distribution (Simmonds et al. 2011). Until recently, the only other
virus classified as a hepacivirus was GBV-B, a virus recovered from a laboratory
The Origin of Hepatitis C Virus 3
Fig. 1 Phylogenetic tree of members of the family Flaviviridae showing its primary division
into four genera. The tree was based on comparison of conserved regions of the RNA polymer-
ase sequences (positions 7704–8550 numbered as in the HCV-1 genome, AF011751) from repre-
sentative variants within each genus and species. The unrooted phylogenetic tree was constructed
by neighbour-joining of (uncorrected) amino acid p-distances. A divergence of 0.1 (10 % amino
acid sequence divergence) is depicted on the scale bar. Variants variously described as GBV-A,
GBV-C and hepatitis G virus have been assigned to the new proposed genus, Pegivirus as
recently proposed (Stapleton et al. 2011)
housed tamarind [New World primate; (Simons et al. 1995; Muerhoff et al. 1995)].
Only the one isolate of GBV-B has been identified to date and its ultimate origin
(primate or non-primate) remains unclear.
Members of the Hepacivirus genus are distinct genetically and in genome organ-
isation from members of the recently assigned Pegivirus genus (Stapleton et al.
2011) (Fig. 1). This group comprises a number of non-pathogenic viruses infecting
humans apes (Adams et al. 1998; Birkenmeyer et al. 1998), non-human primates
(Simons et al. 1995) and more recently, bats (Epstein et al. 2010). The recent pro-
posal to re-designate these viruses as human, simian and bat pegiviruses (Stapleton
et al. 2011) was designed to dispel the confusion in their original nomenclature
(terms such as GB virus C and hepatitis G virus have both been applied to pegivi-
ruses infecting humans) and to differentiate these viruses clearly from GBV-B, a
member of the Hepacivirus genus.
HCV genotypes are substantially divergent in sequence from each other and fall
into 7 phylogenetic clades, designated as genotypes (Fig. 2). Within these, a variable
number of sub-groupings are apparent. HCV variants circulating in Western countries
have been designated as subtypes, of which 1a, 1b, 2a, 2b, 3a, 4a and 6a are the most
frequently identified. HCV subtypes are epidemiologically distinct, with differences
in risk group targeting and geographical distributions that reflect their recent epidemic
spread. As examples, genotype 3a (along with 1a) typically infects injecting drug users
in Northern Europe and 4a in most frequently found in the Middle East. Genotypes
Fig. 2 Evolutionary tree of NS5B sequences of HCV genotypes 1–7 (positions 8276–8615 as
numbered as in the H77 reference sequence). High diversity areas in sub-Saharan Africa and
South East Asia contain a large number of variants additional to subtypes such as 1a, 1b and 3a
found in Western countries, displaying an endemic pattern of diversity. The tree was constructed
by neighbour-joining using maximum composite likelihood distances as implemented in the
MEGA 4 program (Tamura et al. 2007). The scale bar depicts an evolutionary distance of 0.05
The Origin of Hepatitis C Virus
1b, 2a and 2b infections are in contrast most prevalent in older population groups
throughout Europe and Asia and are most frequently linked to past blood transfusions.
A distinct pattern of viral diversity is observed in areas such as sub-Saharan
Africa and South East Asia, where infections with individual genotype predomi-
nate over large geographical areas (such as genotype 1 in Central Africa, geno-
type 2 in West Africa and genotype 6 in South East Asia), within which there is
substantial genetic diversity. The pattern of diversity observed within HCV is thus
both the consequence of its very recent epidemic spread into new risk groups,
overlaid on top of the much older “endemic” circulation of HCV in sub-Saharan
Africa and South East Asia. These different ways to conceptualise “origins” of
HCV are discussed in the next two sections.
2 The Recent Spread of HCV
The discovery of HCV in 1989 (Choo et al. 1989) was a remarkable achievement
that heralded the use of molecular methods for virus aetiological studies refractory
to previously used virus isolation methods. The very active research programme
throughout the 1970s and 1980s that culminated in the discovery of HCV was pri-
marily driven by pressing concerns of clinicians and epidemiologists who increas-
ingly recognised chronic non-A, non-B hepatitis associated with blood transfusion
and therapy with plasma-derived blood products (Prince et al. 1974; Feinstone et
al. 1975; Alter et al. 1975). Since the development of effective diagnostic tests
for HCV, the full scale of the spread of HCV became rapidly apparent. It is cur-
rently thought that HCV chronically infects 170 million people worldwide, 3 %
of the world’s population and creates a huge disease burden from chronic progres-
sive liver disease (Pawlotsky 2003; Hoofnagle 2002; Seeff 2002). In addition to
recipients of blood transfusion and medical treatment with unsterilised needles,
diagnostic screening has identified the extensive spread of infection through nee-
dle-sharing drug abuse, an epidemic starting in the 1960s or earlier in Western
countries and the primary route of ongoing transmission of infection following
the introduction of effective blood donor screening and blood product inactivation
steps in the 1990s (Nelson et al. 2011).
Both the time of initial spread of HCV into Western countries and the popula-
tion dynamics of the epidemic can only be indirectly inferred. However, available
evidence is consistent with relatively recent dates for its worldwide spread although
it likely preceded the AIDS epidemic by some decades. A lack of samples available
for screening collected before the Second World War has prevented a direct demon-
stration of this hypothesis and reconstruction of the HCV epidemic has been largely
based on modelling evolutionary histories of currently circulating variants and by
identifying historical factors such as widespread use of blood transfusion and other
parenterally delivered treatments and vaccinations that facilitated HCV transmission.
In epidemiological terms, transmission of HCV through sexual contact or from
mother to child is inefficient and infrequent (Wasley and Alter 2000; Pradat and Trepo
2000; Thomas 2000). The restriction of HCV transmission through primarily paren-
teral routes therefore implicates medical treatment with unsterilised needles (including
large-scale vaccination programmes), blood transfusion and more recently inject-
ing drug use as routes as the principal means of HCV spread and a relatively recent
timescale (Drucker et al. 2001). None of these risk factors were common before the
Second World War and supports the current model for the spread of genotypes 1b and
type 2 subtypes from the 1940s–1950s, overlaid by more recent transmission among
IDUs from the 1960s onwards (Pybus et al. 2001; Cochrane et al. 2002).
This scenario is strongly supported by genetic analysis of HCV genotypes and
subtypes most frequently detected among IDUs and those infected previously
through medical treatment. A recent large-scale coalescent analysis of 1a and 1b
subtypes demonstrated relatively small and constant population sizes for both sub-
types from the early twentieth century followed by an exponential period of popu-
lation growth between the 1940s and 1980s in the USA (Magiorkinis et al. 2009).
