AIDS RESEARCH AND HUMAN RETROVIRUSES
Volume 21, Number 5, 2005, pp. 371–378
© Mary Ann Liebert, Inc.
The X4 Phenotype of HIV Type 1 Evolves from R5
in Two Children of Mothers, Carrying X4, and Is Not
Linked to Transmission
P. CLEVESTIG,1I. MALJKOVIC,1,2C. CASPER,1,3E. CARLENOR,1S. LINDGREN,4L. NAVÉR,5
A.-B. BOHLIN,5E.M. FENYÖ,6T. LEITNER,7and A. EHRNST1,8
Previously, we found that emergence of the X4 viral phenotype in HIV-1-infected children was related to the
presence of X4 in their mothers (C.H. Casper et al., J Infect Dis 2002; 186:914–921). Here, we investigated
the origin of the X4 phenotype in the child, analyzing two mother–child pairs (Ma–Ca, Mb–Cb) where the
mothers carried X4 and their children developed X4 after an initial presence of R5. We used nested poly-
merase chain reaction of the env V3 region to generate 203 HIV-1 clones for sequencing (Ma, n ? 44; Ca,
n ? 73; Mb, n ? 61; Cb, n ? 25) from DNA of peripheral blood mononuclear cell (PBMC) lysates, altogether
167 clones, or from cDNA of plasma RNA, 36 clones. PBMC and plasma isolate sequences from each time
point enabled us to assign the probable phenotype to clone sequences in a phylogenetic tree. The transmis-
sion and evolution were reconstructed using the maximum likelihood method. In mother–child pair Ma–Ca,
one maternal R5 isolate clustered with the child’s R5 sequences, at the earliest time when R5 was isolated in
the child, confirming this as a likely source of the transmitted R5 phenotype. At age 3, an X4 population was
present in the child that had evolved from the child’s own R5-associated population, clearly distinct from the
maternal X4 sequences. The second mother–child pair (Mb–Cb) displayed a similar pattern. Amino acid sub-
stitution patterns corroborated the conclusions from the phylogenetic tree. Thus, in both children, the X4
virus developed from their own R5 population, and was not caused by transmission of X4.
coreceptors CCR5 and CXCR4.1The coreceptor use classifies the
virus into the R5 or X4 phenotype.2HIV-1 can also be classified
as dual-tropic, using both coreceptors.2
The variable region 3 (V3) of the gp120 envelope glyco-
protein of HIV-1 has been identified as a critical determinant
for coreceptor use.3,4Thus, the V3 loop of X4 virus sequences
has been observed to have a greater amino acid net charge than
IV-1 INFECTS CELLS BY USING THE T CELL DIFFERENTIATION
ANTIGENCD4 as receptor. In addition, HIV-1 uses chemokine
the V3 loop of R5 viruses,5and the presence of positive sig-
nature amino acids has been linked to syncytium induction.4
A link between the progression of the disease and corecep-
tor use in children5and adults has been reported.6We have
found that the emergence of X4 in children occurred after de-
velopment of the immune deficiency.7Also, the emergence of
X4 phenotypes in the children was linked to the presence of X4
in their mothers.8Here, we investigated whether the X4 phe-
notype in the children was transmitted from the mother and de-
layed in appearance until the immune system deteriorated, or
if it evolved independently.
1Microbiology and Tumor Biology Center, Karolinska Institutet, SE-171 77 Stockholm, Sweden.
2Department of Clinical Virology, The Swedish Institute of Infectious Disease Control, SE-172 82 Solna, Sweden.
3Unit of Neonatal Health, Department of Women’s and Children’s Health, Karolinska Institutet, SE-171 77 Stockholm, Sweden.
4Division of Obstetrics and Gynecology and 5Division of Pediatrics, Department of Clinical Science, Karolinska University Hospital Hud-
dinge, Karolinska Institutet, SE-171 77 Stockholm, Sweden.
6Department of Medical Microbiology, Dermatology, and Infection, Lund University Hospital, Lund, Sweden.
7Theoretical Biology and Biophysics Group, Los Alamos National Laboratory, Los Alamos, New Mexico 87545.
