Elucidation of an HIV -1 Transmission from Mother to Child in West Africa by Sequence Analysis

Zbl. Bakt. 284, 307-317 (1996) © Gustav Fischer Verlag, Stuttgart· .lena· New York

Elucidation of an HIV-l Transmission from Mother to Child in West Africa by Sequence Analysis'1PIA KASPERl, ROLF KAISER!, ANNE STEINBECK-KLOSE 2 , BERTFRIED MATZl, and KARL E. SCHNEWEIS 1 1 Institute 2

of Med. Microbiology and Immunology, University of Bonn, Germany General practitioner, Bonn, Germany

Received March 1, 1996 . Accepted March 13, 1996

Summary A pregnant woman living in Germany went to Ghana for several months, where she received 4 blood transfusions. Her newborn child also received one blood transfusion in West Africa. After return to Germany, HIV-l infection was detected in both of them. Serotyping with V3 peptides revealed that the sera reacted only poorly with the subtype B-specific antigens. To investigate whether the child had been infected by vertical or parenteral transmission, we amplified different proviral HIV-l gene segments from samples obtained 1-3 years after infection. Sequence analysis of the hypervariable regions VI and V2 of the proviral env gene was misleading, since the viral population of the mother was highly heterogeneous, whereas only one predominant viral variant was found in the child. In contrast, sequences of the gag pI7 gene and the regulatory genes nef and vif were homogeneous and revealed a very high homology, suggesting that the child had been infected by the mother. This was confirmed by phylogenetic tree analysis showing that sequences of mother and child clustered together and that both were infected by HIV-l subtype A which is common in West Africa. The results suggest that sequence analysis of the hypervariable regions VI and V2 alone can lead to unclear results, especially if not single genomes are analysed but a mixture of quasi-species. It is recommended that investigations into HIV transmission should be based on sequence analysis of several HIV genes.

Introduction Although individual cases in AIDS occurred in residents of West Africa early in the pandemic, the epidemic spread of HIV (human immunodeficiency virus) in this region is a recent one. Unique features of the situation in West Africa include the circulation of both types of HIV (HIV-I and HIV-2) and the great differences in the prevalence of infections in different countries (8).

,. Dedicated to Prof. Dr. H. Brandis on the occasion of his 80 th birthday.

308

P. Kasper et al.

Extreme genetic variability is a characteristic feature of HIV. Phylogenetic analyses have revealed that the globally circulating strains of HIV-1 can be divided into two major groups (M and 0). The M group contains numerous sequence subtypes designated from A to I and displaying a complex geographic distribution (19, 22, 24). Most of the published HIV-l strains from West Africa fall into subtype A and have been isolated in Cote d'Ivoire (14, 22). The worldwide spread of the different HIV subtypes has been studied by sequence analysis, which is also used to study the mode of individual HIV transmissions. Viral populations in seroconverters and short-term-infected individuals appear to be less variable than those found in long-term-infected individuals. Sexual or parenteral HIV infection seems to be initiated by a limited number of viral variants due to selection upon transmission (15,33,35). Recent reports of mother-child transmission pairs have also suggested that either one minor or major virus population of the mother can initiate infection within a child (29, 34). We studied a case of HIV transmission that occurred during an extended stay in West Africa. A pregnant woman from Germany developed acute malaria in Ghana, where she received 4 blood transfusions in the 5th month of her pregnancy (May/June 1991). Her anaemic newborn (born in September 1991) was also treated with one blood transfusion. Before her pregnancy, the mother had proved to be HIV antibodynegative. The HIV- 1 infection of both mother and child was detected after their return to Germany, when the 8 month-old infant developed a candidal infection on his little finger. Since vertical HIV-l transmission is estimated to occur in 13 to 30% of infants born to HIV-l-infected mothers (27), the question arose whether the child had been infected vertically by the mother or parenterally by the blood transfusion. We investigated this case of HIV transmission by serotyping and sequence analysis of several proviral genes of mother and child.

