Human immunodeficiency virus type 1 subtype B ancestral envelope protein is functional and elicits neutralizing antibodies in rabbits similar to those elicited by a circulating subtype B envelope.

Human immunodeficiency virus type 1 (HIV-1) is a difficult target for vaccine development, in part because of its ever-expanding genetic diversity and attendant capacity to escape immunologic recognition. Vaccine efficacy might be improved by maximizing immunogen antigenic similarity to viruses likely to be encountered by vaccinees. To this end, we designed a prototype HIV-1 envelope vaccine using a deduced ancestral state for the env gene. The ancestral state reconstruction method was shown to be >95% accurate by computer simulation and 99.8% accurate when estimating the known inoculum used in an experimental infection study in rhesus macaques. Furthermore, the deduced ancestor gene differed from the set of sequences used to derive the ancestor by an average of 12.3%, while these latter sequences were an average of 17.3% different from each other. A full-length ancestral subtype B HIV-1 env gene was constructed and shown to produce a glycoprotein of 160 kDa that bound and fused with cells expressing the HIV-1 coreceptor CCR5. This Env was also functional in a virus pseudotype assay. When either gp160- or gp140-expressing plasmids and recombinant gp120 were used to immunize rabbits in a DNA prime-protein boost regimen, the artificial gene induced immunoglobulin G antibodies capable of weakly neutralizing heterologous primary HIV-1 strains. The results were similar for rabbits immunized in parallel with a natural isolate, HIV-1 SF162. Further design efforts to better present conserved neutralization determinants are warranted.

Substitution model of sequence evolution for the human immunodeficiency virus type 1 subtype B gp120 gene over the C2-V5 region.

Phylogenetic analyses frequently rely on models of sequence evolution that detail nucleotide substitution rates, nucleotide frequencies, and site-to-site rate heterogeneity. These models can influence hypothesis testing and can affect the accuracy of phylogenetic inferences. Maximum likelihood methods of simultaneously constructing phylogenetic tree topologies and estimating model parameters are computationally intensive, and are not feasible for sample sizes of 25 or greater using personal computers. Techniques that initially construct a tree topology and then use this non-maximized topology to estimate ML substitution rates, however, can quickly arrive at a model of sequence evolution. The accuracy of this two-step estimation technique was tested using simulated data sets with known model parameters. The results showed that for a star-like topology, as is often seen in human immunodeficiency virus type 1 (HIV-1) subtype B sequences, a random starting topology could produce nucleotide substitution rates that were not statistically different than the true rates. Samples were isolated from 100 HIV-1 subtype B infected individuals from the United States and a 620 nt region of the env gene was sequenced for each sample. The sequence data were used to obtain a substitution model of sequence evolution specific for HIV-1 subtype B env by estimating nucleotide substitution rates and the site-to-site heterogeneity in 100 individuals from the United States. The method of estimating the model should provide users of large data sets with a way to quickly compute a model of sequence evolution, while the nucleotide substitution model we identified should prove useful in the phylogenetic analysis of HIV-1 subtype B env sequences.

Human immunodeficiency virus type 1 env sequences from Calcutta in eastern India: identification of features that distinguish subtype C sequences in India from other subtype C sequences.

India is experiencing a rapid spread of human immunodeficiency virus type 1 (HIV-1), primarily through heterosexual transmission of subtype C viruses. To delineate the molecular features of HIV-1 circulating in India, we sequenced the V3-V4 region of viral env from 21 individuals attending an HIV clinic in Calcutta, the most populous city in the eastern part of the country, and analyzed these and the other Indian sequences in the HIV database. Twenty individuals were infected with viruses having a subtype C env, and one had viruses with a subtype A env. Analyses of 192 subtype C sequences that included one sequence for each subject from this study and from the HIV database revealed that almost all sequences from India, along with a small number from other countries, form a phylogenetically distinct lineage within subtype C, which we designate C(IN). Overall, C(IN) lineage sequences were more closely related to each other (level of diversity, 10.2%) than to subtype C sequences from Botswana, Burundi, South Africa, Tanzania, and Zimbabwe (range, 15.3 to 20.7%). Of the three positions identified as signature amino acid substitution sites for C(IN) sequences (K340E, K350A, and G429E), 56% of the C(IN) sequences contained all three amino acids while 87% of the sequences contained at least two of these substitutions. Among the non-C(IN) sequences, all three amino acids were present in 2%, while 22% contained two or more of these amino acids. These results suggest that much of the current Indian epidemic is descended from a single introduction into the country. Identification of conserved signature amino acid positions could assist epidemiologic tracking and has implications for the development of a vaccine against subtype C HIV-1 in India.