8Department of Laboratory Medicine, Division of Clinical Virology, Karolinska University Hospital Huddinge, Karolinska Institutet, SE-171
77 Stockholm, Sweden.
MATERIALS AND METHODS
Subjects and samples
The two mother–child pairs, Ma–Ca and Mb–Cb, have both
been described in detail previously.7–9One mother (Ma) came
from sub-Saharan Africa and carried HIV-1 subtype D and the
other (Mb) came from Southeast Asia and was infected with
HIV-1 CRF01_AE. The mothers had not received preventive
antiretroviral therapy during pregnancy. Defibrinated blood sam-
ples, in heparin before 1990, or EDTA after,9were obtained dur-
ing different trimesters of pregnancy, at delivery, and 6 months
later. Both mothers denied have breast-feeding their infants.
Both children had samples taken at birth, in which HIV was
not detected by virus isolation from peripheral blood mononu-
clear cells (PBMC), plasma, and by polymerase chain reaction
(PCR) on DNA. Thus, both children were most likely infected
around the time of birth. They were followed to the age of 5
and 6.5 years, respectively. RNA was investigated retrospec-
Sample preparation, virus isolation, determination
of coreceptor use, and cDNA synthesis
PBMC and plasma were used for virus isolation.11Core-
ceptor use of virus isolates was tested on engineered cell lines:
U87.CD4 cells, expressing chemokine receptors CCR5 or
CXCR4, and GHOST(3) cells, expressing CCR5 or
CXCR4.1,7,8For preparation of DNA, thawed lymphocytes
were subjected to lysis buffer, containing proteinase K, and de-
tergent. Plasma was frozen at ?70°C. RNA was extracted by
the NASBA/NucliSens technique (Organon Teknika, Boxtel,
Holland). For cDNA synthesis (Applied Biosystems, Foster
City, CA) the primer JA170: GTG ATG TAT T(A/G)C
A(A/G)T AGA AAA ATT C (MedProbe, Oslo, Norway) was
Nested polymerase chain reaction
The HIV-1 env V3 region was amplified in a nested PCR12
with outer primers JA170 and JA167: TAT C(C/T)T TTG AGC
CAA TTC C(C/T)A TAC A (MedProbe, Oslo, Norway).12In-
ner primers were JA168: ACA ATG (C/T)AC ACA TGG AAT
TA(A/G) GCC A and JA169: AGA AAA ATT C(C/T)C CTC
(C/T)AC AAT TAA A (MedProbe, Oslo, Norway). The am-
plified DNA was purified using QIAquick PCR purification kit
(QUIAGEN GmbH, Hilden, Germany).
To obtain V3 clones, a limiting dilution PCR was performed,
followed by PCR on multiple replicas (n ? 54) for each sam-
ple of DNA and HIV-1 RNA, respectively. In the case when
only one nucleotide ambiguity occurred, the possible sequences
were designated a and b, respectively.
Cycle sequencing PCR was performed, using the Big Dye
Terminator Cycle Sequencing Ready Reaction Kit (Applied
Biosystems, Foster City, CA). The inner primers JA168 and
JA169 (0.8 pmol/?l) were used. The amplified DNA was puri-
fied, using Sephadex G-50 Superfine filtering gel (Amersham
Pharmacia Biotech, Sweden) in a MultiScreen 96-well filtration
plate (Millipore Corporation, Bedford, MA). The ABI Prism
3100 DNA capillary sequencer was used. The sequence chro-
matograms were verified (Sequencher software, Genecodes,
Ann Arbor, MI) and aligned using the program Se-Al.13
Hypermutations in mother and child
So that the natural occurrence of G ? A hypermutations in
HIV-114would not affect our phylogenetic analysis, these were
identified using Hypermut (http://www.hiv-web.lanl.gov). Ab-
normal G ? A mutations (#G ? A ? 10), normal value 0–5,
which were observed in the first case (Ma: n ? 3; Ca: n ? 5),
Phylogenetic analysis of biological clones
and viral isolates
Phylogenetic trees were inferred using a maximum likeli-
hood method under the F84 substitution model and a parallel
computing system16(T. Leitner et al., unpublished observa-
tions). The branch lengths were recalculated under a general-
time-reversible model with ?-distributed rates among sites
(? ? 0.3).17,18Each tree was rooted to selected maternal se-
quences to present a relation to genetic evolution over time. The
robustness of our trees was tested by bootstrap analysis with
The binomial test and the Fisher’s exact test were used.