The slowing of population growth thereafter is additionally consistent with reduc-
tions in blood transfusion risk through HIV-1 followed by HCV screening and the
expansion of needle exchange programmes that have led to significant falls in HCV
incidence among IDUs. Emphasising the global nature of the recent spread of HCV,
parallel phylogeographic analyses have revealed similar demographic histories
of these subtypes in Brazil, Indonesia and Japan (Nakano et al. 2004). A detailed
analysis of reconstructed population sizes of HCV and the emergence of parenteral
routes of exposure in Japan Egypt and the USA further strengthens these conclu-
sions (Mizokami et al. 2006), including the close links between HCV emergence
and parenteral antischistosomal therapy in Japan and subsequently in Egypt. In the
latter, the extremely high population prevalence of HCV is dominated by genotype
4a, whose spread can be reconstructed to have occurred between the 1930s–1950s,
a period that coincides with targeted extensive antischistosomal injection cam-
paigns using largely unsterilised injection equipment (Pybus et al. 2003).
Collectively, these and several further combined phylogenetic and epidemiologi-
cal reconstructions provide a convincing narrative for the spread of HCV worldwide.
Although earlier by some decades, its spread is paralleled by the explosive world-
wide spread of HIV-1 from Africa from the 1980s onwards leading to the current
AIDS pandemic. In one sense, the question of the origins of HCV has likely already
been answered. However, where HCV was before then and what factors led to its
emergence are much less well understood and are discussed in the next section.
3 Endemic Circulation of HCV
While the epidemic spread of HCV is associated with specific, very prevalent
subtypes such as 1a, 1b, 3a and 4a, these represent a small part of the diversity
existing with HCV. In sub-Saharan Africa and South East Asia, a quite distinct pat-
tern of genetic diversity exists (Fig. 2). Discounting recent introductions, infec-
tions in large, geographically contiguous areas among several countries in Central
Africa or the South East Asian peninsula are dominated by individual genotypes
(genotypes 1 and 6 respectively in these examples). Individual variants within
these genotypes show striking genetic diversity from each other matching the
genetic divergence observed between subtypes such as 1a and 1b found in Western
countries. For example, sequence characterisation of genotype 2 variants infecting
23 blood donors in Ghana (West Africa) revealed the presence of 20 highly diverse
variants that would merit their assignment as new subtypes, as divergent from
each other as 2a is from 2b (≈25 % nucleotide sequence divergence) (Candotti
et al. 2003). Although far from fully mapped systematically, infections throughout
Western Africa are predominantly by genotype 2 (Candotti et al. 2003; Jeannel et
al. 1998; Mellor et al. 1995; WansbroughJones et al. 1998; Ruggieri et al. 1996),
while those in Central Africa, such as the Congo, Cameroon and Gabon are by
genotypes 1 and 4 (Mellor et al. 1995; Bukh et al. 1993; Fretz et al. 1995; Stuyver
et al. 1993; Menendez et al. 1999; Xu et al. 1994; Ndjomou et al. 2003; Li et al.
2009, 2012). Genotype 3 and 6 are typically found in the Indian sub-continent
and South East Asia (Mellor et al. 1995; Tokita et al. 1994, 1994, 1995; Lu et al.
2008). It is further suspected, although with very limited data that genotypes 5 and
7 are concentrated in Central/Southern Africa.
The extensive genetic heterogeneity of HCV in these regions has been described
as an “endemic” pattern of diversity and is consistent with its long-term pres-
ence and diversification in these populations. As such, it is currently hypothesised
that they represent source areas fuelling the worldwide spread of HCV in the last
100–200 years. Indeed, the distinct subtypes that have been described in Western
countries such as 1a, 1b and 3a might simply represent the explosive expansion
of certain variants within new risk groups for infection. Although we do not know
and may never be able to reconstruct their ultimate origins and initial transmis-
sion pathways, 1a, 1b, 3a and others may simply happen to be the most success-
ful of variants that entered previously unexposed and highly susceptible individuals
exposed parenterally. In the same way that HIV-1 subtype B entered and spread
within male homosexuals and IDUs in the USA and subsequently in Europe (Gao
et al. 1999), our current collection of classified subtype might similarly represent
founder viruses that were among the first to spread epidemically in the last century
in Western countries where HCV was first genetically characterised.
Supporting this model are the more recently described examples of introduc-
tions and varying degrees of local spread of a range of otherwise undescribed
“subtypes” of HCV. As examples, substantial diversity and restricted distribu-
tions of genotype 2 variants infecting have been described in Europe (Thomas
et al. 2007), Indonesia (Utama et al. 2010) and throughout the Caribbean (Sulbaran
et al. 2010; Martial et al. 2004), the latter examples in particular perhaps represent-
ing the shipment of infected West Africans through the slave trade in the eighteenth
and nineteenth centuries (Markov et al. 2009). The more recent spread of genotype
4 variants within Cameron and Egypt through medical treatment (Pybus et al. 2003;
Pepin and Labbe 2012), into Mediterranean countries and the recent rapid spread of
genotype 4 variants among IDUs in Southern Europe (Nicot et al. 2005; de Bruijne
et al. 2009) provide further examples of this model (Ndjomou et al. 2003).
What remains unexplained is the nature of the “endemic” circulation of HCV
in these implicated source areas and in particular the transmission routes that have
sustained long-term circulation of HCV in what have been until recently relatively
frequently highly isolated human communities. As discussed, transmission by either
sexual contact or from mother to child is inefficient at least in areas where it has been
studied (Wasley and Alter 2000; Pradat and Trepo 2000; Thomas 2000) and various
factors that may enhance transmission have been proposed. Examples include sexu-
ally transmitted infections (STIs), circumcision, excision and scarification practices
(Shepard et al. 2005) which at least in Central Africa show associations with HCV
infection and more remote possibilities such as mosquito or other arthropod vectors
(Pybus et al. 2007). These various hypotheses are yet to be resolved.
The time depth of “endemic” circulation of HCV remains similarly uncertain.
