Re: Chikungunya Outbreak Confined To Bagan Panchor Village
Journal of General Virology <hr size="1" style="color: rgb(204, 204, 204);"><!-- / icon and title --><!-- message -->
Re-emergence of chikungunya and o?nyong-nyong viruses: evidence for distinct geographical lineages and distant evolutionary relationships
<nobr>Ann M. Powers<sup>1</sup></nobr>, <nobr>Aaron C. Brault<sup>1</sup></nobr>, <nobr>Robert B. Tesh<sup>1</sup></nobr> and <nobr>Scott C. Weaver<sup>1</sup></nobr>
[SIZE=-1]Department of Pathology and Center for Tropical Diseases, University of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-0609, USA<sup>1</sup> [/SIZE]
[SIZE=-1]Author for correspondence: Ann M. Powers. Fax +1 409 747 2415. e-mail
ampowers@culex.utmb.edu<script type="text/javascript"><!-- var u = "ampowers", d = "culex.utmb.edu"; document.getElementById("em0").innerHTML = '<a href="mailto:' + u + '@' + d + '">' + u + '@' + d + '<\/a>'//--></script> [/SIZE]
<!-- null -->
<table cellspacing="0" cellpadding="0" width="100%" bgcolor="#e1e1e1"><tbody><tr><td valign="center" align="left" width="5%" bgcolor="#ffffff">
</td><th valign="center" align="left" width="95%">[SIZE=+2]Abstract [/SIZE]</th></tr></tbody></table><table cellpadding="5" align="right" border="1"><tbody><tr><th align="left">[SIZE=-1]
Top
Abstract
Introduction
Methods
Results
Discussion
References
[/SIZE]
</th></tr></tbody></table>
Chikungunya (CHIK) virus is a member of the genus
Alphavirus<sup> </sup>in the family
Togaviridae. Serologically, it is most closely<sup> </sup>related to o?nyong-nyong (ONN) virus and is a member of<sup> </sup>the Semliki Forest antigenic complex.
CHIK virus is believed<sup> </sup>to be enzootic throughout much of Africa and historical evidence<sup> </sup>indicates that it spread to other parts of the world from this<sup> </sup>origin. Strains from Africa and Asia are reported to differ<sup> </sup>biologically, indicating that distinct lineages may exist. To<sup> </sup>examine the relatedness of CHIK and ONN viruses using genetic<sup> </sup>data, we conducted phylogenetic studies on isolates obtained<sup> </sup>throughout Africa and Southeast Asia. Analyses revealed that<sup> </sup>ONN virus is indeed distinct from CHIK viruses, and these viruses<sup> </sup>probably diverged thousands of years ago.
Two distinct CHIK<sup> </sup>virus lineages were delineated, one containing all isolates<sup> </sup>from western Africa and the second comprising all southern and<sup> </sup>East African strains, as well as isolates from Asia. Phylogenetic<sup> </sup>trees corroborated historical evidence that CHIK virus originated<sup> </sup>in Africa and subsequently was introduced into Asia. Within<sup> </sup>the eastern Africa and southern Africa/Asia lineage,
Asian strains<sup> </sup>grouped together in a genotype distinct from the African groups.<sup> </sup>These different geographical genotypes exhibit differences in<sup> </sup>their transmission cycles: in Asia, the virus appears to be<sup> </sup>maintained in an urban cycle with Aedes aegypti mosquito vectors,<sup> </sup>while CHIK virus transmission in Africa involves a sylvatic<sup> </sup>cycle, primarily with
Ae.
furcifer and
Ae.
africanus mosquitoes.<sup> </sup>
<!-- null -->
<table cellspacing="0" cellpadding="0" width="100%" bgcolor="#e1e1e1"><tbody><tr><td valign="center" align="left" width="5%" bgcolor="#ffffff">
</td><th valign="center" align="left" width="95%">[SIZE=+2]Introduction [/SIZE]</th></tr></tbody></table><table cellpadding="5" align="right" border="1"><tbody><tr><th align="left">[SIZE=-1]
Top
Abstract
Introduction
Methods
Results
Discussion
References
[/SIZE]
</th></tr></tbody></table>
Chikungunya (CHIK) virus, a member of the
Alphavirus genus in<sup> </sup>the family
Togaviridae, was first isolated from the serum of<sup> </sup>a febrile human in Tanganyika (Tanzania) in 1953 (Karabatsos,<sup> </sup>1985

). Between the 1960s and 1980s, the virus was isolated<sup> </sup>repeatedly from numerous countries in central and southern Africa<sup> </sup>as well as in Senegal and Nigeria in western Africa. During<sup> </sup>this same period, the virus was also identified in many areas<sup> </sup>of Asia. Since 1953, CHIK virus has caused numerous well-documented<sup> </sup>outbreaks and epidemics in both Africa and Southeast Asia, involving<sup> </sup>hundreds of thousands of people (Halstead
et al., 1969
a
,
b
<sup> </sup>; Rao, 1966