The Ethics Committee of the Karolinska Institutet, Stock-
holm, Sweden had approved of the study with registration num-
bers Dnr 1989:246 and Dnr 1998:184.
Overview of the samples used in the
An overview of each mother–child pair is displayed in Table
1. The first available isolate from child Ca was the only isolate
of the R5 phenotype. Therefore the phenotype was determined
on two separate subcultures, collected 14 days apart from the
original isolation, and processed at different times.
In the child (Cb), virus isolation was positive in plasma at 6
months of age. The isolate failed to grow, but PBMCs were
available for amplification of clones. There was an association
of phenotype change to disease progression in both children,
but the change occurred after the immune deficiency had been
established, as reported previously.7
R5 in the child was linked to the maternal R5
The relation between maternal sequences and the sequences
from the child’s first available HIV-1-positive sample is shown
in Fig. 1A. The phylogenetic tree analysis showed a distinct
group of clones, containing the only R5 isolate identified in the
CLEVESTIG ET AL.
mother derived from plasma at delivery. The other maternal iso-
lates were X4 and spread among the remaining clone sequences
without an evident time linearity. The clones from the child
grouped close to its R5 isolates as well as with the maternal
R5-associated cluster. This provided a strong link between the
maternal R5 viruses and the transmitted R5 virus.
Since the population of R5-associated clones was in a mi-
nority, the question of whether there was a selective transmis-
sion of R5 virus was tested. Assuming that the R5 and X4 clones
had an equal chance of being infectious and all also being in-
fectious, the binomial chance of transmission of an R5 virus in
a population dominated by X4, 10 versus 32, was unlikely,
p ? 0.025. However, these assumptions may not correspond
sufficiently well to reality to rule out chance.
The X4 population in the child was not derived
from the maternal X4 population
The analysis of the maternal sequences with the second
time point in the child (Ca) (Fig. 1B) showed that the child’s
X4-associated clones were genetically distinct from the ma-
ternal X4-associated populations. The branching point of the
child’s cluster of X4-associated sequences had its origin
within the maternal R5-associated population, suggesting that
the X4 population in the child was derived from the child’s
Figure 2 displays all sequences of the child (Ca) in one tree.
Distinct subpopulations consisted of all R5 clones from the first
time point, an R5(X4) population appearing later, and the largest
cluster. This was assigned an X4-associated phenotype, because
five X4 isolates grouped within it. The branching point of the
child’s genetically well-defined cluster of X4-associated sequences
had its origin within the child’s R5-associated population.
Similar outcome in the mother–child pair (Mb–Cb)
In mother–child pair Mb–Cb, we observed a similar pattern
of evolution from R5 to X4 after the initial appearance of R5.
As we had fewer clones, all sequences were combined into one
phylogenetic tree (Fig. 3).
The three maternal X4 isolate sequences grouped together
with 14 clone sequences, assigned an X4-associated population.
The remaining maternal sequences were scattered without a
probable phenotype association nor any obvious time linearity.
The sequences from the first time point in the child formed
a homogeneous cluster together with its R5 isolate, determin-
ing it to be an R5 population. We also had one R5X4 isolate
from 4 years and one X4 isolate from 6.5 years of age. These
protruded from a branching point in the tree just before the R5
cluster. The genetic distance after the branching point was small
(negligible), which was consistent with the fact that the change
in coreceptor use occurred in the child.