Molecular evolutionary reconstructions of the recent spread of HCV have pro-
duced robust and reproducible estimates of its substitution rate (see previous sec-
tion). Substitution rates extrapolated to the much larger sequence distances that
exist within genotypes (such as between subtypes 1a and 1b) have been used to
provide some kind of estimate of the minimum period over which the observed
“endemic” diversity developed (Pybus et al. 2001; Markov et al. 2009; Smith
et al. 1997; Pybus et al. 2009). Reconstructed dates for the common ancestor of
different genotypes vary but are estimated to be several hundred years ago for gen-
otype 2 and even longer for genotype 6. In the opinion of the author of this review,
such estimates should be treated with extreme caution and minimum estimates at
best. Extrapolating substitution rates measured over short observation intervals to
the much longer periods of subtype and genotype diversification makes assump-
tions about the evolutionary process that are not self-evidently justified. Factors
such as extreme rate variation between sites, large-scale RNA secondary structure,
greater selective constraints and fitness optimisation of viruses association with
large population sizes may create substantial underestimates of the real period of
virus diversification [reviewed in (Sharp and Simmonds 2011)]. While it is beyond
the scope of the current article to discuss this in detail, what can be said is that the
subtype diversification in HCV and thus the likely period of endemic circulation
in sub-Saharan Africa and Southern Asia is prolonged and likely long before long
distance travel and interactions with colonial powers. These genotypes are there-
fore likely to be truly indigenous to areas where they currently endemically cir-
culate. This takes us a step further back to the question of the ultimate source of
HCV. This much more speculative area will be reviewed in the next section.
4 Origins of Human Infections and HCV Homologues
in Other Mammals
A compelling scenario which has driven much research endeavour in the last
decade and a half is the hypothesis for a non-human primate source for HCV
infections in humans. The theory makes epidemiological sense in that high diver-
sity areas of endemic circulation in humans are those where human, ape and Old
World monkey populations overlap. Before long range travel and the means for
wider dissemination, human infections acquired zoonotically from non-human
primates may have remained geographically focussed and thus account for the
specific association of each of the genotypes in defined areas of sub-Saharan
Africa and Southern Asia. The idea of a non-human primate source for humans
is additionally consistent with the observation of its poor transmissibility between
humans, largely confined to parenteral routes and a reflection perhaps of its lack of
host adaptation as might also be its severe, immune-mediated liver pathology.
This model is, of course, also driven by the tempting analogy with the origin
of HIV-1, which similarly exploded worldwide out of Central Africa in the twen-
tieth century through infections directly or indirectly from chimpanzees (Gao et
al. 1999). As might be imagined for HCV, HIV-1 infections acquired through con-
tact with Central African chimpanzees (Pan troglodytes) may have been occurring
for centuries or millennia, but only in the last 50–70 years were demographic and
societal changes suitable for its wider pandemic spread. Important differences from
the HIV-1 model of origins would be the earlier spread of HCV worldwide and
the existence of multiple potential source areas and possibly different primate spe-
cies. These would be necessary to account for the distinct endemic distributions
of HCV genotypes in different parts of sub-Saharan Africa and also South East
Asia. Finally, into this model would come GBV-B, which might perhaps represent
a much more divergent homologue of HCV in a New World primate species.
Despite the elegance, plausibility and potential medical relevance of the
primate origin hypothesis, the fundamental problem has always been that HCV
or homologues cannot be found in ape or monkey species, at least to the author’s
knowledge. Extensive screening programmes both published (Makuwa et al.
2003, 2006) and unpublished have failed to document either seropositivity or viral
sequences in literally hundreds or thousands of plasma samples collected from dif-
ferent ape and monkey species. As a possibly related problem, GBV-B has to date
never been recovered nor serological evidence for past infections obtained from any
tamarind or other New World primate among wild populations in South America.
Without an obvious primate source for infection and the genetic evidence for
circulation of HCV in what would have been largely isolated human populations
in distinct parts of the world for centuries or more likely millennia, studies of the
ultimate origins of HCV have reached something of a frustrating impasse. As with
many other virus discoveries, however, its resolution is likely to be considerably
stranger than could have been imagined even as recently as last year. By pure ser-
endipity, attempts by Kapoor and colleagues to identify viral causes of respiratory
disease in dog held in kennels by deep sequencing revealed the existence of an
RNA virus extraordinarily similar to HCV (Fig. 3) but with suspected biologi-
cal and epidemiological properties quite different from what had been previously
described for both HCV and GBV-B (Kapoor et al. 2011).
The virus, initially termed canine hepacivirus (CHV) showed approximately
50 % nucleotide sequence divergence from HCV. Data presented in that study
demonstrated high viral loads in respiratory samples and an implied respiratory
route of transmission and association with respiratory disease, none of which have
been observed in HCV (or GBV-B) infections. Infections were found in dogs from
Fig. 3 Amino acid sequence divergence scan of members of the Hepacivirus genus, with
genome diagram drawn to scale underneath plot. NPHV is more similar to HCV throughout the
genome (red line) than GBV-B (dark green line). However, NPHV/HCV divergence is substan-
tially greater than between genotypes (dark and light blue lines respectively). This figure has
been adapted from Fig. 2 in (Kapoor et al. 2011). For details related to the HCV polyprotein and
the cleavage products see chapter “Hepatitis C virus Proteins From Structure to Function” by
Moradpour and Penin, this volume
different regions of the USA but partial genome characterisation demonstrated a
virtual absence of genetic diversity that would be expected for an RNA virus like
HCV. Whether the virus spread systemically or persisted was not demonstrated
although imaging of viral RNA in liver by in situ hybridisation was presented.
More recently, further studies of the host range of hepaciviruses in a range of
mammalian species were performed by the same group using a serological assay
for antibodies to a peptide expressed from the NS3 region of the CHV genome
(Burbelo et al. 2012). This produced further unexpected findings. From the 80
dogs, 81 deer, 84 cows, 103 horses and 14 rabbits screened, only horses showed
frequent seropositivity (35 %) with one weak positive sample from a cow while,
remarkably, all 80 dogs were seronegative. Of the 103 horse samples, 8 were PCR-
positive (all seropositive) and from each of these near-complete genome sequences
were obtained. Sequences showed moderate sequence diversity from each other
(6.4–17.2 % nucleotide sequence divergence) with the CHV sequence grouping
with horse-derived variants. As viruses similar to CHV were frequently found in
horses, the investigators coined the name NPHV to describe this group.
The diversity of NPHV variants was somewhere between inter-subtype and within
subtype divergence of HCV, certainly not the equivalent of HCV genotypes (Fig. 3).
The high degree of amino acid sequence conservation did, however, contrast mark-
edly with the degree of sequence variability at synonymous (non-coding) sites in the
genome. The extraordinarily low ratio between synonymous to non-synonymous
substitutions (0.03–0.06) indicates however that its evolution has been more severely
constrained and/or less subject to positive selection pressures than HCV. These
low sequence distances are therefore not necessarily an indication of their recent
divergence.