Testing the hypothesis of a recombinant origin of human immunodeficiency virus type 1 subtype E.

The human immunodeficiency virus type 1 (HIV-1) epidemic in Southeast Asia has been largely due to the emergence of clade E (HIV-1E). It has been suggested that HIV-1E is derived from a recombinant lineage of subtype A (HIV-1A) and subtype E, with multiple breakpoints along the E genome. We obtained complete genome sequences of clade E viruses from Thailand (93TH057 and 93TH065) and from the Central African Republic (90CF11697 and 90CF4071), increasing the total number of HIV-1E complete genome sequences available to seven. Phylogenetic analysis of complete genomes showed that subtypes A and E are themselves monophyletic, although together they also form a larger monophyletic group. The apparent phylogenetic incongruence at different regions of the genome that was previously taken as evidence of recombination is shown to be not statistically significant. Furthermore, simulations indicate that bootscanning and pairwise distance results, previously used as evidence for recombination, can be misleading, particularly when there are differences in substitution or evolutionary rates across the genomes of different subtypes. Taken jointly, our analyses suggest that there is inadequate support for the hypothesis that subtype E variants are derived from a recombinant lineage. In contrast, many other HIV strains claimed to have a recombinant origin, including viruses for which only a single parental strain was employed for analysis, do indeed satisfy the statistical criteria we propose. Thus, while intersubtype recombinant HIV strains are indeed circulating, the criteria for assigning a recombinant origin to viral structures should include statistical testing of alternative hypotheses to avoid inappropriate assignments that would obscure the true evolutionary properties of these viruses.

Consistent viral evolutionary changes associated with the progression of human immunodeficiency virus type 1 infection.

To understand the high variability of the asymptomatic interval between primary human immunodeficiency virus type 1 (HIV-1) infection and the development of AIDS, we studied the evolution of the C2-V5 region of the HIV-1 env gene and of T-cell subsets in nine men with a moderate or slow rate of disease progression. They were monitored from the time of seroconversion for a period of 6 to 12 years until the development of advanced disease in seven men. Based on the analysis of viral divergence from the founder strain, viral population diversity within sequential time points, and the outgrowth of viruses capable of utilizing the CXCR4 receptor (X4 viruses), the existence of three distinct phases within the asymptomatic interval is suggested: an early phase of variable duration during which linear increases ( approximately 1% per year) in both divergence and diversity were observed; an intermediate phase lasting an average of 1.8 years, characterized by a continued increase in divergence but with stabilization or decline in diversity; and a late phase characterized by a slowdown or stabilization of divergence and continued stability or decline in diversity. X4 variants emerged around the time of the early- to intermediate-phase transition and then achieved peak representation and began a decline around the transition between the intermediate and late phases. The late-phase transition was also associated with failure of T-cell homeostasis (defined by a downward inflection in CD3(+) T cells) and decline of CD4(+) T cells to

Presence of multiple HIV subtypes and a high frequency of subtype chimeric viruses in heterosexually infected women.

The HIV-1 subtype distribution was determined in 41 HIV-positive women (-8% of all HIV-infected women in Denmark) belonging to different risk groups. HIV p17 gag and env gene subtypes were determined by DNA sequence analysis. Five different HIV subtypes were detected across all patients. Most HIV-1-positive women of Danish origin carried subtype B viruses, and a minority had virus belonging to subtypes A or C. All injecting drug users (IDUs) were infected with HIV subtype B viruses, whereas all non-B subtypes were present in cases linked to heterosexual transmission. In contrast, all women of African origin carried non-B HIV subtypes (subtypes A, C, D, or G) regardless of transmission mode. Of these women, 21% infected with non-B HIV subtypes appeared to be infected by subtype chimeric viruses and 7% were jointly infected with viruses belonging to two different subtypes (A and C). Data demonstrate a preferential representation of non-B HIV subtypes in women infected through heterosexual contact, as well as a high degree of recombination between viruses derived from endemic areas in which several HIV subtypes predominate. Combined with the increased incidence of heterosexual transmission of HIV, the results imply that an increased subtype diversity can be anticipated in newly infected individuals.