X4 EVOLUTION OF HIV-1 IN MOTHER–CHILD PAIRS
TABLE 1.SAMPLE CHARACTERISTICS OF THE MOTHER–CHILD PAIRS
Case Time pointcopies/mlPBMC Plasma PBMC Plasma
Mother a (Ma)
1 (second trimester)
2 (third trimester)
4 (6 months later)
Child a (Ca)
0 (age 6 months)
1 (1 year)
2 (3 years)
3 (4 years)
4 (5 years)
R5 (n ? 2)
Mother b (Mb)
1 (second trimester)
3 (6 months later)
Child b (Cb)
1 (6 months)
2 (1 year)
3 (4 years)
4 (6.5 years)
aThis isolate did not fulfill all criteria for CXCR4 use. There was an increase in p24 antigen between days 1 and 7, but a cyto-
pathic effect was apparent only on day 7 for the cell line U87.CD4.CXCR4. The results were negative for the GHOST(3).CXCR4.7
In the child (Ca), samples after birth were taken at 3 months and 6 months of age, but in none of them was HIV-1 detected.31
In retrospective analysis HIV-1 RNA was detected at 6 months of age.10HIV was first isolated at 8 months of age, but this iso-
late failed to grow out.7
Due to the presence of hypermutations, there are more sequences than in the subsequent phylogenetic trees. There is one more
sequence from maternal Ma time point 1 and there are two more from the second, and from the child Ca there are two more from
the first time point, two from the second, and one from the fourth.
Viral phenotype in
No. of clones in
Amino acid patterns of the V3 sequences of the
mother–child pair Ma–Ca
For a better evaluation, we compared the amino acid patterns
of the V3 loop (Fig. 4A). There were major differences between
the maternal (Ma) X4 isolate sequences and the X4 isolate se-
quences in the child (Ca), which showed that there had been a
separate evolution of X4 in the child from the mother. Only one
amino acid substitution, an isoleucine (I) at position 22, was
shared between the maternal and the child’s X4 isolates. It was
also present in one R5 clone each from mother and child. The
phylogenetic tree (Fig. 2) showed that it was present in the ma-
ternal R5 clone furthest away from the child’s sequences. Thus
it did not contribute to the overall phylogeny, i.e., it was a ho-
mology. In the child, this R5 clone was one of the most distant
sequences. Thus, this substitution most likely had occurred af-
ter transmission. An additional observation was the presence of
the N-linked glycosylation motif in the maternal X4 isolates,
but this was absent in the child’s X4, providing further evidence
of dissimilarities between the two X4 viral populations.
Amino acid patterns of the V3 sequences of the
mother–child pair Mb–Cb
In the mother–child pair (Mb–Cb), the maternal X4 isolates
and all maternal X4-associated clones (n ? 14) had arginine
(R) in position 18, while of the remaining clones (n ? 46), only
one had arginine and 45 had glutamine (Q), p ? 1.7 ? 10?12.
Similarly, while all of the latter maternal clones and all of the
child’s sequences had alanine (A) in the following position, the
maternal X4 isolates and associated clones had valine (V), p ?
3.1 ? 10?13. This suggested that the non-X4-associated clones
from the mother were R5-associated. The clones closest to the
child were identical to 24 of 25 of the child’s R5-associated
clones (Fig. 4B), making the R5-association of those maternal
clones undisputable. This corroborated the observation that
CLEVESTIG ET AL.
ternal (Ma) tree with the sequences from the child (Ca) from the first time point, when only R5 virus was detected. When omit-
ting the middle sequences M3.23D and M1.5D, the bootstrap value was recalculated to be 69%. (B) Maternal (Ma) tree with the
sequences of the child from the second time point, when X4 virus was detected for the first time. When the clones M3.23D and
M1.5D were removed from the R5 cluster, the bootstrap value increased to 74%. The isolates are designated ci, c for cellular
and i for isolate), when derived from PBMC and pi when derived from plasma. R5 and/or X4 ? phenotype of isolates. All clones
were derived from PBMC, which was marked by a D for DNA. Clones with the same clone number, but designated a or b, dif-
fer by one nucleotide. Phenotype association of a cluster is shown by a bracket. % ? Bootstrap percentage value, which is shown
for the major branch between the R5-associated population and the X4-associated populations. C1 and C2 ? the two first time
points of the child. All clones were derived from PBMC, marked by a D.
Phylogenetic tree, showing the relationship between maternal (Ma) sequences and sequences of the child (Ca). (A) Ma-
there was no linkage of the X4 appearing in the child to ma-
ternal X4 and that the R5 in the child was derived from a ma-
ternal R5 population. Both maternal and child X4 isolate
sequences lacked an intact N-linked glycosylation motif. How-
ever, the substituting amino acids differed between the two,
which provided further evidence that they were unique X4 vi-
We tracked the origin of the X4 phenotype, following the
initial presence of R5 in two perinatally HIV-1-infected chil-
dren, infected with subtype D and CRF01_AE, respectively. In
both children the X4 phenotype evolved from the R5 virus and
was not transmitted from their mothers.
