We compared 31 complete and nearly complete globally derived HSV-1 genomic sequences using HSV-2 HG52 as an outgroup to investigate their phylogenetic relationships and look for evidence of recombination. The sequences were retrieved from NCBI and were then aligned using Clustal W. The generation of a maximum likelihood tree resulted in a six clade structure that corresponded with the timing and routes of past human migration. The East African derived viruses contained the greatest amount of genetic diversity and formed four of the six clades. The East Asian and European/North American derived viruses formed separate clades. HSV-1 strains E07, E22 and E03 were highly divergent and may each represent an individual clade. Possible recombination was analyzed by partitioning the alignment into 5 kb segments, performing individual phylogenetic analysis on each partition and generating a.phylogenetic network from the results. However most evidence for recombination spread at the base of the tree suggesting that recombination did not significantly disrupt the clade structure. Examination of previous estimates of HSV-1 mutation rates in conjunction with the phylogenetic data presented here, suggests that the substitution rate for HSV-1 is approximately 1.38×10−7 subs/site/year. In conclusion, this study expands the previously described HSV-1 three clade phylogenetic structures to a minimum of six and shows that the clade structure also mirrors global human migrations. Given that HSV-1 has co-evolved with its host, sequencing HSV-1 isolated from various populations could serve as a surrogate biomarker to study human population structure and migration patterns.
Citation: Kolb AW, Ané C, Brandt CR (2013) Using HSV-1 Genome Phylogenetics to Track Past Human Migrations. PLoS ONE 8(10): e76267. https://doi.org/10.1371/journal.pone.0076267
Editor: Sudhindra R. Gadagkar, Midwestern University, United States of America
Received: October 2, 2012; Accepted: August 24, 2013; Published: October 16, 2013
Copyright: © 2013 Kolb et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. These studies were supported by grants from the NIH (R01EY07336 and R01EY018597) to CRB, a Core Grant for Vision Research (P30EY016665), a Research to Prevent Blindness Senior Scientist Award to CRB from Research to Prevent Blindness, Inc. (RPB), and an unrestricted grant to the Department of Ophthalmology and Visual Sciences from RPB, Inc.
Competing interests: The authors have declared that no competing interests exist.
Herpesviruses are large, enveloped double stranded DNA viruses with genomes that range in size from 124–295 kilobases. The alphaherpesvirus subfamily is characterized by the capacity to establish latent infections in the sensory nerve ganglia. Previous phylogenetic studies have shown that herpesviruses have co-evolved with their hosts.  Herpes simplex viruses type 1 (HSV-1) is a member of the alphaherpesviruses and has a genome size of approximately 152 Kb. HSV-1 causes oral mucocutaneus lesions as well as keratitis and encephalitis and is a significant human pathogen. ,  Animal studies in mice have shown that HSV-1 disease severity relies on three factors; innate host resistance, host immune response and viral strains. , , , , , , , , , , Neurovirulence studies with different viral strains in infected mice show that disease severity varies from no disease to lethal encephalitis. ,  Further phylogenetic and genomic analysis of viral strains may aid in understanding the genetic aspects virulence.
Previous studies of HSV-1 phylogeny have analyzed viral strains from primarily one geographic region; Europe or North America with modest sample numbers. Phylogenetic analyses with single genes , ,  or with small numbers of genomes  have consistently yielded a three clade pattern. However, phenotypic analysis using single genes or small clusters of genes may not present an accurate picture of relationships due to recombination. More accurate information on genetic relationships requires the use of whole or nearly complete genomes. Recently, next-generation sequencing techniques have been used to sequence several HSV-1 genomes , ,  with more being directly deposited into GenBank. Currently complete, or nearly complete genomic sequences are available from North America, Europe, East Asia and Eastern Africa. The goal of this study was to examine the phylogeny of the strains as well as look for evidence of recombination. The resulting analysis revealed a minimum six clade structure for HSV-1, as well as a topology based on geographic origin of the isolate. Inspection of the phylogenetic data presented here along with previous estimations of HSV-1 substitution rates suggests a rate of approximately 1.38×10−7 subs/site/year. Recombination analysis showed evidence of both inter- and intra-clade recombination.
In this study, for the first time a global sampling of HSV-1 strains has been used for phylogenetic analysis and supports the conclusion that HSV-1 strains have co-migrated with their human hosts, leading to geographically separated clades. The recent demonstration that multiplex sequencing of HSV-1 genomes is feasible  significantly reduces the cost per genome and using HSV-1 as a surrogate biomarker would reduce the cost and facilitate studies of human migration.