There was no information available on the clinical features of infection with
NPHV in horses. To address this we have recently surveyed horses in Scotland by
PCR and identified 3 viraemic horses from 136 screened (Lyons et al. 2012). Using
veterinary records and further sampling, these have been evaluated for evidence of
hepatitis or other systemic disease manifestations. Positive horses were originally
referred for reasons such as lameness, foot abscess or respiratory infections with no
evidence of the ill-health that might be associated with severe systemic infections.
Although most liver indices were in the normal range, gamma glutamyl transferase
(GGT) levels, a sensitive marker of liver inflammation were marginally or signifi-
cantly elevated along with elevation in bile acids, perhaps providing some tentative
evidence for an aetiological role of NPHV in hepatitis. Repeated sampling from one
of the study horses demonstrated persistence over at least a 6-month period and viral
loads comparable to those observed in HCV infections (7 × 10 −5 × 10 RNA
copies/ml). Respiratory samples and peripheral blood mononuclear cells from the
infected horse have proven uniformly negative although no opportunity to perform a
liver biopsy of the horse has yet presented itself. Overall, these more recent findings
provide some reassurance that hepacivirus infections in horses are both persistent and
potentially associated with mild liver disease rather than the respiratory disease and
viral secretion found originally in dogs. However, large-scale PCR-based screening of
other mammalian species using primers conserved between NPHV and HCV failed to
detect hepaciviruses in dogs (nearly 200 screened), cats, pigs and rodents (Lyons et al.
2012), very much as found in the previous serology-based study (Burbelo et al. 2012).
This, to date, represents current published knowledge of non-human hepacivi-
ruses, a series of findings that present several conflicting interpretations and dif-
ficulties. This author believes that, despite the negative results from screening so
far, domestic horses are most unlikely to be the only mammalian species (other
than human or tamarins) infected with hepaciviruses and there is clearly much to
be learned in short term from more extensive screening.
5 Concluding Thoughts
Our understanding of the ultimate origins of HCV infection in humans will doubt-
less be hugely enhanced once proper mammalian screening for other hepaciviruses
has been performed and the genetic diversity and, more importantly, the specific-
ity of different hepaciviruses to individual host species is more clearly established.
From such studies, it may well turn out that hepaciviruses are highly catholic in
their host range perhaps capable of jumping between horses and dogs as suggested
by the published screening data (Burbelo et al. 2012; Kapoor et al. 2011) and per-
haps all species in-between. An ability of hepaciviruses to jump species is consistent
with the observation that the NPHV protease is able to cleave human MAVS and
TRIF (Parera et al. 2012); this ability to prevent interferon signalling is essential for
HCV replication (Foy et al. 2005) and may therefore function across species bar-
riers and potentially favour zoonotic transmission. A wide mammalian host range
is also characteristic of vector-borne flaviviruses and pestiviruses, the latter at least
within ruminant species. In this scenario, HCV infections in humans may well have
a zoonotic origin consistent with its relatively recent emergence (at least in Western
countries). While being still relatively poorly adapted for infecting its new host, this
may further account for its peculiar, inefficient transmission routes.
Alternatively, it may be that each hepacivirus species is uniquely adapted to
one target species, HCV in humans, NPHV in horses and perhaps further hepa-
civiruses in other mammalian species waiting to be discovered. The ability of
HCV to persist lifelong in humans, an attribute that greatly enhances its transmis-
sibility and evidence for subtle virus/host interactions such as the enhancing role
of human micro RNA, miR-122 expressed in liver on virus replication (Jopling
et al. 2005) certainly hints at long-term virus/host co-adaptation. HCV may always
have infected humans throughout their evolution and it is only through greater life
expectancy, scope for epidemic transmission and better surveillance and under-
standing of causes of hepatitis that it has come to current medical attention. In that
sense, HCV does not have an “origin”, it is just one of those viruses like herpes-
viruses that have always infected humans and before them hominoids, proto-apes
and potentially right back to the ancestor of mammals themselves.
Future research will be truly important in resolving these two diametrically
opposed possibilities.
Source: http://www.roslin.ed.ac.uk/assets/profile-pages/peter-simmonds/the-origin-of-hcv.pdf
Peter Simmonds
Abstract The origin of hepatitis C virus (HCV) can be conceptualised at several
levels. Firstly, origins might refer to its dramatic spread throughout the Western
world and developing countries throughout the twentieth century. As a blood-
borne virus, this epidemic was fuelled by new parenteral transmission routes
associated with medical treatments, immunisation, blood transfusion and more
recently injecting drug use. At another level, however, origins might refer to the
immediate sources of HCV associated with its pandemic spread, now identified
as areas in Central and West sub-Saharan Africa and South and South East Asia
where genetically diverse variants of HCV appear to have circulated for hun-
dreds of years. Going back a final step to the actual source of HCV infection in
these endemic areas, non-human primates have been long suspected as harbour-
ing viruses related to HCV with potential cross-species transmission of variants
corresponding to the 7 main genotypes into humans. Although there is tempting
analogy between this and the clearly zoonotic origin of HIV-1 from chimpanzees
in Central Africa, no published evidence to date has been obtained for infection of
HCV-like viruses in either apes or Old World monkey species. Indeed, a radical re-
think of both the host range and host-specificity of hepaciviruses is now required
following the very recent findings of a non-primate hepacivirus (NPHV) in horses
Contents
1 HCV Genetic Diversity and Genotype Classification .......................................................... 2
2 The Recent Spread of HCV ................................................................................................. 5
3 Endemic Circulation of HCV .............................................................................................. 6
4 Origins of Human Infections and HCV Homologues
in Other Mammals ............................................................................................................... 8
5 Concluding Thoughts ........................................................................................................... 11
References .................................................................................................................................. 12
P. Simmonds (*)
Infection and Immunity Division, Roslin Institute, University of Edinburgh,
Easter Bush, Edinburgh EH25 9RG, UK
e-mail: peter.simmonds@ed.ac.uk
R. Bartenschlager (ed.), Hepatitis C Virus: From Molecular Virology 1
to Antiviral Therapy, Current Topics in Microbiology and Immunology 369,
DOI: 10.1007/978-3-642-27340-7_1, © Springer-Verlag Berlin Heidelberg 2013
and potentially in dogs. Further research on a much wider range of mammals is
needed to better understand the true genetic diversity of HCV-like viruses and
their host ranges in the search for the ultimate origin of HCV in humans.