Evolution of envelope sequences from the genital tract and peripheral blood of women infected with clade A human immunodeficiency virus type 1.

The development of viral diversity during the course of human immunodeficiency virus type 1 (HIV-1) infection may significantly influence viral pathogenesis. The paradigm for HIV-1 evolution is based primarily on studies of male cohorts in which individuals were presumably infected with a single virus variant of subtype B HIV-1. In this study, we evaluated virus evolution based on sequence information of the V1, V2, and V3 portions of HIV-1 clade A envelope genes obtained from peripheral blood and cervical secretions of three women with genetically heterogeneous viral populations near seroconversion. At the first sample following seroconversion, the number of nonsynonymous substitutions per potential nonsynonymous site (dn) significantly exceeded substitutions at potential synonymous sites (ds) in plasma viral sequences from all individuals. Generally, values of dn remained higher than values of ds as sequences from blood or mucosa evolved. Mutations affected each of the three variable regions of the envelope gene differently; insertions and deletions dominated changes in V1, substitutions involving charged amino acids occurred in V2, and sequential replacement of amino acids over time at a small subset of positions distinguished V3. The relationship among envelope nucleotide sequences obtained from peripheral blood mononuclear cells, plasma, and cervical secretions was evaluated for each individual by both phylogenetic and phenetic analyses. In all subjects, sequences from within each tissue compartment were more closely related to each other than to sequences from other tissues (phylogenetic tissue compartmentalization). At time points after seroconversion in two individuals, there was also greater genetic identity among sequences from the same tissue compartment than among sequences from different tissue compartments (phenetic tissue compartmentalization). Over time, temporal phylogenetic and phenetic structure was detectable in mucosal and plasma viral samples from all three women, suggesting a continual process of migration of one or a few infected cells into each compartment followed by localized expansion and evolution of that population.

Genetic evaluation of suspected cases of transient HIV-1 infection of infants.

Detection of human immunodeficiency virus-type 1 (HIV-1) on only one or a few occasions in infants born to infected mothers has been interpreted to indicate that infection may be transient rather than persistent. Forty-two cases of suspected transient HIV-1 viremia among 1562 perinatally exposed seroreverting infants and one mother were reanalyzed. HIV-1 env sequences were not found in specimens from 20; in specimens from 6, somatic genetic analysis revealed that specimens were mistakenly attributed to an infant; and in specimens from 17, phylogenetic analysis failed to demonstrate the expected linkage between the infant's and the mother's virus. These findings argue that transient HIV-1 infection, if it exists, will only rarely be satisfactorily documented.

Evolution of human immunodeficiency virus type 1 envelope sequences in infected individuals with differing disease progression profiles.

Sequence variation displayed by the human immunodeficiency virus type 1 (HIV-1) has been proposed to be linked to the pathogenesis of acquired immunodeficiency syndrome (AIDS). To assess viral evolution during the course of infection, we evaluated sequence variability in the env variable domains in four HIV-1-infected individuals exhibiting differing profiles of CD4+ T cell decline when followed from seroconversion until the development of AIDS or loss of followup. Proviral sequences encoding the V3-V5 region of gp 120 were obtained following PCR amplification of peripheral blood mononuclear cell DNA and cloning. Virus in each patient was relatively homogeneous early in infection and then diverged with time, more consistently at its nonsynonymous sites. Just prior to or coincident with a rapid decline in CD4+ T cell numbers, sequences were found with basic amino acid substitutions clustered within and downstream of the gp 120 V3 domain. Within the constraints of the current data set, we conclude that the virus appears to continually accumulate changes in its amino acid sequences well into the time of marked CD4+ T cell decline.

Genetic subtypes of HIV-1 in the Philippines.