Materials and Methods
The genomic sequences used for analysis were obtained from the NCBI Reference Database. The genomes of HSV-2 HG52 and 31 HSV-1 strains (Figure 1) were aligned with Clustal W  using Mega 5.  The mean genetic distances between HSV-1 and HSV-2, as well as between all HSV-1 strains were calculated using the maximum composite likelihood option with “complete deletion” of alignment gaps using Mega 5. Pairwise distances between all the HSV-1 and HSV-2 strains were calculated using the maximum composite likelihood option. Complete deletion of alignment gaps was performed when HSV-2 was compared to the HSV-1 strains as a group. Pairwise deletion was performed rather than complete deletion when comparing HSV-1 strains to each other in order to minimize overestimates of distance.
A maximum likelihood (ML) phylogenetic tree was constructed with 31 HSV-1 whole or partial genomic sequences, using HSV-2 as an outgroup. (B) An expansion of the HSV-1 specific node from the ML tree in (A). The ML tree was generated from aligned sequences using the Mega 5 package. Clade I includes European/North American strains, Clade II comprises East Asian strains and III, IV, V and VI are East African. HSV-2 was used as an outgroup. The viral isolates are colored according to country of origin and are as follows: U.S.A: light blue, U.K.: dark blue, China: red, South Korea: purple, Japan: orange, and Kenya: green.
Phylogenetic and Recombinational Analysis
Prior to phylogenetic analysis, gaps in the Clustal W genomic alignment were deleted, yielding 126,608 bp in the alignment. We performed maximum likelihood (ML) analysis on the genomic alignment using the RAxMLGUI package with the GTRCAT+I model and 500 replicates . A phylogenetic network was then generated from the 500 bootstrap replicates using Splitstree 4 .
To address possible recombination between the viral strains, the alignments were broken up into twenty five, 5 kb partitions and one, 1.6 kb partition. The RAxMLGUI package was then used to analyze each of the twenty six partitions with the GTRCAT+I model and 500 bootstrap replicates. Utilizing Dendroscope 3 , a consensus tree was generated with a 70% confidence threshold from the 500 bootstrap replicates for each of the twenty six partitions. A consensus network was then assembled from the twenty six, 70% confidence threshold consensus trees with Splitstree 4.
Estimated Divergence Times
The estimated divergence times of the HSV-1 and HSV-2 were calculated using the genomic alignment with the gaps deleted and the BEAST 1.7.4 software package . First the substitution rates were allowed to vary along lineages using the uncorrelated lognormal relaxed clock model (UCLN) , using an exponential prior distribution with a mean of 5×10−5 substitutions/site/thousand years and an offset of 1×10−7, which yielded a 5% quantile value of 2.67×10−6, a median value of 3.5×10−5 and a 95% quantile value of 1.5×10−4. Additionally, the age of the Asian and European/North American population strain split was given a prior distribution with a mean of 34.0 thousand years and a standard deviation of 10.5 thousand years. Two BEAST runs were performed with 10 million generations. The resulting tree and log files were combined with LogCombiner v. 1.7.4 (http://beast.bio.ed.ac.uk/LogCombiner) with a burnin of 6 million for each run. The combined log and tree files were visualized by Tracer v. 1.5 (http://beast.bio.ed.ac.uk/Tracer). The resulting mean substitution rate was 1.34×10−4±6.4×10−7 substitutions/site/thousand years. BEAST analysis was run a second time with the UCLD model, an exponential prior distribution for substitution rates with prior mean 1E-4, and a prior age with a mean of 34.0 and a standard deviation of 5.5 to the Asian and European/North American population strain split. Two BEAST runs were performed with 10 million generations. The resulting tree and log files were combined with LogCombiner v. 1.7.4 with a burnin of 12 million. The combined log and tree files were visualized by Tracer v. 1.5 and Figtree v. 1.4 (http://beast.bio.ed.ac.uk/FigTree) respectively.
The mean genetic distance between HSV-2 and HSV-1 was calculated at 23.16% using the maximum composite likelihood and complete deletion of gaps. The pairwise genetic distances between the HSV-1 strains ranged from 0% (E10 vs. E11) to 1.31% (CJ360 vs. E03), with a mean distance of 0.8%.