Materials and Methods Patients. Blood samples were obtained from mother and child 1-3 years after infection of the mother (May/June 1991). Both patients were in the asymptomatic phase of HIV infection. Serotyping. Serotyping of sera or plasma samples was done with subtype-specific biotinylated V3 loop peptides (100 ng/mL) previously described by Chengsong-Popov et al. (7). We used a modified Enzymun-Test Anti-HIV 1 + 2 Gen. 2 assay (ELISN2-step sandwich assay based on streptavidin technology, Boehringer Mannheim). A negative control was used to establish the cut-off ratio for each assay which was the triple of the mean value. PCR and sequencing. DNA was extracted from peripheral blood mononuclear cells using a standard protocol. The proviral genes were amplified from 1 f!g PBMC-DNA or after limiting dilution to obtain single molecules (31). Primers used for the amplification of the V1N2 region were SKl22/SK123 (10) or SK 122/env 6 (20). The second PCR was performed with ]P124 (18) and biotinylated SK123. The proviral p 17 gene was amplified as described (35). Primers used for PCR of nef and vif were published by Shugars et al. (30) and Wieland et al. (32), respectively. PCR product for sequencing was generated by nested amplification with 10 pmol biotinylated antisense primer and a sense primer which included the universal M13 sequence. Amplified DNA was immobilized on magnetic beads (15) and sequenced with fluorescent M13 primers using the Autoread™ Sequencing Kit and the ALF sequencer according to the instructions of the manufacturer (Pharmacia LKB).

HIV Transmission from Mother to Child

309

Sequence data. Sequence alignments and tree analysis were carried out using ClustalV (11). Genetic distances were also calculated with this program and corrected by Kimura's empirical method. All sites with gaps were deleted. Phylogenetic trees were constructed with the neighbour-joining method. In order to assess the relationship of mother and child sequences to other HIV strains, a background population of 13 published HIV-l sequences was also included (MAL, CI51, CI4, VI525, VI557, VI191, DJ259, UG268, VI174, BZ162, BZ200, UG274 and VI203). The published HIV isolates belonged to different HIV-l subtypes and were collected in geographic areas very remote from each other (22). Bootstrapping (1000 replicates) was done to consider if the groupings shown in the tree were statistically significant.

Results Sequence analysis of the VlIV2 region with primers SK 122/SKl23. First we analysed the hypervariable regions V1 and V2 of the proviral env gene from mother and child. The samples of mother and child were obtained 21 and 12 months after the infection of the mother. We used primers SK122/SK123 for the first PCR and SK 123/JP124 for the second amplification. The products of the second PCR were analysed by gel electrophoresis, which showed that the peR product from the child was smaller than the amplified fragment from the mother. Direct sequencing of both products resulted in clear sequences with no background. The length variation shown by gel electrophoresis was also verified by the sequencing method. The PCR product from the child had a length of 260 bp, whereas the amplified V1N2 region from the mother one of 300 bp. Moreover, the proviral sequences of mother and child were very different, resulting in a genetic distance of 42.3 % in this area. Moreover, the predominant proviral variant of the mother displayed a large insertion with respect to the proviral sequence of the child (Fig. 1A). This preliminary result suggested that the predominant V1N2 sequences of mother and child were not related and that the child had probably not been infected by the mother. Sequence analysis of regulatory genes and the p 17 region. To confirm this result we analysed further proviral genes from the samples of mother and child. The amplification of the regulatory genes nef and vif resulted in fragments of the same length for mother and child. Surprisingly, the predominant sequences of mother and child were very similar (Fig. lB, C): the proviral nef sequences displayed a genetic distance of 0.8% and the vif sequences, a distance of only 0.1 %. In contrast to the analysis of the V1N2 region, these small genetic distances suggested that mother and child were epidemiologically linked. This result was also confirmed by the analysis of the p17 gene in the proviral gag region. The genetic distances between the proviral sequences of mother and child reached 2.2 % in this gene. Repeated analysis of the V1 IV2 region with primers SK 122/env6 and with the limiting dilution procedure. To clear the contradictory results described above, we repeated the analysis of the V1N2 region with primers from conserved regions of the env gene. The primers env 5 and env6 were shown to amplify the V1N2 region of HIV strains from diverse geographical locations (20). The primer combination SK122/env6 resulted in a successful amplification of samples from mother and child. Sequence analysis of the PCR products revealed that the child harboured a homogeneous virus population resulting in a clear sequence with no background. This sequence was identical to the sequence we detected with primers SK122/SK123. In contrast, the mother displayed again a heterogeneous virus population in this area: sequences from different viral variants were mixed and no predominant sequence could be found.