X4 EVOLUTION OF HIV-1 IN MOTHER–CHILD PAIRS
C4 ? the four time points of the child. All clones were derived from PBMC and marked by a D. Ten clone sequences from the
maternal R5-associated population were included for rooting purposes. For other explanations see Fig. 1.
Phylogenetic tree, showing the internal population dynamics with all clone and isolate sequences from child Ca. C1 to
Comparing sequences from isolates with known phenotypes and
a number of molecular clones at different time points provides a
powerful tool to follow the viral evolution within individuals and
after transmission, and the transmission history between cases.19–21
This allowed us to investigate the relation of viral populations in
mother-to-child transmission, similar to what has been reported
previously in a family infected with CRF01_AE,21,22but not to
our knowledge in cases with subtype D.
In both children, a distinct R5-associated population was dis-
played at 1 year of age, which was the first time point with
available samples here. Despite the time that had elapsed since
birth, both children’s isolates and clones clearly originated from
the maternal R5 (Ma–Ca) or likely R5 (Mb–Cb) clones.
The amino acid alignment gave further evidence of these
results when also considering the topology of the phyloge-
netic tree. The numerous amino acid differences between the
maternal X4 isolates and the X4 isolates from the children
showed that the development of these populations within the
children had occurred independently of the maternal X4 pop-
ulations and were thus unlikely a result of a transmission. In-
terestingly, the N-linked glycosylation motif within the V3
loop differed within each mother–child pair among the X4
Altogether, these data support the general conclusion that X4
developed in the children from the early R5 population. How-
ever, we cannot exclude the possibility of transmission of more
CLEVESTIG ET AL.
points 1–4 in the mother. C1–C4 ? time points 1–4 in the child. R denotes plasma RNA-derived clones. For other explanations
see Fig. 1.
Phylogenetic tree with all clone and isolate sequences from the second mother–child pair (Mb–Cb). M1 to M4 ? time
than one variant, but this cannot have given rise to the major
X4 isolates and associated clones that we observed.
Earlier studies on mother-to-child transmission have sug-
gested that either single or multiple variants are transmit-
ted.23–26Usually, transmission of one variant has been regarded
as a sign of selective transmission, while transmission of mul-
tiple variants has been regarded as reflecting another mode or
mechanism of transmission. Both may well be the result of sto-
chastic processes depending primarily on viral load, or expo-
sure and multiplicity of infection. However, the question may
be asked whether R5 has an advantage in transmission27as also
in one case here, where an R5 virus was transmitted from a pre-
dominant X4 population.
Interestingly, an arginine residue in position 25 among 5 R5-
associated clones from the mother Ma was also present in the
third time point plasma isolate from her child, which was de-
termined as an R5(X4) phenotype. This might imply that trans-
mission of another R5 virus had occurred with or without the
ability to use the CXCR4 coreceptor, but originating from the
maternal R5 population. This may strengthen the case of a se-
lection of R5 in transmission.
In addition, why was X4 isolated when the clones suggested
that R5 was dominant, as in the mother Mb? A possible reason
could be a selective transmission of R5 in vivo, and a preferred
outgrowth of X4 in vitro in the PBMC cultures.28
In both cases the delivery sample was most closely related
to the first virus detected in the child. This corresponds well to
the hypothesis of transmission at delivery for children lacking
signs of the presence of HIV in the blood within the first week
of life,29as was the case in both children here.
It is not easy to explain what links the parallel development of
X4 in the mother and in the child, when it is not related to trans-
mission of the virus. Genetic factors of the hosts might be factors
related to infection control, such as the HLA system. Genetic fac-
tors of the virus might be related to amino acid patterns, predis-
posing for a switch from R5 to X4. If specific amino acid pat-
terns occur, possibly subtype specific, this could point to a
convergent evolution of HIV-1 with regard to coreceptor use.21,30
The GenBank sequence accession numbers for mother–child
pair Ma–Ca was AY 619730 to AY 619853 and for mother–
child pair Mb–Cb AY 619854 to AY 619945.