To investigate the phylogenetic relationships of the available complete or nearly complete Herpes Simplex Type 1 genomic sequences, 1 HSV-2 and 31 HSV-1 sequences with origins from North America, Europe, East Africa and East Asia were obtained from NCBI (Table 1). The sequences were first aligned with Clustal W and a maximum likelihood (ML) tree (Figure 1) was generated. Figure 1A shows the initial ML tree, using HSV-2 as an outgroup and Figure 1B is an expansion of the HSV-1 specific node from the tree in Figure 1A. The resulting trees revealed a six clade pattern based on the geographic origin of the isolates. The European/North American viruses formed clade I, East Asian strains formed clade II and the East African viruses comprised clades III, IV, V and VI (Figure 1B). Only one virus did not sort according to geographic isolation and this was strain KOS, a North American derived strain which was placed in the East Asian clade II. While the East African strains E07 was placed into a node with clade IV viruses, it is genetically distant with low bootstrap values, thus we did not assign it to a clade.
To examine the phylogenetic dissonance of the maximum likelihood analysis, a phylogenetic network was constructed using the 500 bootstrap replicate trees generated from the RAxML analysis (Figure 2). The six main phylogenetic clades were recovered and the isolated position of the African strain E07 was supported. The Eurasian clades I and II form one pole, while the African clades III, IV, V and VI form a continuum to the opposite pole.
The HSV-1 strains in the network form the same six clades as in Figure 1. Clade I includes European/North American strains, Clade II comprises East Asian strains and III, IV, V and VI are East African. HSV-2 was used as an outgroup. Splitstree 4 was used to generate the network. The viral isolates are colored according to country of origin and are as follows: U.S.A: light blue, U.K.: dark blue, China: red, South Korea: purple, Japan: orange, and Kenya: green.
To address possible recombination between the HSV-1 strains, the genomic alignment was broken up into 5 kb partitions. Each partition was then subjected to maximum likelihood analysis. A consensus tree with a 70% confidence threshold was generated from 500 bootstrap replicates for each partition. A consensus network was constructed by combining the consensus trees from each of the 26 partitions into a single file. The individual consensus trees are found in Figure S1. The resulting network is shown in Figure 3. This partition derived network closely resembles the unpartitioned network in Figure 2, however there are some key differences. The first difference is a recombination bottleneck between the Eurasian clades and the remaining African strains. The European/North American and Asian viruses form two distinct clades, however three North American strains (134, CJ311 and CJ360) were placed into the Asian clade II node. The connections to the clade II node are near to the node base, suggesting ancient recombination events. The network suggests recombination has occurred between; i) Asian clade II viruses KOS and CR38 ii) African clade VI strains E03 and E22 iii), African clade IV viruses E08 and E19.
The genome alignment was partitioned into 5(70% confidence threshold) was constructed for each partition, and a consensus network was generated from the combined results using Splitstree 4. Clade I includes European/North American strains, Clade II comprises North American/East Asian strains and III, IV, V and VI are East African HSV-2 was used as an outgroup. The viral isolates are colored according to country of origin and are as follows: U.S.A: light blue, U.K.: dark blue, China: red, South Korea: purple, Japan: orange, and Kenya: green.
Following the initial phylogenetic tree analysis, we sought to estimate the relative divergence time for the strains in our analysis. Three preceding studies estimated the mutation rate of HSV-1 and herpes viruses in general to be 1.82×10−8 , 3×10−8  and 3×10−9  substitutions/site/year respectively. To determine what substitution rate fits best with the human population split data, BEAST analysis was performed using a wide substitution rate range, corresponding to 2.67×10−9 to 1.5×10−7 subs/site/year. A prior assumed that the HSV-1 European/North American and Asian clade split was 34,000±10,500 years BP so as to correspond with the human European and Asian population split 23–45 thousand years ago , , , . The HSV-2 strain HG52 remained in the analysis as an outgroup. The BEAST analysis subsequently inferred an overall substitution rate of 1.34×10−7 (95% HPD upper: 2.14×10−7; 95% HPD lower: 7.48×10−8)subs/site/year. With the optimal substitution rate calculated, BEAST analysis was performed a second time with a prior mean substitution rate of 1×10−7 and an additional prior of 34,000±5,500 years BP. The BEAST analysis subsequently calculated an overall substitution rate of 1.38×10−7 (95% HPD upper: 1.89×10−7; 95% HPD lower: 9.5×10−8) subs/site/year. The resulting tree produced by BEAST is found in figure S2. The estimated divergence times are summarized in Table 2. Briefly, the estimated HSV-1 and HSV-2 divergence time was 2.18±0.753 million years BP, HSV-1 began to expand 50,300±16,700 years BP and the Eurasian strains diverged 32,800±10,900 years BP.
For the first time, a phylogenetic and recombinational analysis of all the available HSV-1 genomic sequences has been conducted. The results suggest that there are at least six clades of viruses with evidence for the possible existence of others. Our results also show that the clade structure is consistent with other data concerning human population structure and migration patterns and the results are also in agreement with the previous conclusion that HSV-1 has co-evolved with its host. .