310

P. Kasper et al.

A Child -RCVTLICH--------------NLTSSNSSTYNNSNISNDMQ-GE1KNCSFNMT Mother PL .... N.TD1WNTTASNSTNTT.1.GNGTTPAPT.RSWEK.ER ...... T .. A.

55

Child TELRDKEQKVYPLFYKLDVVQ1QNND------TQYRL1NCNTST1TQACPKVSFE 110 Mother .DME.RR.E.HAT .. N ... EPMDKANTTY ... RR .... S .. S.V ....... 1 .. .

B Child KWSKGSIVGWPKVRERIRQTPTAATGVGAASQDLDRHGA1TSSNTSTTNADCAWL Mother ....... g. . .......................................... .

55

Child EAQEDEEVGFPVRPQVPLRPMTfKAALDLSHFLKEKGGLEGL1YSKKRQEILDLW 110 Mother ...................... Y ... V ...................... d .... . Child VYHTQGFFPDWQNYTPGPGTRYPLTFGWCFKLVPMDPAETEEATEGENNSLLHPI 165 Mother ............. i . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Child CQHGMEDEDREVLVWRFDSXLAFRHTAREMHPEFYKDC*HRSC*QG Mother .... d ........ R ......... L ............... .

211

c Child KGQ**YRTIVI*R*YQEEKQRSLGTMENRWQVM1VWQVDRMRINTWKSLVKYHMH Mother

55

Child ISKKAKRWSYRHHYDSNHPK1SSEVHIPLGAAELVVTTYWGLHTGERKWHLGQGV 110 Mother Child S1EWRLKRYRTQVDPGLADQLIH1HYFDCFSESA1RKA1LGHTVSPRCDYQAGHN 165 Mother Child KVGSLQYLALTALINPKPTKPPLPSVQKLVEDRWNKPQRTRGHRENQTMNGH*SF 220 Mother Child WRSLKMKLLDIFLGPGSMA*DNISMKL Mother ...................... T ... .

247

D Child RLRPGGNKKYKLKHLVWASRELERFALNPGLLETAEGCQQIIAQIQPALKTGSEE Mother S ..... K ... R .............................. M.. 1 ......... . Child LKSLYNTVVTLYCVHQKIeVRDTKEALDKIEEiQDKSKQKTQQTVANTGSN Mother ........ A ....... R.d ............. V .... Q ..... A ...... .

55

106

Fig. 1. Sequence alignments from different proviral genes of mother and child. Only sequence differences are displayed, points denoting identical positions. Lower case letters indicate that the sequence of the mother or the child is also present. - :deletion, *: stop codon. A: env VIN2 sequences obtained by PCR with primers SK 122/123. B: proviral nef sequences. C: vif sequence alignment. D: gag p17 sequences of mother and child.

HIV Transmission from Mother to Child

311

To investigate whether the heterogeneous virus population from the mother harboured a viral variant related to the virus population in the child, we performed the limiting dilution technique. The proviral DNA of the mother was diluted before the peR to amplify single molecules. This method enabled us to obtain sequence information from single proviruses. To get a detailed analysis of the HIV quasi-species within the mother, we included available samples taken at different times (18, 21 and 28 months after infection of the mother). The sequence alignment (Fig. 2) revealed that the mother displayed a heterogeneous virus population at all times studied as indicated by the large insertions and deletions in the VIN2 sequences. We also included the sequence from the mother obtained by the first VIN2 analysis with primers SK122/SK123 in the alignment (sequence R). It was possible to find a very similar sequence (R6) by the limiting dilution technique and with primers SK122/env6, but this sequence represented a minor population within the viral variants of the investigated sample from the mother. This result suggests that selective amplification may have occurred with primers SK122/SK123 leading to the isolation of a minor variant

Child (12) 21

28

18

Child (12) 21

28

p74S

{::1 prS pr6 prB pm2 pm3 { pm4 pm6 pm?