This review is written at a highly significant time in evolutionary studies of
HCV and its origins. The discovery of closely related viruses to human HCV in
horses and possibly dogs termed non-primate hepacivirus (NPHV) (Burbelo
et al. 2012; Kapoor et al. 2011) throws an entirely new light on the species dis-
tribution of hepaciviruses and their host range. Despite the significance of these
very recent discoveries, however, in many ways it is a particularly difficult time
to write a review of HCV origins and evolutionary history. Frequent infection of
horses worldwide with a virus reasonably similar to HCV breaks a key assump-
tion of much previous research that the closest relative of HCV would be found in
non-human primates. In its place we now have total uncertainty; domestic horses
seem an incongruous host species and the suspicion must be that hepaciviruses are
much more widely distributed in other mammals. At present, however, we simply
do not know what these are. More importantly, we do not know whether viruses
more similar to human HCV than NPHV exist and what species these may infect.
Discovering a zoonotic source for the epidemic of HCV infection that has swept
through the human population in the last century would be a truly important step
in our understanding of host relationships, adaptation and pathogenicity.
This review of HCV origins therefore concentrates initially on the better character-
ised recent epidemic transmission of HCV in the twentieth century and the existence
of suspected source areas for infection in sub-Saharan Africa and South-East Area.
Some aspects of the much less well understood history of HCV before this recent
spread will be speculatively discussed, as will the existence of HCV-like viruses in
non-human species. Inevitably any comments made about the latter will, through fur-
ther research, be revealed as either hopelessly cautious or naively overstated in a very
short space of time, errors for which I apologise in advance. However, the findings
cannot be simply omitted from a review with this title and I hope that they spur a
greater interest in the wider group of hepaciviruses and whether the attributes of HCV
that make it such an important human pathogen (persistence and hepatotropism) are
shared with other members of the genus.
1 HCV Genetic Diversity and Genotype Classification
HCV is classified as the type member of the genus Hepacivirus within the virus
family Flaviviridae (Fig. 1) (Bukh 1995; Simmonds et al. 1993, 2005). Although
variants of HCV show substantial genetic diversity from each other, the 7 cur-
rently classified genotypes are all classified as one species under current ICTV
rules notwithstanding their considerably antigenic variability and geographical
differences in distribution (Simmonds et al. 2011). Until recently, the only other
virus classified as a hepacivirus was GBV-B, a virus recovered from a laboratory
The Origin of Hepatitis C Virus 3
Fig. 1 Phylogenetic tree of members of the family Flaviviridae showing its primary division
into four genera. The tree was based on comparison of conserved regions of the RNA polymer-
ase sequences (positions 7704–8550 numbered as in the HCV-1 genome, AF011751) from repre-
sentative variants within each genus and species. The unrooted phylogenetic tree was constructed
by neighbour-joining of (uncorrected) amino acid p-distances. A divergence of 0.1 (10 % amino
acid sequence divergence) is depicted on the scale bar. Variants variously described as GBV-A,
GBV-C and hepatitis G virus have been assigned to the new proposed genus, Pegivirus as
recently proposed (Stapleton et al. 2011)
housed tamarind [New World primate; (Simons et al. 1995; Muerhoff et al. 1995)].
Only the one isolate of GBV-B has been identified to date and its ultimate origin
(primate or non-primate) remains unclear.
Members of the Hepacivirus genus are distinct genetically and in genome organ-
isation from members of the recently assigned Pegivirus genus (Stapleton et al.
2011) (Fig. 1). This group comprises a number of non-pathogenic viruses infecting
humans apes (Adams et al. 1998; Birkenmeyer et al. 1998), non-human primates
(Simons et al. 1995) and more recently, bats (Epstein et al. 2010). The recent pro-
posal to re-designate these viruses as human, simian and bat pegiviruses (Stapleton
et al. 2011) was designed to dispel the confusion in their original nomenclature
(terms such as GB virus C and hepatitis G virus have both been applied to pegivi-
ruses infecting humans) and to differentiate these viruses clearly from GBV-B, a
member of the Hepacivirus genus.
HCV genotypes are substantially divergent in sequence from each other and fall
into 7 phylogenetic clades, designated as genotypes (Fig. 2). Within these, a variable
number of sub-groupings are apparent. HCV variants circulating in Western countries
have been designated as subtypes, of which 1a, 1b, 2a, 2b, 3a, 4a and 6a are the most
frequently identified. HCV subtypes are epidemiologically distinct, with differences
in risk group targeting and geographical distributions that reflect their recent epidemic
spread. As examples, genotype 3a (along with 1a) typically infects injecting drug users
in Northern Europe and 4a in most frequently found in the Middle East. Genotypes
Fig. 2 Evolutionary tree of NS5B sequences of HCV genotypes 1–7 (positions 8276–8615 as
numbered as in the H77 reference sequence). High diversity areas in sub-Saharan Africa and
South East Asia contain a large number of variants additional to subtypes such as 1a, 1b and 3a
found in Western countries, displaying an endemic pattern of diversity. The tree was constructed
by neighbour-joining using maximum composite likelihood distances as implemented in the
MEGA 4 program (Tamura et al. 2007). The scale bar depicts an evolutionary distance of 0.05
The Origin of Hepatitis C Virus
1b, 2a and 2b infections are in contrast most prevalent in older population groups
throughout Europe and Asia and are most frequently linked to past blood transfusions.
A distinct pattern of viral diversity is observed in areas such as sub-Saharan
Africa and South East Asia, where infections with individual genotype predomi-
nate over large geographical areas (such as genotype 1 in Central Africa, geno-
type 2 in West Africa and genotype 6 in South East Asia), within which there is
substantial genetic diversity. The pattern of diversity observed within HCV is thus
both the consequence of its very recent epidemic spread into new risk groups,
overlaid on top of the much older “endemic” circulation of HCV in sub-Saharan
Africa and South East Asia. These different ways to conceptualise “origins” of
HCV are discussed in the next two sections.