We want to thank Gun Sundin and Ulla Lips from the
Karolinska University Hospital Huddinge and Kajsa Aperia and
Maj Kroon from the Swedish Institute for Infectious Disease
Control for technical support. Financial support was given by
the Karolinska Institute Research Funds, the Sven Jerring Foun-
X4 EVOLUTION OF HIV-1 IN MOTHER–CHILD PAIRS
R5 isolate sequence (as reference) from the child was aligned with the other child isolate and clone sequences and with mater-
nal isolate and R5-associated clone sequences. The N-linked glycosylation site is underlined. The association of the clones with
a given phenotype is given within parentheses. See Figs. 1 and 2 for clone positions in the phylogenetic trees and
for further explanations of symbols. (B) Alignment of the mother–child pair Mb–Cb, carrying CRF01_AE. The child (Cb) iso-
late and clone sequences and maternal isolate and maternal R5-associated and X4-associated clone sequences, respectively, are
aligned with the child R5 isolate sequence, as reference. See Figs. 1 and 3 for further explanations of symbols and for clone po-
sitions in the phylogenetic tree.
V3 sequence alignment of the mother–child pairs. (A) Alignment of the mother–child Ma–Ca, carrying subtype D. The
dation (to A.E. and P.C. each), the Swedish Research Council, Download full-text
No. 521-2002-6449 (to T.L.) and No. K2003-16VX-14724-01A
(to A.E.), and the Los Alamos National Laboratory, LAUR No.
1. Deng H, Unutmaz D, Kewal-Ramani VN and Littman DR: Ex-
pression cloning of new receptors used by simian and human im-
munodeficiency viruses. Nature 1997;388:296–300.
2. Berger EA, Doms RW, Fenyö EM, et al.: A new classification for
HIV-1. Nature 1998;391:240.
3. Hwang SS, Boyle TJ, Lyerly HK, and Cullen BR: Identification of
the envelope V3 loop as the primary determinant of cell tropism
in HIV-1. Science 1991;253(5015):71–74.
4. De Jong JJ, De Ronde A, Keulen W, Tersmette M, and Goudsmit
J: Minimal requirements for the human immunodeficiency virus
type 1 V3 domain to support the syncytium-inducing phenotype:
Analysis by single amino acid substitution. J Virol 1992;66:
5. Scarlatti G. Tresoldi E, Björndal Å, et al.: In vivo evolution of
HIV-1 co-receptor usage and sensitivity to chemokine-mediated
suppression. Nat Med 1997;3:1259–1265.
6. Connor RI, Sheridan KE, Ceradini D, Choe S, and Landau NR:
Change in coreceptor use correlates with disease progression in
HIV-1-infected individuals. J Exp Med 1997;185:621–628.
7. Casper C, Naver L, Clevestig P, et al.: Coreceptor change appears
after immune deficiency is established in children infected with
different HIV-1 subtypes. AIDS Res Hum Retroviruses 2002;18:
8. Casper CH, Clevestig P, Carlenor E, et al.: Link between the X4
phenotype in human immunodeficiency virus type 1-infected moth-
ers and their children, despite the early presence of R5 virus in the
child. J Infect Dis 2002;186:914–921.
9. Lindgren S, Martin C, Anzen B, Strand H, Bredberg-Raden U, and
Ehrnst A: Pattern of HIV viraemia and CD4 levels in relation to
pregnancy in HIV-1 infected women. Scand J Infect Dis 1996;28:
10. Naver L, Ehrnst A, Belfrage E, et al.: Long-term pattern of HIV-1
RNA load in perinatally infected children. Scand J Infect Dis
11. Ehrnst A, Sönnerborg A, Bergdahl S, and Strannegård O: Efficient
isolation of HIV from plasma during different stages of HIV in-
fection. J Med Virol 1988;26:23–32.
12. Albert J and Fenyö EM: Simple, sensitive and specific detection
of human immunodeficiency virus type 1 in clinical specimens by
polymerase chain reaction with nested primers. J Clin Microbiol
13. Rambaut A: Se-Al: Sequence Alignment Editor. vl.dl. Department
of Zoology, University of Oxford, 1996. Available at http://evolve.