). CHIK virus infection produces an illness in humans<sup> </sup>that is characterized by fever, headache, nausea, vomiting,<sup> </sup>myalgia, rash and arthralgia. Because the clinical symptoms<sup> </sup>of CHIK infection often mimic those of dengue fever and because<sup> </sup>CHIK virus circulates in regions where dengue virus is endemic,<sup> </sup>it has been postulated that many cases of dengue virus infection<sup> </sup>are misdiagnosed and that the incidence of CHIK virus infection<sup> </sup>is much higher than reported (Carey, 1971

).<sup> </sup>
In Africa, CHIK virus appears to be maintained in a sylvatic<sup> </sup>cycle involving wild primates and forest-dwelling
Aedes spp.<sup> </sup>mosquitoes. Serological studies have repeatedly demonstrated<sup> </sup>the presence of antibodies in humans and wild primates throughout<sup> </sup>the moist forests and semi-arid savannas of Africa (Adesina<sup> </sup>& Odelola, 1991

; Jupp & McIntosh, 1988

; Rodhain
et<sup> </sup>al., 1989

; Salim & Porterfield, 1973

; Karabatsos, 1975

<sup> </sup>).
To date, a vertebrate reservoir or sylvan transmission cycle<sup> </sup>has not been identified outside Africa, supporting the historical<sup> </sup>evidence (Carey, 1971
) that CHIK virus originated in Africa<sup> </sup>and was subsequently introduced into Asia, where it is now typically<sup> </sup>associated with Ae. aegypti mosquitoes. Strains from Africa<sup> </sup>and Asia are reported to differ biologically (Jupp & McIntosh,<sup> </sup>1988

), indicating that
distinct lineages may exist.<sup> </sup>
In 1996, a closely related alphavirus, o?nyong-nyong (ONN)<sup> </sup>virus, caused a major epidemic in southern Uganda (Lanciotti<sup> </sup>
et al., 1998

). This was the first epidemic of ONN virus infection<sup> </sup>since 1959, when a large epidemic swept across East Africa involving<sup> </sup>over 2 million reported cases (Johnson, 1988

). Unlike CHIK<sup> </sup>and all other alphaviruses, ONN virus is unique in its transmission<sup> </sup>patterns: the virus is not transmitted by culicine mosquitoes,<sup> </sup>but rather by anophelines, typically
Anopheles funestus and<sup> </sup>
An.
gambiae. A vertebrate reservoir for ONN virus has not yet<sup> </sup>been identified. The transmission of ONN virus by two common<sup> </sup>vectors that inhabit much of tropical Africa and that live in<sup> </sup>close association with humans may be a factor in the rapid spread<sup> </sup>of the virus during epidemics.<sup> </sup>
With the exception of information derived from a limited number<sup> </sup>of serosurveys, little is known about the relationships of CHIK<sup> </sup>and ONN viruses (Chanas
et al., 1979

; Karabatsos, 1975

; Porterfield,<sup> </sup>1961

). ONN is considered to be a subtype of CHIK virus: serological<sup> </sup>tests reveal a one-way antigenic cross-reactivity between the<sup> </sup>two agents. Antibody to CHIK virus reacts almost equally with<sup> </sup>both CHIK and ONN viral antigens while ONN virus antibodies<sup> </sup>react weakly against CHIK virus antigen (Blackburn
et al., 1995

<sup> </sup>; Chanas
et al., 1979

; Lee
et al., 1997

; Karabatsos, 1985

<sup> </sup>).
It was once postulated that mutations in CHIK virus led to<sup> </sup>the emergence of ONN virus and its ability to be transmitted<sup> </sup>by anopheline mosquitoes (Johnson, 1988
). However, genetic<sup> </sup>studies by Lanciotti et al. (1998)
as well as the phylogenetic<sup> </sup>analyses presented here clearly demonstrate that ONN and CHIK<sup> </sup>viruses are genetically distinct. The phylogenetic and serological<sup> </sup>studies presented here were designed to help elucidate the evolutionary<sup> </sup>relationships of these viruses and to aid in understanding their<sup> </sup>epidemic and maintenance transmission patterns.<sup> </sup>
<!-- null -->
<table cellspacing="0" cellpadding="0" width="100%" bgcolor="#e1e1e1"><tbody><tr><td valign="center" align="left" width="5%" bgcolor="#ffffff">
</td><th valign="center" align="left" width="95%">[SIZE=+2]Methods [/SIZE]</th></tr></tbody></table><table cellpadding="5" align="right" border="1"><tbody><tr><th align="left">[SIZE=-1]
Top
Abstract
Introduction
Methods
Results
Discussion
References
[/SIZE]
</th></tr></tbody></table>
Virus preparation.
The CHIK and ONN virus strains used in this study are described<sup> </sup>in Table 1

. Viruses were diluted and grown on either BHK-21<sup> </sup>or Vero 76 cells at an m.o.i. less than 1. After approximately<sup> </sup>75% of the cells exhibited cytopathic effects, the virus present<sup> </sup>in the supernatant was concentrated by polyethylene glycol precipitation<sup> </sup>(Killington
et al., 1996