2 The Recent Spread of HCV
The discovery of HCV in 1989 (Choo et al. 1989) was a remarkable achievement
that heralded the use of molecular methods for virus aetiological studies refractory
to previously used virus isolation methods. The very active research programme
throughout the 1970s and 1980s that culminated in the discovery of HCV was pri-
marily driven by pressing concerns of clinicians and epidemiologists who increas-
ingly recognised chronic non-A, non-B hepatitis associated with blood transfusion
and therapy with plasma-derived blood products (Prince et al. 1974; Feinstone et
al. 1975; Alter et al. 1975). Since the development of effective diagnostic tests
for HCV, the full scale of the spread of HCV became rapidly apparent. It is cur-
rently thought that HCV chronically infects 170 million people worldwide, 3 %
of the world’s population and creates a huge disease burden from chronic progres-
sive liver disease (Pawlotsky 2003; Hoofnagle 2002; Seeff 2002). In addition to
recipients of blood transfusion and medical treatment with unsterilised needles,
diagnostic screening has identified the extensive spread of infection through nee-
dle-sharing drug abuse, an epidemic starting in the 1960s or earlier in Western
countries and the primary route of ongoing transmission of infection following
the introduction of effective blood donor screening and blood product inactivation
steps in the 1990s (Nelson et al. 2011).
Both the time of initial spread of HCV into Western countries and the popula-
tion dynamics of the epidemic can only be indirectly inferred. However, available
evidence is consistent with relatively recent dates for its worldwide spread although
it likely preceded the AIDS epidemic by some decades. A lack of samples available
for screening collected before the Second World War has prevented a direct demon-
stration of this hypothesis and reconstruction of the HCV epidemic has been largely
based on modelling evolutionary histories of currently circulating variants and by
identifying historical factors such as widespread use of blood transfusion and other
parenterally delivered treatments and vaccinations that facilitated HCV transmission.
In epidemiological terms, transmission of HCV through sexual contact or from
mother to child is inefficient and infrequent (Wasley and Alter 2000; Pradat and Trepo
2000; Thomas 2000). The restriction of HCV transmission through primarily paren-
teral routes therefore implicates medical treatment with unsterilised needles (including
large-scale vaccination programmes), blood transfusion and more recently inject-
ing drug use as routes as the principal means of HCV spread and a relatively recent
timescale (Drucker et al. 2001). None of these risk factors were common before the
Second World War and supports the current model for the spread of genotypes 1b and
type 2 subtypes from the 1940s–1950s, overlaid by more recent transmission among
IDUs from the 1960s onwards (Pybus et al. 2001; Cochrane et al. 2002).
This scenario is strongly supported by genetic analysis of HCV genotypes and
subtypes most frequently detected among IDUs and those infected previously
through medical treatment. A recent large-scale coalescent analysis of 1a and 1b
subtypes demonstrated relatively small and constant population sizes for both sub-
types from the early twentieth century followed by an exponential period of popu-
lation growth between the 1940s and 1980s in the USA (Magiorkinis et al. 2009).
The slowing of population growth thereafter is additionally consistent with reduc-
tions in blood transfusion risk through HIV-1 followed by HCV screening and the
expansion of needle exchange programmes that have led to significant falls in HCV
incidence among IDUs. Emphasising the global nature of the recent spread of HCV,
parallel phylogeographic analyses have revealed similar demographic histories
of these subtypes in Brazil, Indonesia and Japan (Nakano et al. 2004). A detailed
analysis of reconstructed population sizes of HCV and the emergence of parenteral
routes of exposure in Japan Egypt and the USA further strengthens these conclu-
sions (Mizokami et al. 2006), including the close links between HCV emergence
and parenteral antischistosomal therapy in Japan and subsequently in Egypt. In the
latter, the extremely high population prevalence of HCV is dominated by genotype
4a, whose spread can be reconstructed to have occurred between the 1930s–1950s,
a period that coincides with targeted extensive antischistosomal injection cam-
paigns using largely unsterilised injection equipment (Pybus et al. 2003).
Collectively, these and several further combined phylogenetic and epidemiologi-
cal reconstructions provide a convincing narrative for the spread of HCV worldwide.
Although earlier by some decades, its spread is paralleled by the explosive world-
wide spread of HIV-1 from Africa from the 1980s onwards leading to the current
AIDS pandemic. In one sense, the question of the origins of HCV has likely already
been answered. However, where HCV was before then and what factors led to its
emergence are much less well understood and are discussed in the next section.
3 Endemic Circulation of HCV
While the epidemic spread of HCV is associated with specific, very prevalent
subtypes such as 1a, 1b, 3a and 4a, these represent a small part of the diversity
existing with HCV. In sub-Saharan Africa and South East Asia, a quite distinct pat-
tern of genetic diversity exists (Fig. 2). Discounting recent introductions, infec-
tions in large, geographically contiguous areas among several countries in Central
Africa or the South East Asian peninsula are dominated by individual genotypes
(genotypes 1 and 6 respectively in these examples). Individual variants within
these genotypes show striking genetic diversity from each other matching the
genetic divergence observed between subtypes such as 1a and 1b found in Western
countries. For example, sequence characterisation of genotype 2 variants infecting
23 blood donors in Ghana (West Africa) revealed the presence of 20 highly diverse
variants that would merit their assignment as new subtypes, as divergent from
each other as 2a is from 2b (≈25 % nucleotide sequence divergence) (Candotti
et al. 2003). Although far from fully mapped systematically, infections throughout
Western Africa are predominantly by genotype 2 (Candotti et al. 2003; Jeannel et
al. 1998; Mellor et al. 1995; WansbroughJones et al. 1998; Ruggieri et al. 1996),
while those in Central Africa, such as the Congo, Cameroon and Gabon are by
genotypes 1 and 4 (Mellor et al. 1995; Bukh et al. 1993; Fretz et al. 1995; Stuyver
et al. 1993; Menendez et al. 1999; Xu et al. 1994; Ndjomou et al. 2003; Li et al.
2009, 2012). Genotype 3 and 6 are typically found in the Indian sub-continent
and South East Asia (Mellor et al. 1995; Tokita et al. 1994, 1994, 1995; Lu et al.
2008). It is further suspected, although with very limited data that genotypes 5 and
7 are concentrated in Central/Southern Africa.
The extensive genetic heterogeneity of HCV in these regions has been described
as an “endemic” pattern of diversity and is consistent with its long-term pres-
ence and diversification in these populations. As such, it is currently hypothesised
that they represent source areas fuelling the worldwide spread of HCV in the last
100–200 years. Indeed, the distinct subtypes that have been described in Western
countries such as 1a, 1b and 3a might simply represent the explosive expansion
of certain variants within new risk groups for infection. Although we do not know
and may never be able to reconstruct their ultimate origins and initial transmis-
sion pathways, 1a, 1b, 3a and others may simply happen to be the most success-
ful of variants that entered previously unexposed and highly susceptible individuals
exposed parenterally. In the same way that HIV-1 subtype B entered and spread
within male homosexuals and IDUs in the USA and subsequently in Europe (Gao
et al. 1999), our current collection of classified subtype might similarly represent
founder viruses that were among the first to spread epidemically in the last century
in Western countries where HCV was first genetically characterised.