14. Vartanian JP, Meyerhans A, Asjo B, and Wain-Hobson S: Selec-
tion, recombination, and G ? A hypermutation of human immu-
nodeficiency virus type 1 genomes. J Virol 1991;65:1779–1788.
15. Rose PP and Korber BT: Detecting hypermutations in viral se-
quences with an emphasis on G ? A hypermutation. Bioinfor-
16. Felsenstein J: PHYLIP: Phylogenetic Inference Package v3.52c,
1993. University of Washington, Seattle, WA. Available at
17. Swofford DL: PAUP*: Phylogenetic Analysis Using Parsimony (*
and Other Methods). Version 4, 2002, Sinauer Associates, Sun-
18. Leitner T, Kumar S, and Albert J: Tempo and mode of nucleotide
substitutions in gag and env gene fragments in human immunode-
ficiency virus type 1 populations with a known transmission his-
tory. J Virol 1997;71:4761–4770.
19. Leitner T, Escanilla D, Franzen C, Uhlen M, and Albert J: Accu-
rate reconstruction of a known HIV-1 transmission history by phy-
logenetic tree analysis. Proc Natl Acad Sci USA 1996;93:
20. Albert J, Wahlberg J, Leitner T, Escanilla D, and Uhlen M: Anal-
ysis of a rape case by direct sequencing of the human immunode-
ficiency virus type 1 pol and gag genes. J Virol 1994;68:
21. Sato H, Shiino T, Kodaka N, et al.: Evolution and biological char-
acterization of human immunodeficiency virus type 1 subtype E
gp120 V3 sequences following horizontal and vertical virus trans-
mission in a single family. J Virol 1999;73:3551–3559.
22. Kato K, Sato H, and Takebe Y: Role of naturally occurring basic
amino acid substitutions in the human immunodeficiency virus type
1 subtype E envelope V3 loop on viral coreceptor usage and cell
tropism. J Virol 1999;73:5520–5526.
23. Wade CM, Lobidel D, and Brown AJ: Analysis of human immu-
nodeficiency virus type 1 env and gag sequence variants derived
from a mother and two vertically infected children provides evi-
dence for the transmission of multiple sequence variants. J Gen Vi-
24. Wolinsky SM, Wike CM, Korber BT, et al.: Selective transmis-
sion of human immunodeficiency virus type-1 variants from moth-
ers to infants. Science 1992;255:1134–1137.
25. Dickover RE, Garratty EM, Plaeger S, and Bryson YJ: Perinatal
transmission of major, minor, and multiple maternal human im-
munodeficiency virus type 1 variants in utero and intrapartum. J
26. Scarlatti G, Leitner T, Halapi E, et al.: Comparison of variable re-
gion 3 sequences of human immunodeficiency virus type 1 from
infected children with the RNA and DNA sequences of the virus
populations of their mothers. Proc Natl Acad Sci USA
27. Ahmad N, Baroudy BM, Baker RC, and Chappey C: Genetic anal-
ysis of human immunodeficiency virus type 1 envelope V3 region
isolates from mothers and infants after perinatal transmission. J Vi-
28. Koning FA, van Rij RP, and Schuitemaker H: Biological and mo-
lecular aspects of HIV-1 coreceptor usage. Review: HIV Sequence
Compendium, 2002; pp. 24–42. Available at http://www.hiv.lanl.
29. Dunn DT, Brandt CD, Krivine A, et al.: The sensitivity of HIV-1
DNA polymerase chain reaction in the neonatal period and the rel-
ative contributions of intra-uterine and intra-partum transmission.
30. Buckley KA, Li PL, Khimani AH, et al.: Convergent evolution of
SIV env after independent inoculation of rhesus macaques with in-
fectious proviral DNA. Virology 2003;312:470–480.
31. Navér L, Ehrnst A, Belfrage E, et al.: Broad spectrum in time of
detection, primary symptoms, and disease progression in infants
with HIV-1 infection. Eur J Clin Microbiol Infect Dis 2001;20:
Address reprint requests to:
Microbiology and Tumor Biology Center
SE-171 77 Stockholm, Sweden
CLEVESTIG ET AL.