). The virus pellet was resuspended<sup> </sup>in 150 ?l TEN buffer and 2 ml Trizol LS (Gibco-BRL)<sup> </sup>was added in preparation for RNA extraction according to the<sup> </sup>manufacturer?s protocol.<sup> </sup>
<!-- null -->
<center><table cellspacing="0" cellpadding="0" width="95%"><tbody><tr bgcolor="#e1e1e1"><td><table cellspacing="2" cellpadding="2"><tbody><tr bgcolor="#e1e1e1"><td valign="top" align="middle" bgcolor="#ffffff">
View this table:
<nobr>
[in this window]
[in a new window]
</nobr>
</td><td valign="top" align="left">Table 1. Viruses used in phylogenetic analyses
</td></tr></tbody></table></td></tr></tbody></table></center>
RNA extraction and RT?PCR.
RNA was extracted from the virus/Trizol suspension according<sup> </sup>to manufacturer?s protocols as described previously (Cilnis<sup> </sup>
et al., 1996

). cDNA was synthesized from the RNA using a poly(T)<sup> </sup>oligonucleotide primer (either T<sub>19</sub>V or T<sub>25</sub>V-
Mlu; 5' TTACGAATTCACGCGT<sub>25</sub>V<sup> </sup>3'). PCR amplification was performed on the first strand cDNA<sup> </sup>using the poly(T) primer and a forward primer designed to anneal<sup> </sup>to genome positions 10344 to 10360 (5' TACCCNTTYATGTGGGG 3')<sup> </sup>of ONN strain SG650, covering the carboxy-terminal portion of<sup> </sup>the E2 envelope glycoprotein gene. The following parameters<sup> </sup>were used: 30 cycles of denaturation at 95 ?C for 30 s,<sup> </sup>primer annealing at 50 ?C for 30 s, and extension<sup> </sup>at 72 ?C for 3 min. A 10 min final extension<sup> </sup>was used to ensure complete product synthesis.<sup> </sup>
Sequencing/phylogenetic analyses.
PCR products ranging from 1?2 to 1?7 kb<sup> </sup>were isolated from 1% agarose gels. The cleaned DNA fragments<sup> </sup>were cloned into the pCR2.1 TA cloning vector (Invitrogen) and<sup> </sup>white bacterial colonies screened for plasmids containing inserts<sup> </sup>of the correct size. Selected clones were sequenced using the<sup> </sup>plasmid-specific T7 promoter and m13 reverse primers combined<sup> </sup>with internal, CHIK virus-specific primers (C3205, 5' GCRACAAACCCSGTAAG<sup> </sup>3'; C3152, 5' ACTGGCTRAAAGAACGAGG 3'). Sequencing was performed<sup> </sup>using an Applied Biosystems Prism 377 sequencer and automated<sup> </sup>DNA sequencing kit. The deduced amino acid sequences were aligned<sup> </sup>by using the PILEUP program in the Wisconsin Package (Genetics<sup> </sup>Computer Group) with default parameters, and the nucleotide<sup> </sup>sequences were aligned manually based on codon homology. Phylogenetic<sup> </sup>analyses were performed using maximum parsimony, neighbour joining<sup> </sup>and maximum likelihood programs implemented in the PAUP 4.0<sup> </sup>software (Swofford, 1998

). Distance analyses used the Kimura<sup> </sup>2-parameter formula to correct for multiple substitutions of<sup> </sup>the same nucleotides. Unordered and ordered characters (transition/transversion<sup> </sup>ratio of 4:1; based on previous alphavirus estimates) were used<sup> </sup>in the parsimony analysis. Alphaviruses in the Venezuelan equine<sup> </sup>encephalitis, Barmah Forest and eastern equine encephalitis<sup> </sup>antigenic complexes were used as an outgroup. Bootstrap analysis<sup> </sup>(Felsenstein, 1985

) was performed with 1000 replicates to determine<sup> </sup>confidence values on the clades within trees.<sup> </sup>
Estimation of divergence times.
An average divergence rate for CHIK and ONN virus lineages<sup> </sup>was estimated by identification of sister-sequence pairs that<sup> </sup>were robust (bootstrap values >>90%), closely related and isolated<sup> </sup>at least 7 years apart in the same geographical region. The<sup> </sup>number of differences in synonymous changes depicted in branch<sup> </sup>lengths separating each sister sequence from the predicted common<sup> </sup>ancestor?s sequence was divided by the number of years<sup> </sup>between isolations to yield rates expressed as changes per nucleotide<sup> </sup>per year, and several estimates were compared to provide an<sup> </sup>estimated mean and standard deviation. Synonymous nucleotide<sup> </sup>divergence estimates for pair-wise sequence comparisons were<sup> </sup>generated using the formula of Li
et al. (1985)