When we aligned the 31 HSV-1 sequences with Clustal W the mean genetic distance within the HSV-1 isolates was 0.8%. The mean genetic distance calculated here between HSV-1 and HSV-2 (23.16%) is lower than what has been reported previously.  This may be explained as an artifact due to the partial nature of several of the genomes in this study as well as the deletion of gaps within the aligned sequences. When the more fully sequenced HSV-1 strains from clade I were compared to HSV-2 using pairwise deletion, the distance increased to 27%. Therefore, the mean genetic distances between the HSV-1 strains reported here (0.8%) are likely an underestimate by approximately 15%, with the true number likely being about 0.92%.
Maximum likelihood based phylogenetic analysis of the HSV-1 strains produced a six clade tree topology that correlated with the geographic origin of the isolate with one exception, HSV-1 KOS. Earlier phylogenetic work with single genes from the Unique Short region of the genome (US1, US4, US7 and US8) consistently produced a three clade pattern. , ,  Recently, analysis with modest numbers of genomes from Europe and North Americans of European ancestry also yielded a three clade tree topology.  These results provide support for three sub-clades originating in Europe. Here, for the first time, a global sampling was used for a phylogenetic analysis. The topologic placement of the isolates in this study broke down along strict geographic lines (Figures 1 and 2); Europe/North America, East Asia, and Africa. This finding supports the hypothesis that Alphaherpesviruses co-evolved with their hosts  and the “out of Africa” theory of human evolution , . The only strain that did not fit the geographic pattern was the North American derived strain KOS which broke the geographic topology pattern because it sorted into the East Asian clade II lineage. There are at least two potential explanations for the KOS lineage; it could represent recent global dissemination related to travel or, KOS may originally have been from the native Amerindian population. This is discussed more fully in the subsequent human migration section.
Our recovery of a six clade topology is not surprising and is likely temporary given the small number of sequences in the dataset. For example the East African strain E07 may represent a 7th clade. The additional collection and sequencing of isolates other parts of the world, notably Western/Southern Africa, India, Melanesia, Central/South America, and Amerindian populations will probably yield new clades and may reveal firmer details of the history and migration patterns in these populations.
Estimating Divergence Times
The observation that the HSV-1 viral strains sorted according to geographic origin and supported the “out of Africa” theory of human evolution suggested that a relaxed molecular clock could be applied to determine of date of divergence. Three previous estimates of either general herpesvirus or HSV-1 mutation rates have been reported as 1.82×10−8 , 3×10−8  and 3×10−9  substitutions/site/year. As such we sought to determine the substitution rate which best fit the human population divergence data. BEAST analysis was first performed with a wide substitution rate range, 2.67×10−9 to 1.5×10−7 subs/site/year and a prior (34,000±10,500 years BP) linking the viral European/North American and Asian strain split to that of the corresponding human strain split 23–45 thousand years BP , , , . The resulting substitution rate was 1.34×10−7 subs/site/year, which is at least an order of magnitude greater than the previous estimates. It is unclear as to the discrepancy of substitution rates, however previous estimates examined only a small subset of genes as well as one or a small number of HSV-1 strains.
The BEAST analysis estimated an HSV-1 and HSV-2 divergence time of 2.184±0.753 million years BP. This time period corresponds roughly to the advent of the genus homo . It unclear as to what would have precipitated the split between HSV-1 and 2, however a cognitive or behavioral change could be speculated as a cause.
Herpes simplex virus genomes are known to undergo high rates of recombination  and this can confound the phylogenetic analysis and the use of such data to calculate divergence times. However, most, if not all, of the data on HSV-1 recombination has been generated in laboratory settings where co-infection with large amounts of virus is used , , , , , , , , . Such laboratory studies however, are highly artificial and it is not clear if these data can be extrapolated to natural infections.
There is little, if any, information available regarding how common recombination occurs in humans and there are several features of the natural history of HSV-1 that would act to reduce the chances of co-infection or superinfection with two different strains of virus. Transmission occurs by close contact, most commonly through infected saliva and is thus strongly interfamilial. Viral replication is restricted and localized to the site of infection and the innervating sensory nerve ganglia. The virus does not disseminate in its host. This reduces the number of infected cells available for recombination to occur. Viral replication subsides within a week or two and latent infection is established where a small percentage of neurons contain the virus and replication is suppressed. The low number of cells involved reduces the chances of co-infection of a single neuron with two strains of virus. Primary infection generates an adaptive immune response that suppresses replication and could reduce the probability of superinfection. Finally, expression of the viral glycoprotein D in cells renders them resistant to superinfection  and the latency-associated transcript (LAT), which is expressed in latently infected neurons interferes with superinfection . When considered together, the probability of a circumstance that could lead to the generation of recombinant viruses could be quite low in natural infections.