OBJECTIVES: To determine the genetic variability of HIV-1 amongst infected Filipinos and to analyze phylogenetic relationships, temporal introductions and transmission dynamics of identified variants. METHODS: Polymerase chain reaction amplification and direct sequencing of a 204 base-pair fragment of the env C2-V3 region from uncultured peripheral blood mononuclear cells obtained from 51 HIV-1-positive Filipinos infected from 1987 to mid-1996. Evolutionary distance and phylogenetic relationships among the DNA sequences were estimated. RESULTS: The 51 Philippine strains were classified into five env V3 subtypes, namely subtype B (n = 37), subtype E (n = 8), subtype A (n = 3), subtype C (n = 2) and subtype D (n = 1). The overall env nucleotide divergence ranged from 11.7 to 32.2%. The nucleotide variation appeared to be random and no temporal ordering was observed. The variation of the sequences at the tip of the V3 loop was very broad. Subtypes B and C isolates did not show close genetic relationship to other Asian variants. Only three of the subtype E strains had close affinity to known Asian sequences. The majority (94%) of the subjects acquired the infection by sexual transmission. About two-thirds were presumably infected outside the Philippines, whereas the remaining were infected indigenously. Information was limited to allow segregation of the identified subtypes by mode of transmission or risk groups. CONCLUSION: Our findings demonstrate the presence of multiple genetic subtypes of HIV-1 in the Philippines. The apparent geographic range of previously reported genotypes in South and South-east Asia was extended and has obvious implications for env-based antiviral interventions.

Evolution of a simian immunodeficiency virus pathogen.

Analysis of disease induction by simian immunodeficiency viruses (SIV) in macaques was initially hampered by a lack of molecularly defined pathogenic strains. The first molecularly cloned SIV strains inoculated into macaques, SIVmacBK28 and SIVmacBK44 (hereafter designated BK28 and BK44, respectively), were cases in point, since they failed to induce disease within 1 year postinoculation in any inoculated animal. Here we report the natural history of infection with BK28 and BK44 in inoculated rhesus macaques and efforts to increase the pathogenicity of BK28 through genetic manipulation and in vivo passage. BK44 infection resulted in no disease in four animals infected for more than 7 years, whereas BK28 induced disease in less than half of animals monitored for up to 7 years. Elongation of the BK28 transmembrane protein (TM) coding sequence truncated by prior passage in human cells marginally increased pathogenicity, with two of four animals dying in the third year and one dying in the seventh year of infection. Modification of the BK28 long terminal repeat to include four consensus nuclear factor SP1 and two consensus NF-kappaB binding sites enhanced early virus replication without augmenting pathogenicity. In contrast, in vivo passage of BK28 from the first animal to die from immunodeficiency disease (1.5 years after infection) resulted in a consistently pathogenic strain and a 50% survival time of about 1.3 years, thus corresponding to one of the most pathogenic SIV strains identified to date. To determine whether the diverse viral quasispecies that evolved during in vivo passage was required for pathogenicity or whether a more virulent virus variant had evolved, we generated a molecular clone composed of the 3' half of the viral genome derived from the in vivo-passaged virus (H824) fused with the 5' half of the BK28 genome. Kinetics of disease induction with this cloned virus (BK28/H824) were similar to those with the in vivo-passaged virus, with four of five animals surviving less than 1.7 years. Thus, evolution of variants with enhanced pathogenicity can account for the increased pathogenicity of this SIV strain. The genetic changes responsible for this virulent transformation included at most 59 point mutations and 3 length-change mutations. The critical mutations were likely to have been multiple and dispersed, including elongation of the TM and Nef coding sequences; changes in RNA splice donor and acceptor sites, TATA box sites, and Sp1 sites; multiple changes in the V2 region of SU, including a consensus neutralization epitope; and five new N-linked glycosylation sites in SU.

Human immunodeficiency virus type 1 populations in blood and semen.

Transmission of human immunodeficiency virus type 1 (HIV-1) usually results in outgrowth of viruses with macrophage-tropic phenotype and consensus non-syncytium-inducing (NSI) V3 loop sequences, despite the presence of virus with broader host range and the syncytium-inducing (SI) phenotype in the blood of many donors. We examined proviruses in contemporaneous peripheral blood mononuclear cells (PBMC) and non-spermatozoal semen mononuclear cells (NSMC) of five HIV-1-infected individuals to determine if this preferential outgrowth could be due to compartmentalization and thus preferential transmission of viruses of the NSI phenotype from the male genital tract. Phylogenetic reconstructions of approximately 700-bp sequences covering the second constant region through the fifth variable region (C2 to V5) of the viral envelope gene revealed distinct variant populations in the blood versus the semen in two patients with AIDS and in one asymptomatic individual (patient 613), whereas similar variant populations were found in both compartments in two other asymptomatic individuals. Variants with amino acids in the V3 loop that predict the SI phenotype were found in both AIDS patients and in patient 613; however, the distribution of these variants between the two compartments was not consistent. SI variants were found only in the PBMC of one AIDS patient but only in the NSMC of the other, while they were found in both compartments in patient 613. It is therefore unlikely that restriction of SI variants from the male genital tract accounts for the observed NSI transmission bias. Furthermore, no evidence for a semen-specific signature amino acid sequence was detected.