to correct<sup> </sup>for multiple substitutions of the same nucleotides.<sup> </sup>
Production of immune sera.
Syrian golden hamsters and BALB/C mice were used to generate<sup> </sup>immune sera to three strains of CHIK virus (37997, Ross and<sup> </sup>1455/75) and one strain of ONN virus (Igbo Ora, IbH12628). Animals<sup> </sup>received a single injection of virus (~10<sup>5</sup> p.f.u./ml),<sup> </sup>either intraperitoneally (i.p.) alone or subcutaneously with<sup> </sup>a mixture of virus and an
Ae.
aegypti mosquito salivary gland<sup> </sup>suspension to enhance the infection. Approximately 4 weeks post-inoculation,<sup> </sup>blood was obtained from the rodents from the retroorbital sinus<sup> </sup>and tested for antibody to CHIK virus by an immunofluorescent<sup> </sup>antibody assay (IFA) or by neutralization test (NT). Mice that<sup> </sup>were positive for CHIK virus antibody by IFA were injected i.p.<sup> </sup>with sarcoma 180 cells to produce hyperimmune ascitic fluid.<sup> </sup>Abdominal fluid was removed between 1 and 2 weeks after injection<sup> </sup>of the sarcoma cells and was used in IFA and plaque reduction<sup> </sup>neutralization tests (PRNT) to determine homologous titres.<sup> </sup>
Titration of neutralizing antibody.
Three of the four viruses (37997, Ross and 1455/75) generated<sup> </sup>a detectable homologous antibody response as determined by IFA<sup> </sup>(in mice) or NT (in hamsters). Only two of these, 37997 and<sup> </sup>Ross, had IFA titres sufficient to perform additional serological<sup> </sup>analyses. These two viruses were used in 80% PRNTs to determine<sup> </sup>both the homologous and heterologous neutralizing antibody titres.<sup> </sup>
<!-- null -->
<table cellspacing="0" cellpadding="0" width="100%" bgcolor="#e1e1e1"><tbody><tr><td valign="center" align="left" width="5%" bgcolor="#ffffff">
</td><th valign="center" align="left" width="95%">[SIZE=+2]Results [/SIZE]</th></tr></tbody></table><table cellpadding="5" align="right" border="1"><tbody><tr><th align="left">[SIZE=-1]
Top
Abstract
Introduction
Methods
Results
Discussion
References
[/SIZE]
</th></tr></tbody></table>
Phylogenetic analysis
Cloned PCR products ranging from approximately 1200 to 1700<sup> </sup>nucleotides of the E1 envelope glycoprotein gene and the entire<sup> </sup>3' noncoding region (NCR) were sequenced and aligned using the<sup> </sup>PILEUP program in the GCG software package. Because alignment<sup> </sup>of the 3'NCR was poor, only the E1 coding nucleotides were used<sup> </sup>in the phylogenetic analyses. Both distance matrix programs<sup> </sup>and maximum parsimony generated trees with the same basic topology,<sup> </sup>differing only in the arrangement of the CHIK virus isolates<sup> </sup>from the Asian clade.<sup> </sup>
Initial parsimony analyses revealed that several isolates, previously<sup> </sup>designated as CHIK virus, were genetically quite distinct from<sup> </sup>the prototype strain and from all other isolates examined. Inclusion<sup> </sup>of representative members of the Semliki Forest, Venezuelan<sup> </sup>equine encephalitis, Barmah Forest and eastern equine encephalitis<sup> </sup>virus antigenic complexes showed that these viruses were actually<sup> </sup>ONN (strain IPD A234), Semliki Forest (DAK ArB16878) and Sindbis-like<sup> </sup>(ArMg812 and B448) viruses.
Additionally, CHIK virus strains<sup> </sup>3412/78 and C-0392/95 were isolated from patients in Thailand<sup> </sup>suspected of having dengue virus infection, reinforcing the<sup> </sup>uncertainties of viral diagnosis based upon clinical presentation.<sup> </sup>
All of the CHIK and ONN virus isolates examined formed a monophyletic<sup> </sup>group within the Semliki Forest virus antigenic complex (Fig.<sup> </sup>1

), supported by a 100% bootstrap value. The ONN virus isolates<sup> </sup>formed a robust, distinct clade (100% bootstrap support) apart<sup> </sup>from all isolates of CHIK virus. ONN and CHIK virus sequences<sup> </sup>were approximately 28% and 13% divergent at the nucleotide and<sup> </sup>amino acid levels, respectively, underscoring the distinct nature<sup> </sup>of the two virus groups. Igbo Ora virus (strain IbH12628) grouped<sup> </sup>closely with the other strains of ONN, supporting previous reports<sup> </sup>that this is indeed an antigenic variant of ONN virus (Lanciotti<sup> </sup>
et al., 1998