To account for the potential effect of recombination one could identify regions of the viral genome that have not recombined but such regions are difficult to identify. In addition, as the number of strains available for analysis increases the probability of recombination free regions decreases. To date there do not appear to be significant hot spots for recombination and recombination appears to be random across the genome , , .
Bowden et.al.  sequenced 3 loci comprising 3% of the HSV-1 genome and reported high rates of recombination in a collection of strains from the United Kingdom or Korea. However, the use of single genes or small sets of genes can result in highly biased phylogenies that do not necessarily identify actual relationships. We took an alternative approach where we divided the genomes of the 31 isolates into twenty-five 5 KB segments and one 1.6 Kb segment. We then constructed 500 individual trees for each segment and then used these to generate a partition based network. The resulting network (Figure 3) suggests a recombination bottleneck, highlighted by a red circle, between the African and Eurasian strains. This finding supports an “out of Africa” model of human population spread, with limited back migration into Africa. The topology of the partition based network (Figure 3) closely resembles the ML bootstrap based network (Figure 2). The same six clades were recovered in the partition based network, however the topology between clades I and II was changed. The European/North American strains 134, CJ311 and CJ360 were clustered near the base of Asian clade II, which suggests ancient recombination events. Further analysis of the partition based network also indicated that the majority of recombination that was detected across the entirety of the tree occurred near the root nodes. Once the individual strains began diverging there was little evidence for recombination. The exceptions included; i) the African strains E03 and E22 ii)E08 and E19 and iii) KOS and CR38. Note that the recombination at the roots in the two European/North American groups occurred within the same cluster. This analysis suggests that recombination is not a confounding factor and can be accounted for in using HSV-1 genome sequences to study human populations.
Previous investigations examining recombination at the genomic level as well as with groups of single genes , , suggested that most if not all of the HSV-1 strains analyzed were recombinants and were genetic mosaics. The partitioned based network (Figures 3) presented here reinforces that conclusion. Recombination appeared to be random across the genome without obvious recombination hotspots or cold spots detected (data not shown).
Relation to Human Migration
Other human pathogens such as JC virus , ,  and Helicobacter pylori , ,  have been shown to co-migrate and diversify with their human hosts. The phylogenetic tree data presented demonstrates that HSV-1 also does the same. HSV-1 establishes a latent, persistent infection and which enables it to easily travel with its host. While preliminary, our data raise the possibility that HSV-1 sequences could serve as a surrogate marker to analyze human migration and population structures. This would greatly facilitate such studies because viral isolates are easy to obtain and multiplex sequencing of viral genomes is much less costly than sequencing human genomes or SNP analysis. The HSV-1 genome is approximately 30 times larger than the JC virus genome, which may allow for finer genetic mapping due to a larger number of SNPs per genome.
The four clade structure recovered from the Kenyan samples shows the high level of diversity in HSV-1 sequence from this area and correlates with the genetic diversity of human populations in East Africa. No data was available from GenBank specifying the ethnic group from which the Kenyan samples were derived. It is tempting to speculate however that the four clades may be a result of the four major ethnic groups which have historically occupied this area of East Africa , . Clade VI could be associated with hunter gatherer groups, which are thought to be the first to appear, clade V with Cushitic people, clade IV with Nilotic peoples and clade III with Bantu groups. The analysis of additional isolates could confirm these speculations and could further validate the use of HSV-1 in studying the history of human populations.
The placement of North American derived strain KOS with the East Asian clade II was the only strain not to follow geographical lines. This could be due to access to modern travel or it could represent an indigenous Amerindian isolate. The BEAST analysis calculated an estimated divergence time of 15,760±5,300 (Table 2) between strain KOS and the Chinese virus CR38. This divergence time fits with the estimated time period in which the North American continent was populated from Asia, approximately 15,000 years BP . As such we would propose that KOS is a representative of an Amerindian HSV-1 strain. A summary figure featuring the geographic location of the HSV-1 clades with respect to human migration is found in Figure 4.
The phylogenetic data supports the “out of Africa model” of human migration with HSV-1 traveling and diversifying with its human host. Each clade is depicted by a roman numeral inside a circle. Land migration is depicted by yellow lines and air/sea migration is shown by the pink line. The countries of origin for the strains in the current study are China (red), Japan (orange), Kenya (dark green), South Korea (purple), UK (dark blue), and USA (light blue). The map was generated using R (version 3.4.2, “maps” package).