{mEL

o

p?72-11 p?72-3

p74S

{::1 prS pr6 prB pm2 pm3 { pm4 pm6

pm? p?72-S

18

o

{:~~~~~o

p?72-11 p?72 -3

o

o

-R-CVTLICHN--------------LTSSNSSTYNNSNISNDMQ-GEIKNCSFNMTTELR TPLCVTLNCTDIWNTTASNSTNTTNITGGTTPAPNTSRSWEKMERGEIKNCSFNATTDME -PLCVTLNCSN--------------FTSSNNNTTSNSNISRDMQ-GEIKNCSFNMTTELR -PLCVTLNCSN- - - - - - - - - - - -SNFTSSNNNTSSNSNISSEIIQ-GEIKN-SFNITTELR -PLCVTLNCTNIWNTTASNSTNTTNITGNGTTPAPTSRSWEKMERKEIKNCSFNATTDIE -PLCVTLNCSN------------SNFTSSDNNTSSNSNISSEMQ-GEIKNCSFNMTTELR ----VTLNCSN------------SNFTSSNSSTSNNSNISIDMQ-GEIKNCSFNMTTELR - - - - - -LNCHN- - - - - - - - - - - - - -LTSSNNNTSSNSSISSEMP-GEIKNCSLQI- - - --PLCVTLNCSN------------SNFTSSNSSTSNNSNISIDMQ-GEIKNCSFNMTTELR -PLCVTLNCSN- - - - - - - - - - - -SNFTSNNSSTSNNSNISIDMQ-GEIKNCSFNMTTELR ------LNCH--------------NFTSSNNNTSSNRNISSEMR-GEIKNCSFNMTTELR ____ -TLNCHD- - - - - - - - - - - - - -FNSSNSNTSSNSNISSDMQ-GEIKNCSFNMTTBLR -PLCVTLNCH--------------NLTSSNNNTFSNSSISSEMR-----NCSFNMTTELR -PLCVTLNCG- - - - - - - - - - - - - -NFTSSKSNTSSIRNISSEMQ-GEIKNCSFNMTTELR -PLCVTLNCS- - - - - - - - - - - - - -NF- - - -SNTSSNSNISSEMQ-GEIKNCSFNMTTELR - - - -VTLDCHN- - - - - - - - - - - - - -FTS- - -NTSSNSNISSDMQ-GEIUCSFNMTTELR DKEQKVYPLFYKLDVVQIQNND------TQYRLINCNTSTITQACPKVSFE---DRRQEVRATFYNLDVEPMDKANTTY---RRYRLISCNSSVITQACPKISFEPIPDKERKMYALFYRLDVVQIQNND------TQYRLINCNTSTITQACPKVSFEPIPDKEKKMYALFYKLDV- - - - - - - -QINNDTQYRINN- -TSTIT- - - - - - - - - - - -NRRQEVHATFYNLNVKPMDKANTTY---RRYRLISCNSSVITQACPKISFEPIPDKEKKMYALFYKLDVVQIQNNDS-----TQYRLINCNTSTITQACPKVSFEPIPDKEKKMYALFYKLDVVQLQNNDRQSQNNDSYRLINCNTSTITQACPKVSFEPI-- - - - - - - - - - - - - - - - - - - - - - - - -NNDTQYRLLNCNTSTITQACPKVSFEPIPI DKEKKMYALFYKLDVVQLQNNDSQNNDSRQYRLINCNTSTITQACPKVSFEPIPDKEKKMYALFYKLDVVQIQN------DSTQYRLINCNTSTITQACPKVSFEPIPDKEQKMYALFYKLDVVQIQNNDS-----TQYRLINCNTSTITQACPKVSFEPI-DKEQKMYALFYKLDVVQIQNNDSQNNDSRQYRLINCNTSTITQ-CPK-SFEPIPDKEQKMYALFYKLDVVQIQNNDS-----TQYRLINCNTSTITQACPKVSFEPIPDKEQKMYALFYKLDVVQIQNNDS-----TQYRLINCNTSTITQACPKVSFEPIPDKEQKMYALFYKLDVVQIQNNDS-----TQYRLINCNTSTITQACPKVSFEPIPDUQKVYALFYRLDVVQIQNND------TQYRLINCNTSTITQACPKVSFEPI--

Fig. 2. Sequence alignment of V1N2 sequences from mother and child obtained by PCR with primers SK122/env6 and limiting dilution. Sequences of the mother were from samples taken at different times after infection as indicated: 21 (pRI to pR8) single proviral sequences), 28 (pM sequences), and 18 months (p772 sequences) after infection. pR repesents a sequence obtained by PCR with primers SK1221123. The squares indicate maternal Vl/ V2 sequences that are closely related to those of the child (p745, obtained 12 months after infection of the mother).