).<sup> </sup>
<!-- null -->
<center><table cellspacing="0" cellpadding="0" width="95%"><tbody><tr bgcolor="#e1e1e1"><td><table cellspacing="2" cellpadding="2"><tbody><tr bgcolor="#e1e1e1"><td valign="top" align="middle" bgcolor="#ffffff">
View larger version (23K):
<nobr>
[in this window]
[in a new window]
</nobr>
</td><td valign="top" align="left">Fig. 1. Phylogenetic analysis of CHIK and ONN viruses generated by performing a PAUP analysis on the 1050 bp partial E1 gene sequence. To correct branch lengths for multiple substitutions, the Neighbor distance program was used to draw the tree utilizing the topology of the PAUP phylogram. Numbers indicate bootstrap values for the groups to the right. Letter A indicates the hypothetical ancestor used to estimate the time of divergence of Asian CHIK isolates from the African progenitor. The bar indicates horizontal distance corresponding to 5% nucleotide sequence divergence.
</td></tr></tbody></table></td></tr></tbody></table></center>
All phylogenetic methods divided the CHIK virus isolates into<sup> </sup>three distinct genotypes, based primarily on geographical origins.<sup> </sup>One CHIK virus clade consisted of the isolates from Senegal<sup> </sup>and Nigeria, forming the West Africa genotype (Fig. 1

). These<sup> </sup>were quite distinct from the remaining CHIK virus isolates having<sup> </sup>only 78 to 85% nucleotide sequence identity over the fragment<sup> </sup>analysed. The remaining CHIK virus isolates formed two clades:<sup> </sup>one contained strains from central and eastern Africa, while<sup> </sup>the other contained solely Asian isolates.
The paraphyletic<sup> </sup>grouping of the African CHIK viruses supports the historical<sup> </sup>evidence that the virus was introduced into Asia from Africa.<sup> </sup>
Estimated divergence times
An attempt was made to estimate the average rate of evolution<sup> </sup>of the CHIK and ONN viruses by comparison of sequences of sister<sup> </sup>taxa from the same geographical areas. Analysis of individual<sup> </sup>lineages was not possible because too few strains were available.<sup> </sup>Sister pairs were chosen that had bootstrap values >>90% and<sup> </sup>were isolated at least 7 years apart. Using these sequences,<sup> </sup>an estimated rate of evolution was determined to be 6[FONT=ariel,helvetica]x[/FONT]10<sup>-4</sup> substitutions<sup> </sup>per nucleotide per year with a standard deviation of 4[FONT=ariel,helvetica]x[/FONT]10<sup>-4</sup>.<sup> </sup>The same estimate was obtained when pair-wise comparisons included<sup> </sup>distance corrections using the Kimura two-parameter formula<sup> </sup>or maximum likelihood (see below). The synonymous rate was 5[FONT=ariel,helvetica]x[/FONT]10<sup>-4</sup><sup> </sup>(standard deviation 3[FONT=ariel,helvetica]x[/FONT]10<sup>-4</sup>) and the nonsynonymous rate was 6[FONT=ariel,helvetica]x[/FONT]10<sup>-5</sup><sup> </sup>(standard deviation 5[FONT=ariel,helvetica]x[/FONT]10<sup>-5</sup>). Although these estimates were based<sup> </sup>on only six sister-pair sequences and therefore had a high degree<sup> </sup>of error, they are similar to those previously determined for<sup> </sup>neotropical alphaviruses (Weaver
et al., 1993

, 1997

; Powers<sup> </sup>
et al., 1997

). Using the synonymous rate and the K<sub>s</sub> values<sup> </sup>computed for CHIK strain comparisons (ranging from 0?12<sup> </sup>to 0?25 with standard deviations of 0?03), the<sup> </sup>Asian genotype evolved from a hypothetical African ancestor<sup> </sup>(node A, Fig. 1

) an estimated 50 to 430 (?1 standard<sup> </sup>deviation) years ago. K<sub>s</sub> values for strains from the West African<sup> </sup>vs East African/Asian genotypes ranged from 0?75 to 0?86<sup> </sup>with standard deviations of 0?10 to 0?12. Using<sup> </sup>these values, the ancestor of all of the CHIK virus strains<sup> </sup>is estimated to have emerged between 150 and 1350 years ago.<sup> </sup>Although the divergence time of ONN virus from CHIK virus could<sup> </sup>not be estimated reliably due to excessive variance in the K<sub>s</sub><sup> </sup>values resulting from near saturation of synonymous changes,<sup> </sup>divergence of CHIK and ONN viruses probably occurred at least<sup> </sup>thousands of years ago.<sup> </sup>
A potential flaw in these time estimates is that substitution<sup> </sup>rates may vary across nucleotide sites, including synonymous<sup> </sup>sites, as has been reported for human immunodeficiency virus<sup> </sup>(Leitner
et al., 1997

). Unequal substitution rates across sites<sup> </sup>could result in an underestimation of true sequence divergence<sup> </sup>because the sites undergoing more change may accumulate more<sup> </sup>sequential mutations than are estimated by traditional formulas<sup> </sup>that assume equal rates. Therefore, we estimated the gamma distribution<sup> </sup>shape parameter for unequal rates using maximum likelihood analysis<sup> </sup>applied to all equally parsimonious tree topologies as well<sup> </sup>as the topology generated by neighbour joining. The gamma distribution<sup> </sup>shape parameter estimate was 0?42, and the transition/transversion<sup> </sup>ratio estimate was 4?3, similar to previous alphavirus<sup> </sup>estimates using substitution data from parsimony analyses (Cilnis<sup> </sup>
et al., 1996