Supporting this model are the more recently described examples of introduc-
tions and varying degrees of local spread of a range of otherwise undescribed
“subtypes” of HCV. As examples, substantial diversity and restricted distribu-
tions of genotype 2 variants infecting have been described in Europe (Thomas
et al. 2007), Indonesia (Utama et al. 2010) and throughout the Caribbean (Sulbaran
et al. 2010; Martial et al. 2004), the latter examples in particular perhaps represent-
ing the shipment of infected West Africans through the slave trade in the eighteenth
and nineteenth centuries (Markov et al. 2009). The more recent spread of genotype
4 variants within Cameron and Egypt through medical treatment (Pybus et al. 2003;
Pepin and Labbe 2012), into Mediterranean countries and the recent rapid spread of
genotype 4 variants among IDUs in Southern Europe (Nicot et al. 2005; de Bruijne
et al. 2009) provide further examples of this model (Ndjomou et al. 2003).
What remains unexplained is the nature of the “endemic” circulation of HCV
in these implicated source areas and in particular the transmission routes that have
sustained long-term circulation of HCV in what have been until recently relatively
frequently highly isolated human communities. As discussed, transmission by either
sexual contact or from mother to child is inefficient at least in areas where it has been
studied (Wasley and Alter 2000; Pradat and Trepo 2000; Thomas 2000) and various
factors that may enhance transmission have been proposed. Examples include sexu-
ally transmitted infections (STIs), circumcision, excision and scarification practices
(Shepard et al. 2005) which at least in Central Africa show associations with HCV
infection and more remote possibilities such as mosquito or other arthropod vectors
(Pybus et al. 2007). These various hypotheses are yet to be resolved.
The time depth of “endemic” circulation of HCV remains similarly uncertain.
Molecular evolutionary reconstructions of the recent spread of HCV have pro-
duced robust and reproducible estimates of its substitution rate (see previous sec-
tion). Substitution rates extrapolated to the much larger sequence distances that
exist within genotypes (such as between subtypes 1a and 1b) have been used to
provide some kind of estimate of the minimum period over which the observed
“endemic” diversity developed (Pybus et al. 2001; Markov et al. 2009; Smith
et al. 1997; Pybus et al. 2009). Reconstructed dates for the common ancestor of
different genotypes vary but are estimated to be several hundred years ago for gen-
otype 2 and even longer for genotype 6. In the opinion of the author of this review,
such estimates should be treated with extreme caution and minimum estimates at
best. Extrapolating substitution rates measured over short observation intervals to
the much longer periods of subtype and genotype diversification makes assump-
tions about the evolutionary process that are not self-evidently justified. Factors
such as extreme rate variation between sites, large-scale RNA secondary structure,
greater selective constraints and fitness optimisation of viruses association with
large population sizes may create substantial underestimates of the real period of
virus diversification [reviewed in (Sharp and Simmonds 2011)]. While it is beyond
the scope of the current article to discuss this in detail, what can be said is that the
subtype diversification in HCV and thus the likely period of endemic circulation
in sub-Saharan Africa and Southern Asia is prolonged and likely long before long
distance travel and interactions with colonial powers. These genotypes are there-
fore likely to be truly indigenous to areas where they currently endemically cir-
culate. This takes us a step further back to the question of the ultimate source of
HCV. This much more speculative area will be reviewed in the next section.
4 Origins of Human Infections and HCV Homologues
in Other Mammals
A compelling scenario which has driven much research endeavour in the last
decade and a half is the hypothesis for a non-human primate source for HCV
infections in humans. The theory makes epidemiological sense in that high diver-
sity areas of endemic circulation in humans are those where human, ape and Old
World monkey populations overlap. Before long range travel and the means for
wider dissemination, human infections acquired zoonotically from non-human
primates may have remained geographically focussed and thus account for the
specific association of each of the genotypes in defined areas of sub-Saharan
Africa and Southern Asia. The idea of a non-human primate source for humans
is additionally consistent with the observation of its poor transmissibility between
humans, largely confined to parenteral routes and a reflection perhaps of its lack of
host adaptation as might also be its severe, immune-mediated liver pathology.
This model is, of course, also driven by the tempting analogy with the origin
of HIV-1, which similarly exploded worldwide out of Central Africa in the twen-
tieth century through infections directly or indirectly from chimpanzees (Gao et
al. 1999). As might be imagined for HCV, HIV-1 infections acquired through con-
tact with Central African chimpanzees (Pan troglodytes) may have been occurring
for centuries or millennia, but only in the last 50–70 years were demographic and
societal changes suitable for its wider pandemic spread. Important differences from
the HIV-1 model of origins would be the earlier spread of HCV worldwide and
the existence of multiple potential source areas and possibly different primate spe-
cies. These would be necessary to account for the distinct endemic distributions
of HCV genotypes in different parts of sub-Saharan Africa and also South East
Asia. Finally, into this model would come GBV-B, which might perhaps represent
a much more divergent homologue of HCV in a New World primate species.
Despite the elegance, plausibility and potential medical relevance of the
primate origin hypothesis, the fundamental problem has always been that HCV
or homologues cannot be found in ape or monkey species, at least to the author’s
knowledge. Extensive screening programmes both published (Makuwa et al.
2003, 2006) and unpublished have failed to document either seropositivity or viral
sequences in literally hundreds or thousands of plasma samples collected from dif-
ferent ape and monkey species. As a possibly related problem, GBV-B has to date
never been recovered nor serological evidence for past infections obtained from any
tamarind or other New World primate among wild populations in South America.
Without an obvious primate source for infection and the genetic evidence for
circulation of HCV in what would have been largely isolated human populations
in distinct parts of the world for centuries or more likely millennia, studies of the
ultimate origins of HCV have reached something of a frustrating impasse. As with
many other virus discoveries, however, its resolution is likely to be considerably
stranger than could have been imagined even as recently as last year. By pure ser-
endipity, attempts by Kapoor and colleagues to identify viral causes of respiratory
disease in dog held in kennels by deep sequencing revealed the existence of an
RNA virus extraordinarily similar to HCV (Fig. 3) but with suspected biologi-
cal and epidemiological properties quite different from what had been previously
described for both HCV and GBV-B (Kapoor et al. 2011).