312

P. Kasper et al.

Table 1. Genetic distances between the different proviral sequences of mother and child Gene

Genetic distance (%)

V1N2

5.8-42.3

!lit

0.1 0.8 2.2

net

p17

from the mother which is not related to the proviral sequence in the child. But there are sequences from the mother (Fig. 2, sequences indicated by a square) displaying a very low genetic distance to the sequence in the child. For example, sequence M2 showed the lowest genetic distance of 5.8% to the proviral sequence of the child, suggesting that a minor virus population from the mother had been transmitted to the child. In addition, the limiting dilution showed that the genetic distance of the maternal VIN2-sequences compared to the sequence of the child ranged from 5.8% to 42.8%. This is in sharp contrast to the low genetic distances observed in the other investigated genes (Tab. 1). Serotypes of infecting HIV strains. Serotyping of the samples from mother and child showed a high cross-reactivity of both sera to subtype A, C, D and E peptides. The highest antibody binding reactivity was observed within antigens A and C, but the reactivity to the subtype B peptide was up to 10 times lower than that to the other peptides (Fig. 3). This result suggests that both mother and child were infected by non-subtype B strains.

30

....

25

0I

15

0

ns

--

:t:

20 DMother .Child

:::l

U C

0

10 5 0

A

B

C

D

E

V3 peptides Fig. 3. Pattern of antibody binding reactivity of the sera from mother and child to the panel of subtype-specific peptides A-E.

HIV Transmission from Mother to Child /

I

I

,-

D

313

,, \

UG274 \

/

/

I

/ A

,r \

\ \ \

\ ~14-_ _ _ _ _ _-..::~

,

UG268' , \

\

ci /

,-

,-

/

VI525

/

OJ259 /' /

I

I

I I

... ...

lH \ VI557 ------

10 % divergence

...

VI191

G

... ,

-- -

\

~

Fig. 4. Unrooted neighbour joining tree of p17 DNA sequences from mother and child with published sequences of different gag subtypes designated as A-H according to Myers et al. (24). The numbers next to some branches represent the number of bootstrap replications which support that branch.

Phylogenetic tree analysis. To confirm the relationship between the sequences of mother and child and to get reliable information about the infecting HIV-I subtype, we performed a phylogenetic tree analysis. We used the sequences of the pI? region because this gene has been proposed as an epidemiological marker. The background population included different HIV-l subtypes from diverse geographical areas (Fig. 4). These sequences clustered in the phylogenetic tree according to their subtypes as described in former studies (22). The phylogenetic analysis showed that the p1? sequences of mother and child were most closely related to HIV-l subtype A strains which are common in West Africa. It is also clear that the sequences of mother and child are closely related to each other because they clustered together in the phylogenetic tree.

Discussion Parenteral HIV-J infection of the mother. At all times of investigation, the mother was in the asymptomatic stage of infection. Nevertheless, she displayed a very heterogeneous virus population in V1N2 region, the genetic distances between the sequences ranging from 0 to 35.1%. The VIN2 region is known to be highly variable: HIV-linfected mothers were shown to display a mean variation from 2 % to more than 20% in this region, but these patients had progressed from the asymptomatic stage to AIDS (20). In contrast, genetic variability in the VIN2 region of hemophiliacs, parenterally