In conclusion, for the first time global genome sequences from HSV-1 were subjected to phylogenetic and recombinational analysis. The results suggest the existence a minimum of six clades that sort according to the geographic origin of the strains. The recombinational analysis suggests that both intra- and inter-clade recombination have occurred. These results also suggest that sequencing and analysis of HSV-1 strains could serve as a surrogate marker to study human population structure and migration patterns.
Consensus trees (70% threshold value) for each of 5 kb partitioning of the genomic alignment.
Conceived and designed the experiments: Aaron W. Kolb Cecile Ane Curtis R. Brandt. Performed the experiments: Aaron W. Kolb Cecile Ane Curtis R. Brandt. Analyzed the data: Aaron W. Kolb Cecile Ane Curtis R. Brandt. Contributed reagents/materials/analysis tools: Aaron W. Kolb Cecile Ane Curtis R. Brandt. Wrote the paper: Aaron W. Kolb Cecile Ane Curtis R. Brandt.
- 1. McGeoch DJ, Dolan A, Ralph AC (2000) Toward a comprehensive phylogeny for mammalian and avian herpesviruses. J Virol 74: 10401–10406.
- 2. Whitley RJ (1996) Herpes simplex viruses. In: Fields BN, Knipe DM, Howley PM, editors. Fields Virology. 3 ed. Philadelphia: Lippincott-Raven. 2297–2342.
- 3. Liesegang TJ (2001) Herpes simplex virus epidemiology and ocular importance. Cornea 20: 1–13.
- 4. Bhattacharjee PS, Neumann DM, Foster TP, Bouhanik S, Clement C, et al. (2008) Effect of human apolipoprotein E genotype on the pathogenesis of experimental ocular HSV-1. Exp Eye Res 87: 122–130.
- 5. Han X, Lundberg P, Tanamachi B, Openshaw H, Longmate J, et al. (2001) Gender influences herpes simplex virus type 1 infection in normal and gamma interferon-mutant mice. J Virol 75: 3048–3052.
- 6. Burgos JS, Ramirez C, Sastre I, Valdivieso F (2006) Effect of apolipoprotein E on the cerebral load of latent herpes simplex virus type 1 DNA. J Virol 80: 5383–5387.
- 7. Kastrukoff LF, Lau AS, Puterman ML (1986) Genetics of natural resistance to herpes simplex virus type 1 latent infection of the peripheral nervous system in mice. J Gen Virol 67 (Pt 4): 613–621.
- 8. Lopez C (1975) Genetics of natural resistance to herpesvirus infections in mice. Nature 258: 152–153.
- 9. Lundberg P, Welander P, Openshaw H, Nalbandian C, Edwards C, et al. (2003) A locus on mouse chromosome 6 that determines resistance to herpes simplex virus also influences reactivation, while an unlinked locus augments resistance of female mice. J Virol 77: 11661–11673.
- 10. Sørensen LN, Reinert LS, Malmgaard L, Bartholdy C, Thomsen AR, et al. (2008) TLR2 and TLR9 synergistically control herpes simplex virus infection in the brain. J Immunol 181: 8604–8612.
- 11. Stulting RD, Kindle JC, Nahmias AJ (1985) Patterns of herpes simplex keratitis in inbred mice. Invest Ophthalmol Vis Sci 26: 1360–1367.
- 12. Zhang SY, Jouanguy E, Ugolini S, Smahi A, Elain G, et al. (2007) TLR3 deficiency in patients with Herpes simplex encephalitis. Science 317: 1522–1527.
- 13. Koelle DM, Corey L (2003) Recent progress in herpes simplex virus immunobiology and vaccine research. Clin Microbiol Rev 16: 96–113.
- 14. Pollara G, Katz DR, Chain BM (2004) The host response to herpes simplex virus infection. Curr Opin Infect Dis 17: 199–203.
- 15. Doymaz MZ, Rouse BT (1992) Immunopathology of herpes simplex virus infections. Curr Topics Microbiol Immunol 179: 121–136.
- 16. Streilein JW, Dana MR, Ksander BR (1997) Immunity causing blindness: five different paths to herpes stromal keratitis. Immunol Today 18: 443–449.
- 17. Thomas J, Rouse BT (1997) Immunopathogenesis of herpetic ocular disease. Immunol Res 16: 375–386.
- 18. Brandt CR (2004) Virulence genes in herpes simplex virus type 1 corneal infection. Curr Eye Resh 29: 103–117.