; Weaver
et al., 1994

, 1997

). We used these values<sup> </sup>to generate trees with maximum likelihood branch lengths applied<sup> </sup>to tree topologies generated using maximum parsimony and neighbour<sup> </sup>joining methods. Using this approach, similar divergence time<sup> </sup>estimates were obtained, with
the Asian CHIK virus genotype<sup> </sup>emerging between 50 and 310 years ago, and the West and East<sup> </sup>African genotypes diverging 100 to 840 years ago.<sup> </sup>
Antigenic analysis
To determine the antigenic relatedness of viruses in the CHIK<sup> </sup>and ONN virus clades, one virus from each CHIK and ONN genotype<sup> </sup>was selected and used to generate antibodies in hamsters and<sup> </sup>mice (Table 2

). Four weeks after a single injection of virus,<sup> </sup>animals were bled, and their sera tested for antibodies by IFA.<sup> </sup>The three CHIK viruses all produced specific antibodies while<sup> </sup>the ONN virus-infected mice and hamsters produced no detectable<sup> </sup>antibody response. The homologous IFA titre of CHIK virus strain<sup> </sup>1455/75 was too low to be useful in neutralization assays; however,<sup> </sup>strains 37997 and Ross produced adequate antibody titres and<sup> </sup>were tested by 80% PRNT (Table 3

). Results indicated that these<sup> </sup>viruses have a greater than 4-fold difference in one direction<sup> </sup>suggesting that they are distinct enough to be classified as<sup> </sup>antigenic subtypes (Calisher & Karabatsos, 1988

; Calisher<sup> </sup>
et al., 1980

). While no antibody against ONN virus was generated<sup> </sup>here, eliminating the possibility of performing two-way cross<sup> </sup>neutralization tests between CHIK and ONN viruses, it would<sup> </sup>be reasonable to assume that because distinct genotypes of CHIK<sup> </sup>virus are sufficiently different antigenically to be considered<sup> </sup>subtypes, the ONN virus lineage would be more likely to be considered<sup> </sup>a distinct group of viruses within this antigenic complex.<sup> </sup>
<!-- null -->
<center><table cellspacing="0" cellpadding="0" width="95%"><tbody><tr bgcolor="#e1e1e1"><td><table cellspacing="2" cellpadding="2"><tbody><tr bgcolor="#e1e1e1"><td valign="top" align="middle" bgcolor="#ffffff">
View this table:
<nobr>
[in this window]
[in a new window]
</nobr>
</td><td valign="top" align="left">Table 2. Initial serological analysis of CHIK/ONN viruses
</td></tr></tbody></table></td></tr></tbody></table></center>
<!-- null -->
<center><table cellspacing="0" cellpadding="0" width="95%"><tbody><tr bgcolor="#e1e1e1"><td><table cellspacing="2" cellpadding="2"><tbody><tr bgcolor="#e1e1e1"><td valign="top" align="middle" bgcolor="#ffffff">
View this table:
<nobr>
[in this window]
[in a new window]
</nobr>
</td><td valign="top" align="left">Table 3. Serological analysis of CHIK viruses (80% PRNT titres)
</td></tr></tbody></table></td></tr></tbody></table></center>
<!-- null -->
<table cellspacing="0" cellpadding="0" width="100%" bgcolor="#e1e1e1"><tbody><tr><td valign="center" align="left" width="5%" bgcolor="#ffffff">
</td><th valign="center" align="left" width="95%">[SIZE=+2]Discussion [/SIZE]</th></tr></tbody></table><table cellpadding="5" align="right" border="1"><tbody><tr><th align="left">[SIZE=-1]
Top
Abstract
Introduction
Methods
Results
Discussion
References
[/SIZE]
</th></tr></tbody></table>
Results of our analyses support the hypothesis and historical<sup> </sup>accounts that CHIK virus probably originated in tropical Africa<sup> </sup>and subsequently was imported into southern Asia. In Africa,<sup> </sup>evidence that the virus circulates continually in sylvatic cycles<sup> </sup>has been documented for decades. The virus has been isolated<sup> </sup>from sylvatic mosquito species in several countries including<sup> </sup>Senegal, Cote d?Ivoire, Central African Republic and South<sup> </sup>Africa (Diallo
et al., 1999

; Jupp & McIntosh, 1990

; McCarthy<sup> </sup>
et al., 1996

).
The mosquito species involved vary geographically<sup> </sup>and with ecological conditions. In Senegal, for example,
Ae.
<sup> </sup>furcifer,
Ae.
taylori,
Ae.
luteocephalus,
Ae.
africanus and<sup> </sup>
Ae.
neoafricanus are the species determined to be of major importance<sup> </sup>in maintaining CHIK virus transmission cycles. Interestingly,<sup> </sup>several of these are the same mosquito species involved in maintaining<sup> </sup>yellow fever virus, perhaps suggesting that outbreaks of CHIK<sup> </sup>virus infection could be concomitant with sylvan yellow fever<sup> </sup>(Traore-Lamizana
et al., 1996