314

P. Kasper et al.

infected from a common source, i. e. a seronegative donor, seems to be restricted to 5% at the time of seroconversion up to one year later in the asymptomatic phase of infection (17, 18). The homogeneity of early virus populations in recipients after parenteral and sexual transmission is also confirmed by sequence analysis of other genes, even if the source of infection has been heterogeneous (33, 35). Selection upon transmission may be the reason for this phenomenon (15, 35). In contrast, the heterogeneity of VIN2 sequences in the mother described here speak in favour of a parenteral transmission of multiple HIV-l variants. Dual HIV-l infection of individuals was reported to occur in transfusion recipients (9) and drug users (28) and after sexual transmission (36). Moreover, a dual infection with HIV-l of distinct envelope subtypes (B and E) was detected in Thailand (4). However, the high heterogeneity of the maternal VIN2 variants was not reflected in the homogeneous sequences of p17, nef and vif. This could be due to the greater degree of variability, which more readily occurs in the hypervariable regions VIN2 of the env gene than in more conserved regions like gag, vif, and nef. Recombination of different viruses in the mother could also be a reason for this because rearrangements due to strand displacement have already been described during a single cycle of retrovirus replication (25). Moreover, there is evidence of recombination in a dually exposed transfusion recipient (9) and a surprising number of apparently recombinant viruses with hybrid gag and env genes belonging to different HIV-l subtypes (26). Selective HIV-J transmission from mother to child. In spite of the heterogeneity found in the mother, the child harboured a homogeneous virus population in the VlI V2 region which was related to a minor variant detected in the mother. This suggests that a selective transmission of a minor maternal env-gene variant occurred as described for previous studies (6, 34). Whereas recent studies on vertical HIV-l transmissions had been mainly based on envelope sequences, especially on the hypervariable region V3 (29, 34), this investigation of further proviral genes like p17, nef and vif showed that the populations of mother and child were very similar. Phylogenetic analysis of p 17 sequences. Recent investigations have shown that phylogenetic analysis with a limited part of the env gene should be interpreted with caution (3, 13, 14). In contrast, the pI7 gene has been proposed as a useful marker for epidemiological studies regardless of whether sequences were obtained from plasma virus RNA, from lymphocyte-associated proviral DNA, or from samples taken at different times during the course of the infection (12, 13, 16). Consequently, we used the pI7 sequences of mother and child for phylogenetic tree analysis. The two sequences clustered together in the phylogenetic tree as indicated by the high percentage of bootstrap replications (99,7%). This provides evidence that mother and child have been epidemiologically linked i. e. that the mother infected the child. The sequences of both belong to HIV-I subtype A which seems to be the major subtype in Cote d'Ivoire whereas genotypes Band D are not so common in this West African region (14). Two of the subtype A strains (CI51 and CI4) used as background population were even isolated in Cote d'Ivoire (22) which is a neighbouring country of Ghana where mother and child were infected. Serotyping and genotyping. Recent publications showed that serotyping can predict the genotypes and vice versa (7,23,37). Since mother and child had been infected in Ghana, we were interested in the subtype of the infecting HIV-l strains. Although our serotyping results suggested that both mother and child were infected by a non-subtype-B strain, it remained difficult to decide to which serotype the samples belonged because of the observed cross-reactivity between subtypes A, C, D and E. The sera

HIV Transmission from Mother to Child

315

even reacted slightly more with peptide C than with A. However, a high degree of cross-reactivity was observed in particular between A and C genotypes (7) due to the high sequence similarity in V3. In such cases, only genotyping can lead to a clear result. Implications for epidemiological analyses on HIV transmissions. The direct-sequencing method can be used as a rapid method to analyze heterogeneous virus populations (21). However, if insertions or deletions occur as in the mother-to-child transmission described here, limiting dilution or cloning of peR products is necessary. Thus, direct sequencing is only useful for studying epidemiological relationships if the virus populations within an individual are rather homogeneous (1, 15). Our sequence analysis of the VlN2 region with primers SKI22/SK123 has shown that selective amplification of minor maternal variants occurred leading to false conclusions. Previously, these primers had only been used to amplify subtype B strains (10). In contrast, primers from more conserved regions of the env gene used in our study resulted in the amplification of further maternal variants which belonged to HIV1 subtype A. Recent publications have shown that amplification with subtype-B-specific gag primers can even lead to false-negative results if samples from other subtypes are investigated (2, 5). To avoid false-negative results and selective amplification, primers from conserved regions of the HIV-l genome should be used. If predominant sequences from patients suspected of being epidemiologically linked are not related, this should be interpreted with caution. This study provides evidence that analysis of predominant sequences from parts of the env gene like VIN2 can lead to discordant results and that studies on HIV-l transmission should be based on the analysis of several proviral genes. Acknowledgements. We thank T. Sauer and E. Faatz for helpful discussion.