- 19. Brandt CR (2005) The role of viral and host genes in corneal infection with herpes simplex virus type 1. Exp Eye Res 80: 607–621.
- 20. Norberg P, Bergstrom T, Rekabdar E, Lindh M, Lijeqvist J (2004) Phylogenetic analysis of clinical herpes simplex virus type 1 isolates identified three genetic groups and recombinant viruses. J Virol 78: 10755–10764.
- 21. Kolb AW, Schmidt TR, Dyer DW, Brandt CR (2011) Sequence variation and phosphorylation sites in the herpes simplex virus US1 ocular virulence determinant. Invest Ophthalmol Vis Sci 52: 4630–4638.
- 22. Norberg P, Tyler S, Severini A, Whitley R, Liljeqvist JA, et al. (2011) A genome-wide comparative evolutionary analysis of herpes simplex virus type 1 and varicella zoster virus. PLoS One 6: e22527.
- 23. Kolb AW, Adams M, Cabot EL, Craven M, Brandt CR (2011) Multiplex sequencing of seven ocular Herpes simplex virus Type-1 genomes: Phylogeny, sequence variability and SNP distribution. Invest Ophthalmol Vis Sci 52: 9061–9073.
- 24. Szpara ML, Parsons L, Enquist LW (2010) Sequence variability in clinical and laboratory isolates of herpes simplex virus 1 reveals new mutations. J Virol 84: 5303–5313.
- 25. Macdonald SJ, Mostafa HH, Morrison LA, Davido DJ (2012) Genome sequence of herpes simplex virus 1 strain KOS. J Virol 86: 6371–6372.
- 26. Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, et al. (2007) Clustal W and Clustal X version 2.0. Bioinformatics 23: 2947–2948.
- 27. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, et al. (2011) MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol 28: 2731–2739.
- 28. Berger SA, Krompass D, Stamatakis A (2011) Performance, accuracy, and Web server for evolutionary placement of short sequence reads under maximum likelihood. Syst Biol 60: 291–302.
- 29. Huson DH, Bryant D (2006) Application of phylogenetic networks in evolutionary studies. Mol Biol Evol 23: 254–267.
- 30. Huson DH, Scornavacca C (2012) Dendroscope 3: an interactive tool for rooted phylogenetic trees and networks. Syst Biol 61: 1061–1067.
- 31. Drummond AJ, Suchard MA, Xie D, Rambaut A (2012) Bayesian phylogenetics with BEAUti and the BEAST 1.7. Mol Biol Evol 29: 1969–1973.
- 32. Drummond AJ, Ho SY, Phillips MJ, Rambaut A (2006) Relaxed phylogenetics and dating with confidence. PLoS Biol 4: e88.
- 33. Sakaoka H, Kurita K, Iida Y, Takada S, Umene K, et al. (1994) Quantitative analysis of genomic polymorphism of herpes-simplex virus type-1 strains from 6 countries-studies of molecular evolution and molecular epidemiology of the virus J Gen Virol. 75: 513–527.
- 34. Cavalli-Sforza LL, Feldman MW (2003) The application of molecular genetic approaches to the study of human evolution. Nat Genet 33 Suppl: 266–275
- 35. Gutenkunst RN, Hernandez RD, Williamson SH, Bustamante CD (2009) Inferring the joint demographic history of multiple populations from multidimensional SNP frequency data. PLoS Genet 5: e1000695.
- 36. Gronau I, Hubisz MJ, Gulko B, Danko CG, Siepel A (2011) Bayesian inference of ancient human demography from individual genome sequences. Nat Genet 43: 1031–1034.
- 37. Mellars P (2006) Going east: new genetic and archaeological perspectives on the modern human colonization of Eurasia. Science 313: 796–800.
- 38. Dolan A, Jamieson FE, Cunningham C, Barnett BC, McGeoch DJ (1998) The genome sequence of herpes simplex virus type 2. J Virol 72: 2010–2021.
- 39. Cann RL, Stoneking M, Wilson AC (1987) Mitochondrial DNA and human evolution. Nature 325: 31–36.
- 40. Stringer CB, Andrews P (1988) Genetic and fossil evidence for the origin of modern humans. Science 239: 1263–1268.
- 41. Dirks PH, Kibii JM, Kuhn BF, Steininger C, Churchill SE, et al. (2010) Geological setting and age of Australopithecus sediba from southern Africa. Science 328: 205–208.
- 42. Thiry E, Meurens F, Muylkens B, McVoy M, Gogev S, et al. (2005) Recombination in alphaherpesviruses. Rev Med Virol 15: 89–103.