The virus, initially termed canine hepacivirus (CHV) showed approximately
50 % nucleotide sequence divergence from HCV. Data presented in that study
demonstrated high viral loads in respiratory samples and an implied respiratory
route of transmission and association with respiratory disease, none of which have
been observed in HCV (or GBV-B) infections. Infections were found in dogs from
Fig. 3 Amino acid sequence divergence scan of members of the Hepacivirus genus, with
genome diagram drawn to scale underneath plot. NPHV is more similar to HCV throughout the
genome (red line) than GBV-B (dark green line). However, NPHV/HCV divergence is substan-
tially greater than between genotypes (dark and light blue lines respectively). This figure has
been adapted from Fig. 2 in (Kapoor et al. 2011). For details related to the HCV polyprotein and
the cleavage products see chapter “Hepatitis C virus Proteins From Structure to Function” by
Moradpour and Penin, this volume
different regions of the USA but partial genome characterisation demonstrated a
virtual absence of genetic diversity that would be expected for an RNA virus like
HCV. Whether the virus spread systemically or persisted was not demonstrated
although imaging of viral RNA in liver by in situ hybridisation was presented.
More recently, further studies of the host range of hepaciviruses in a range of
mammalian species were performed by the same group using a serological assay
for antibodies to a peptide expressed from the NS3 region of the CHV genome
(Burbelo et al. 2012). This produced further unexpected findings. From the 80
dogs, 81 deer, 84 cows, 103 horses and 14 rabbits screened, only horses showed
frequent seropositivity (35 %) with one weak positive sample from a cow while,
remarkably, all 80 dogs were seronegative. Of the 103 horse samples, 8 were PCR-
positive (all seropositive) and from each of these near-complete genome sequences
were obtained. Sequences showed moderate sequence diversity from each other
(6.4–17.2 % nucleotide sequence divergence) with the CHV sequence grouping
with horse-derived variants. As viruses similar to CHV were frequently found in
horses, the investigators coined the name NPHV to describe this group.
The diversity of NPHV variants was somewhere between inter-subtype and within
subtype divergence of HCV, certainly not the equivalent of HCV genotypes (Fig. 3).
The high degree of amino acid sequence conservation did, however, contrast mark-
edly with the degree of sequence variability at synonymous (non-coding) sites in the
genome. The extraordinarily low ratio between synonymous to non-synonymous
substitutions (0.03–0.06) indicates however that its evolution has been more severely
constrained and/or less subject to positive selection pressures than HCV. These
low sequence distances are therefore not necessarily an indication of their recent
divergence.
There was no information available on the clinical features of infection with
NPHV in horses. To address this we have recently surveyed horses in Scotland by
PCR and identified 3 viraemic horses from 136 screened (Lyons et al. 2012). Using
veterinary records and further sampling, these have been evaluated for evidence of
hepatitis or other systemic disease manifestations. Positive horses were originally
referred for reasons such as lameness, foot abscess or respiratory infections with no
evidence of the ill-health that might be associated with severe systemic infections.
Although most liver indices were in the normal range, gamma glutamyl transferase
(GGT) levels, a sensitive marker of liver inflammation were marginally or signifi-
cantly elevated along with elevation in bile acids, perhaps providing some tentative
evidence for an aetiological role of NPHV in hepatitis. Repeated sampling from one
of the study horses demonstrated persistence over at least a 6-month period and viral
loads comparable to those observed in HCV infections (7 × 10 −5 × 10 RNA
copies/ml). Respiratory samples and peripheral blood mononuclear cells from the
infected horse have proven uniformly negative although no opportunity to perform a
liver biopsy of the horse has yet presented itself. Overall, these more recent findings
provide some reassurance that hepacivirus infections in horses are both persistent and
potentially associated with mild liver disease rather than the respiratory disease and
viral secretion found originally in dogs. However, large-scale PCR-based screening of
other mammalian species using primers conserved between NPHV and HCV failed to
detect hepaciviruses in dogs (nearly 200 screened), cats, pigs and rodents (Lyons et al.
2012), very much as found in the previous serology-based study (Burbelo et al. 2012).
This, to date, represents current published knowledge of non-human hepacivi-
ruses, a series of findings that present several conflicting interpretations and dif-
ficulties. This author believes that, despite the negative results from screening so
far, domestic horses are most unlikely to be the only mammalian species (other
than human or tamarins) infected with hepaciviruses and there is clearly much to
be learned in short term from more extensive screening.
5 Concluding Thoughts
Our understanding of the ultimate origins of HCV infection in humans will doubt-
less be hugely enhanced once proper mammalian screening for other hepaciviruses
has been performed and the genetic diversity and, more importantly, the specific-
ity of different hepaciviruses to individual host species is more clearly established.
From such studies, it may well turn out that hepaciviruses are highly catholic in
their host range perhaps capable of jumping between horses and dogs as suggested
by the published screening data (Burbelo et al. 2012; Kapoor et al. 2011) and per-
haps all species in-between. An ability of hepaciviruses to jump species is consistent
with the observation that the NPHV protease is able to cleave human MAVS and
TRIF (Parera et al. 2012); this ability to prevent interferon signalling is essential for
HCV replication (Foy et al. 2005) and may therefore function across species bar-
riers and potentially favour zoonotic transmission. A wide mammalian host range
is also characteristic of vector-borne flaviviruses and pestiviruses, the latter at least
within ruminant species. In this scenario, HCV infections in humans may well have
a zoonotic origin consistent with its relatively recent emergence (at least in Western
countries). While being still relatively poorly adapted for infecting its new host, this
may further account for its peculiar, inefficient transmission routes.
Alternatively, it may be that each hepacivirus species is uniquely adapted to
one target species, HCV in humans, NPHV in horses and perhaps further hepa-
civiruses in other mammalian species waiting to be discovered. The ability of
HCV to persist lifelong in humans, an attribute that greatly enhances its transmis-
sibility and evidence for subtle virus/host interactions such as the enhancing role
of human micro RNA, miR-122 expressed in liver on virus replication (Jopling
et al. 2005) certainly hints at long-term virus/host co-adaptation. HCV may always
have infected humans throughout their evolution and it is only through greater life
expectancy, scope for epidemic transmission and better surveillance and under-
standing of causes of hepatitis that it has come to current medical attention. In that
sense, HCV does not have an “origin”, it is just one of those viruses like herpes-
viruses that have always infected humans and before them hominoids, proto-apes
and potentially right back to the ancestor of mammals themselves.
Future research will be truly important in resolving these two diametrically
opposed possibilities.
Source: http://www.roslin.ed.ac.uk/assets/profile-pages/peter-simmonds/the-origin-of-hcv.pdf
No comments:
Post a Comment