). In addition, it has been reported<sup> </sup>that different populations of
Ae.
aegypti in Senegal have distinct<sup> </sup>susceptibilities to CHIK virus (Diallo
et al., 1999

), suggesting<sup> </sup>vector strain specificity for CHIK viruses. Further characterization<sup> </sup>of the mosquito vectors present in the endemic areas and their<sup> </sup>vector competence for CHIK viruses could provide valuable information<sup> </sup>regarding the potential for re-emergence of the viruses in human<sup> </sup>populations.<sup> </sup>
In contrast to the numerous species involved in maintenance<sup> </sup>of CHIK virus infection in Africa,
Ae. aegypti and Ae. albopictus<sup> </sup>are the only vector species known to transmit CHIK virus in<sup> </sup>Asia. These are urban and peridomestic, anthropophilic mosquitoes<sup> </sup>that maintain close associations with humans. It is therefore<sup> </sup>not surprising that outbreaks of CHIK virus infection are noted<sup> </sup>more frequently in Asia than in Africa. Several studies have<sup> </sup>demonstrated the varying susceptibility of different Asian mosquito<sup> </sup>strains for CHIK viruses (Banerjee
et al., 1988

; Mourya &<sup> </sup>Banerjee, 1987

; Mourya
et al., 1987

; Tesh
et al., 1976

).<sup> </sup>Because CHIK and dengue viruses are transmitted by the same<sup> </sup>mosquito species in Asia and because the clinical symptoms of<sup> </sup>the two viral diseases are similar, the two diseases are difficult<sup> </sup>to differentiate. Furthermore,
there have been documented cases<sup> </sup>of simultaneous coinfection with CHIK and dengue viruses (Halstead,<sup> </sup>1966
; Myers & Carey, 1967
), further complicating the characterization<sup> </sup>of CHIK virus maintenance, evolution and emergence in Asia.<sup> </sup>
Another question concerning the transmission of CHIK virus in<sup> </sup>Asia relates to the high degree of genetic similarity among<sup> </sup>Asian genotype viruses. Although our sampling of the Asian virus<sup> </sup>was limited, sequences from viruses spanning a wide geographical<sup> </sup>range and isolated over a period of almost 35 years showed less<sup> </sup>than 3% nucleotide sequence divergence (Fig. 1
). This genetic<sup> </sup>conservation in Asia is intriguing for a virus that is known<sup> </sup>to be maintained only between humans and peridomestic mosquitoes.<sup> </sup>A similar, high degree of sequence conservation is observed<sup> </sup>within several other groups of alphaviruses: the North American<sup> </sup>eastern equine encephalitis viruses (Weaver
et al., 1994

; Brault<sup> </sup>
et al., 1999

), Highlands J virus from North America (Cilnis<sup> </sup>
et al., 1996

), western equine encephalitis viruses (Weaver<sup> </sup>
et al., 1997

) and the Sindbis-like viruses distributed throughout<sup> </sup>Australia (Sammels
et al., 1999

). As an example, North American<sup> </sup>eastern equine encephalitis viruses are maintained by an avian<sup> </sup>reservoir host; therefore, the increased movement of the virus<sup> </sup>due to migration of the birds may effectively increase the virus<sup> </sup>population size and decrease founder effects and genetic drift.<sup> </sup>This may explain their sequence conservation (Weaver, 1995

;<sup> </sup>Weaver
et al., 1992

; Brault
et al., 1999

).
It is unknown whether<sup> </sup>such an avian transmission cycle exists for CHIK viruses in<sup> </sup>Asia. Migratory patterns of both passerines and shorebirds do<sup> </sup>encompass much of Southeast Asia ranging from the Yellow Sea<sup> </sup>and South China Sea across the Philippines and Indonesia to<sup> </sup>Australia. Additionally, migration routes from India across<sup> </sup>the Indian Ocean to East Africa have been documented (Williams<sup> </sup>& Williams, 1990
). Serological testing of passerines and<sup> </sup>shorebirds in Southeast Asia could reveal if this is a plausible<sup> </sup>means of virus dispersal. Alternatively,
dispersal of the virus<sup> </sup>by travel of humans could account for the presence of virtually<sup> </sup>identical viruses in areas as distant as Indonesia and the Philippines<sup> </sup>to Barsi in central India, as well as the introduction of the<sup> </sup>virus into Asia from Africa.<sup> </sup>
The phylogenetic results presented here clearly demonstrate<sup> </sup>that ONN virus did not emerge via a recent mutation of CHIK<sup> </sup>virus as was once postulated (Johnson, 1988