- 43. Umene K (1985) Intermolecular recombination of the herpes simplex virus type 1 genome analysed using two strains differing in restriction enzyme cleavage sites. J Gen Virol 66 (Pt 12): 2659–2670.
- 44. Javier RT, Sedarati F, Stevens JG (1986) Two avirulent herpes simplex viruses generate lethal recombinants in vivo. Science 234: 746–748.
- 45. Sedarati F, Javier RT, Stevens JG (1988) Pathogenesis of a lethal mixed infection in mice with two nonneuroinvasive herpes simplex virus strains. J Virol 62: 3037–3039.
- 46. Brandt CR, Grau DR (1990) Mixed infection with herpes simplex virus type 1 generates recombinants with increased ocular and neurovirulence. Invest Ophthalmol Vis Sci 31: 2214–2223.
- 47. Brandt CR (1991) Mixed ocular infections identify strains of herpes simplex virus for use in genetic studies. J Virol Methods 35: 127–135.
- 48. Nishiyama Y, Kimura H, Daikoku T (1991) Complementary lethal invasion of the central nervous system by nonneuroinvasive herpes simplex virus types 1 and 2. J Virol 65: 4520–4524.
- 49. Yirrell DL, Rogers CE, Blyth WA, Hill TJ (1992) Experimental in vivo generation of intertypic recombinant strains of HSV in the mouse. Arch Virol 125: 227–238.
- 50. Brown SM, Subaksharpe JH, Harland J, Maclean AR (1992) Analysis of intrastrain recombination in herpes-simplex virus type-1 strain 17 and herpes-simplex virus type-2 strain HG52 using restriction endonuclease sites as unselected markers and temperature-sensitive lesions as selected markers. J Gen Virol 73: 293–301.
- 51. Kintner RL, Brandt CR (1995) The effect of viral inoculum level and host age on disease incidence, disease severity, and mortality in a murine model of ocular HSV-1 infection. Curr Eye Res 14: 145–152.
- 52. Campadelli-Fiume G, Qi S, Avitabile E, Foa-Tomasi L, Brandimarti R, et al. (1990) Glycoprotein D of herpes simplex virus encodes a domain which precludes penetration of cells expressing the glycoprotein by superinfecting herpes simplex virus. J Virol 64: 6070–6079.
- 53. Mador N, Panet A, Steiner I (2002) The latency-associated gene of herpes simplex virus type 1 (HSV-1) interferes with superinfection by HSV-1. J Neurovirol 8 Suppl 297–102.
- 54. Kintner RL, Allan RW, Brandt CR (1995) Recombinants are isolated at high frequency following in vivo mixed ocular infection with two avirulent herpes simplex virus type 1 strains. Arch Virol 140: 231–244.
- 55. Bowden R, Sakaoka H, Donnelly P, Ward R (2004) High recombination rate in herpes simplex virus type 1 natural populations suggests significant co-infection. Infect Genet Evol 4: 115–123.
- 56. Pavesi A (2004) Detecting traces of prehistoric human migrations by geographic synthetic maps of Polyomavirus JC. J Mol Evol 58: 304–313.
- 57. Pavesi A (2005) Utility of JC polyomavirus in tracing the pattern of human migrations dating to prehistoric times. J Gen Virol 86: 1315–1326.
- 58. Shackelton LA, Rambaut A, Pybus OG, Holmes EC (2006) JC virus evolution and its association with human populations. J Virol 80: 9928–9933.
- 59. Falush D, Wirth T, Linz B, Pritchard JK, Stephens M, et al. (2003) Traces of human migrations in Helicobacter pylori populations. Science 299: 1582–1585.
- 60. Linz B, Balloux F, Moodley Y, Manica A, Liu H, et al. (2007) An African origin for the intimate association between humans and Helicobacter pylori. Nature 445: 915–918.
- 61. Moodley Y, Linz B, Bond RP, Nieuwoudt M, Soodyall H, et al. (2012) Age of the association between Helicobacter pylori and man. PLoS Pathog 8: e1002693.
- 62. de Filippo C, Bostoen K, Stoneking M, Pakendorf B (2012) Bringing together linguistic and genetic evidence to test the Bantu expansion. Proc Biol Sci 279: 3256–3263.
- 63. Tishkoff SA, Reed FA, Friedlaender FR, Ehret C, Ranciaro A, et al. (2009) The genetic structure and history of Africans and African Americans. Science 324: 1035–1044.
- 64. Kitchen A, Miyamoto MM, Mulligan CJ (2008) A three-stage colonization model for the peopling of the Americas. PLoS One 3: e1596.