Genetic diversity of feline immunodeficiency virus: dual infection, recombination, and distinct evolutionary rates among envelope sequence clades.

For the rapid genetic analysis of feline immunodeficiency virus (FIV), we developed a heteroduplex mobility assay (HMA) that utilizes a PCR-amplified fragment of the FIV envelope gene spanning the third and fourth variable regions of the envelope surface protein coding sequence. Viral sequences were successfully amplified from blood specimens from 98 naturally infected cats from Australia, Canada, Germany, Italy, South Africa, and the United States. Eighty were clearly assignable to the A or B envelope sequence subtypes. Three belonged to subtype C, one was dually infected with viruses harboring the A and B env subtypes, and several were categorized as outliers to any of the established subtypes or as probable intersubtype recombinants. Some geographic clustering was evident, with subtypes A and B found in greater frequency in the western and eastern regions of the United States, respectively. Subtypes A, B, and C were found on more than one continent, and countries with more than two samples analyzed contained at least two subtypes. The broadest representation of subtypes was found in Munich, Germany, where three subtypes and one virus that was not classifiable by HMA were found. Thirteen samples were selected for DNA sequence determination over the same region of env used for HMA. Analysis of all available FIV env sequences from this and previous studies revealed the existence of recombinant viruses generated from subtype A/B, B/D, and A/C envelope gene sequences. Subtype A env sequences were less diverse than subtype B sequences, although both groups had well-supported clusters. Furthermore, the mutational pattern giving rise to diversification in the two subtypes differed, with the subtype A viruses showing half as many synonymous site mutations compared to subtype B yet showing similar levels of nonsynonymous site changes. These results are consistent with the hypothesis that FIV-B is an older virus group and is possibly more host adapted than FIV-A.

Mother-to-infant transmission of human immunodeficiency virus type 1 involving five envelope sequence subtypes.

Genetic analysis of human immunodeficiency virus type 1 (HIV-1) from cases of mother-to-infant transmission were analyzed in an effort to provide insights into the viral selection that may occur during transmission, as well as the timing and source of transmitted viruses. HIV-1 env genes obtained from seven mothers and their perinatally infected infants in Sweden were studied. Five envelope sequence clades (A to E) were found to be represented. We used a heteroduplex tracking assay (HTA) to assess the genetic relatedness between early viral isolates from the infants and serial maternal virus populations taken during pregnancy and at delivery. HTA findings were used to select for DNA sequence analysis maternal virus populations that were either closely or more distantly related to the infant virus. In each case, nucleotide sequence analysis confirmed the genetic relationships inferred by the HTA. Only maternal peripheral blood was sampled, and large sets of maternal specimens throughout pregnancy were generally not available. However, no consistent correlation was found to support the hypothesis that infant viruses should match blood-derived maternal virus genotypes found early in pregnancy if infants were found to be infected at birth or, conversely, that infant viruses should match blood-derived maternal virus genotypes found at delivery if infants were found to be infected only some time later.

Intrapatient sequence variation of the gag gene of human immunodeficiency virus type 1 plasma virions.

Because certain regions of the gag gene, such as p24, are highly conserved among human immunodeficiency virus (HIV) isolates, many therapeutic strategies have been directed at gag gene targets. Although intrapatient variation of segments of gag have been determined, little is known about the variability of the full-length gag gene for HIV isolated from a single individual. To evaluate intrapatient full-length gag variability, we derived the nucleotide sequences of at least 10 cDNA gag clones of virion RNA isolated from plasma for each of four asymptomatic HIV type 1-infected patients with relatively high CD4+ T-cell counts (300 to 450 cells per mm3). Mean values of intrapatient gag nucleotide variation obtained by pairwise comparisons ranged from 0.55 to 2.86%. For three subjects, this value was equivalent to that reported for intrapatient full-length env variation. The greatest range of intrapatient mean nucleotide variation for individual protein-coding regions was observed for p7. We did not detect any G-to-A hypermutation, as A-to-G and G-to-A transitions occurred at similar frequencies, accounting for 29 and 25%, respectively, of the changes. Mean variation values and phylogenetic analysis suggested that the extent of nucleotide variation correlated with the length of viral infection. Furthermore, no distinct subpopulations of quasispecies were detectable within an individual. The predicted amino acid sequences indicated that there were no regions within a gag protein that were comprised of clustered changes.