). This hypothesis<sup> </sup>was based on serological evidence indicating that the viruses<sup> </sup>could only be distinguished by two-way specific antigenic tests<sup> </sup>(i.e. neutralization assay) or the use of monoclonal antibodies<sup> </sup>(Karabatsos, 1975

; Porterfield, 1961

). Antiserum raised against<sup> </sup>CHIK virus reacted with ONN virus but the reciprocal was not<sup> </sup>true, leading to the hypothesis that mutations in CHIK virus<sup> </sup>generated altered structural configurations in ONN virus affecting<sup> </sup>seroassay results (Johnson, 1988

; Williams & Woodall, 1961

<sup> </sup>; Williams
et al., 1962

). It was suggested that these same<sup> </sup>mutations were responsible for the novel ability of ONN virus<sup> </sup>to replicate in and be transmitted by anopheline mosquitoes.<sup> </sup>However, if ONN virus undergoes periodic emergence from CHIK<sup> </sup>virus progenitors, ONN virus isolates from the outbreak in Uganda<sup> </sup>in 1996 would be predicted to group phylogenetically with CHIK<sup> </sup>virus isolates rather than with the other strains of ONN virus<sup> </sup>as seen in our analysis (Fig. 1

). For example, repeated emergence<sup> </sup>from a common progenitor has been found with epidemic/epizootic<sup> </sup>Venezuelan equine encephalitis viruses, which emerge periodically<sup> </sup>from enzootic viruses and occupy clades nested within the enzootic<sup> </sup>ID lineage (Kinney
et al., 1992

; Powers
et al., 1997

; Weaver<sup> </sup>
et al., 1996

).<sup> </sup>
In addition to the antigenic and sequence differences between<sup> </sup>CHIK and ONN viruses, differences in several other biological<sup> </sup>patterns exist. Studies examining the relative ability of various<sup> </sup>strains of CHIK and ONN to replicate in different cell types<sup> </sup>have shown clear distinctions between these two viruses. CHIK<sup> </sup>viruses can replicate in both
Ae.
aegypti cell lines and numerous<sup> </sup>
Aedes spp. mosquitoes (Chanas
et al., 1979

; Jupp & McIntosh,<sup> </sup>1988

; Mourya
et al., 1987

) while ONN does not appear to replicate<sup> </sup>in
Ae.
aegypti cells (Chanas
et al., 1979

). Interestingly,<sup> </sup>both CHIK and ONN viruses can replicate in
An.
gambiae cells;<sup> </sup>however, only ONN replicates in and is believed to be transmitted<sup> </sup>primarily by
An.
gambiae or
An.
funestus mosquitoes under natural<sup> </sup>conditions (Corbet
et al., 1961

; Williams
et al., 1965

). Differences<sup> </sup>between the plaque sizes of CHIK and ONN viruses on mammalian<sup> </sup>cells have also been described (Chanas
et al., 1979

; Tesh
et<sup> </sup>al., 1976

); however, among CHIK and ONN viruses, plaque size<sup> </sup>may be strain specific (Chanas
et al., 1979

).<sup> </sup>
A putative explanation for the varying biological properties<sup> </sup>among CHIK and ONN viruses is differences in the 3'NCR. All<sup> </sup>alphaviruses sequenced have repeat sequence elements in the<sup> </sup>3'NCR that vary in length and number, often according to serogroup<sup> </sup>(Pfeffer
et al., 1998

; reviewed in Strauss & Strauss, 1994).<sup> </sup>Within some virus groups (e.g. the Sindbis-like viruses) very<sup> </sup>little sequence heterogeneity is detected in the 3'NCR (Shirako<sup> </sup>
et al., 1991

) while other alphaviruses including Ross River<sup> </sup>virus, Venezuelan equine encephalitis and Semliki Forest complex<sup> </sup>viruses show high degrees of 3'NCR variation in both length<sup> </sup>and nucleotide composition (Faragher & Dalgarno, 1986

;<sup> </sup>Pfeffer
et al., 1998

). Kuhn
et al. (1991)

have shown that<sup> </sup>changes in the repeat sequence elements can affect virus replication<sup> </sup>in different cell types, suggesting that this region may be<sup> </sup>important in binding cellular proteins utilized during virus<sup> </sup>replication. Our sequences from 22 strains of CHIK and ONN viruses<sup> </sup>from a diverse geographical and temporal range demonstrated<sup> </sup>such a high degree of variability in the 3'NCR that nucleotide<sup> </sup>sequence alignments in this area were unreliable. This information,<sup> </sup>combined with the knowledge that the replicative ability of<sup> </sup>a given CHIK or ONN viral strain varies tremendously with different<sup> </sup>strains of mosquitoes, may support Kuhn?s hypothesis.<sup> </sup><!-- null -->
http://vir.sgmjournals.org/cgi/content/full/81/2/471
Florida1 says: It would be nice if the genetic data from the chik in La Reunion was released so that a comparison could be made to the chik in India.
From this thread:
http://www.flutrackers.com/forum/showthread.php?p=4362#post4362