Distribution of HIV-1 subtypes seen in an AIDS clinic in Sao Paulo City, Brazil.

OBJECTIVE: To determine the distribution of HIV-1 subtypes in Sao Paulo, Brazil. METHODS: Samples were obtained from 80 consecutive HIV-1-infected individuals attending the Immunodeficiency Clinic at the University of Sao Paulo in 1993. Peripheral blood mononuclear cells (PBMC) were separated by Ficoll-Hypaque gradient and a portion was used for routine CD4 counts; the remainder were frozen. PBMC were proteinase-K-digested and DNA-purified by organic extraction. Samples were amplified for the env region of HIV, and envelope sequence subtypes determined by heteroduplex mobility analysis using prototypic subtypes as references. A subset of these were also sequenced through the C2-V3 region of env. RESULTS: A total 69 of 80 samples yielded env polymerase chain reaction product enabling subtype determination; samples that did not amplify were those with low DNA yields. Among 12 injecting drug users (IDU) or sexual partners of IDU, four were typed as clade F and eight as clade B. Forty-three homosexual men or female sexual partners of bisexual men were typed as clade B. The 14 additional cases without known risk factors were typed as clade B. CONCLUSION: These data suggest that subtype F is related to injecting drug use in Brazil.

PIP: Serum samples from 80 consecutive HIV-1-infected individuals presenting to the Immunodeficiency Clinic at the University of Sao Paulo in 1993 were analyzed to determine the distribution of HIV-1 subtypes in the city. Peripheral blood mononuclear cells (PBMC) were separated using Ficoll-Hypaque gradient, a portion was used for routine CD4 counts, and the rest were frozen. PBMC were proteinase-K-digested and DNA-purified by organic extraction. The samples were amplified for the env region of HIV, and envelope sequence subtypes determined by heteroduplex mobility analysis using prototypic subtypes as references. A subset was also sequenced through the C2-V3 region of env. 69 samples yielded env polymerase chain reaction product enabling subtype determination. The samples which did not amplify had low DNA yields. Among 12 IV-drug users or their sex partners, four were typed as clade F and eight as clade B. 43 homosexual men or female sex partners of bisexual men were typed as clade B. The 14 additional cases with no known risk factor were typed as clade B.

Maintaining the integrity of human immunodeficiency virus sequence databases.

Human immunodeficiency virus type 1 (HIV-1) sequences are accumulating in the literature at a rapid pace. For this ever-expanding resource to be maximally useful, it is critical that researchers strive to maintain a high level of quality assurance, both in experimental design and conduct and in analyses. Here we present detailed analyses of problematic sets of HIV-1 sequences in the database that include sequence anomalies suggestive of mislabeling or sample contamination problems. These data are examined in the context of currently available HIV-1 sequence information to provide an example of how to identify potentially flawed data. Indicators of potential problems with sequences are (i) sequences that are nearly identical that are supposed to be derived from unlinked individuals and that are markedly distinct from other sequences from the putative source or (ii) sequences that are nearly identical to those of laboratory strains. We provide an outline of methods that researchers can use to perform preliminary laboratory and computational analyses that could help identify problematic data and thus help ensure the integrity of sequence databases.

Evolution of the chalcone synthase gene family in the genus Ipomoea.

The evolution of the chalcone synthase [CHS; malonyl-CoA:4-coumaroyl-CoA malonyltransferase (cyclizing), EC 2.3.1.74] multigene family in the genus Ipomoea is explored. Thirteen CHS genes from seven Ipomoea species (family Convolvulaceae) were sequenced--three from genomic clones and the remainder from PCR amplification with primers designed from the 5' flanking region and the end of the 3' coding region of Ipomoea purpurea Roth. Analysis of the data indicates a duplication of CHS that predates the divergence of the Ipomoea species in this study. The Ipomoea CHS genes are among the most rapidly evolving of the CHS genes sequenced to date. The CHS genes in this study are most closely related to the Petunia CHS-B gene, which is also rapidly evolving and highly divergent from the rest of the Petunia CHS sequences.