References 1. Albert,]., J. Wahlberg, and M. Uhlen: Forensic evidence by DNA sequencing. Nature 361 (1993) 595-596 2. Arnold, C, K. L. Barlow, S. Kaye, C Loveday, P. Balfe, and J. P. Clewley: HIV-l sequence subtype G transmission from mother to infant: Failure of variant sequence species to amplify in the Roche Amplicor test. AIDS Res. Human Retroviruses 11 (1995) 999-1001 3. Arnold, C, P. Balfe, and J. P. Clewley: Sequence distances between env genes of HIV-l from individuals infected from the same source: Implications for the investigation of possible transmission events. Virology 211 (1995) 198-203 4. Artenstein, A. w., T. C VanCott, J. R. Mascola, J. K. Carr, P. A. Hegerich, J. Gaywee, E. Sanders-Buell, M. L. Robb, D. E. Dayhoff, S. Thitivichianlert, S. Nitayaphan, J. G. McNeil, D. L. Birx, R. A. Michael, D. S. Burke, and F. E. McCutchan: Dual infection with HIV-l of distinct envelope subtypes in humans. ]. Infect. Dis. 171 (1995) 805-810 5. Barlow, K. L., J. H. C Tosswill, and J. P. Clewley: Analysis and genotyping of PCR products of the Amplicor HIV-l kit.]. Virol. Methol. 52 (1995) 65-74 6. Briant, L., C M. Wade,.f. Puel, A.}. Leigh-Brown, and M. Guyader: Analysis of envelope sequence variants suggests multiple mechanisms of mother-to-child transmission of HIV-1.]. Virol. 69 (1995) 3778-3788 7. Chengsong-Popov, R., S. Lister, D. Callow, P. Kaleebu, S. Beddows, J. Weber, and The WHO Network for HIV Isolation and Characterization: Serotyping of HIV-1 byantibody binding to the V3 loop: relationship to viral genotype. AIDS Res. Human Retroviruses, 10 (1994) 1379-1386

316

P. Kasper et al.

8. DeCock, K. M., F. Brun- Vezinet, and B. Soro: HIV-l and HIV-2 infections and AIDS in West Africa. AIDS 5 (1991) S21-S28 9. Diaz, R. S., E. C Sabino, A. Mayer,j. W. Mosley, M. P. Busch, and The Transfusion Safety Study Group: Dual HIV-l infection and recombination in a dually exposed transfusion recepient . .J. Virol. 69 (1995) 3273-3281 10. Goodenow, M., T. Huet. W. Saurin, S. Kwok, ]. Sninsky, and S. Wain-Hobson: HIV-1 isolates are rapidly evolving quasispecies: evidence for viral mixtures and preferred nucleotide substitutions.]. Acquired immuno. Defic. Syndr. 2 (1989) 344-352 11. Higgins, D. G., A.]. Bleasbyand, and R. Fuchs: CLUSTALV: Improved software for multiple sequence alignment. CABIOS 8 (1992) 189-191 12. Holmes, E. C, L. Q. Zhang, P. Robertson, A. Cleland, E. Harvey, P. Simmonds, and A. ]. Leigh Brown: The molecular epidemiology of HIV-l in Edinburgh . .J. Infect. Dis. 171 (1995) 45-53 13. Holmes, E. C, L Q. Zhang, P. Simmonds, A. S. Rogers, and A.]. Leigh Brown: Molecular investigation of a HIV infection in a patient of an HIV-infected surgeon. ]. Infect. Dis. 167 (1993) 1411-1414 14.janssens, w., L. Heyndrickx, Y. van de Peer, A. Bouckaert, K. Fransen,]. Motte, G.-M. Gershy-Damet, M. Peeters, P. Piot, and G. van der Groen: Molecular phylogeny of part of the env gene of HIV-1 strains isolated in Cote d'Ivoire. AIDS 8 (1994) 21-26 15. Kasper, P., R. Kaiser, ]. Oldenburg, H. H. Brackmann, B. Matz, and K. E. Schneweis: Parallel evolution in the V3 region of HIV-l after infection of hemophiliacs from a common source. AIDS Res. Human Retroviruses 10 (1994) 1669-1678 16. Kasper, P., P. Simmonds, K. E. Schneweis, R. Kaiser, B. Matz, j. Oldenburg, H. H. Brackmann, and E. C. Holmes: The genetic diversification of the HIV-l gag p17 gene in patients after infection from a common source. AIDS Res. Human Retroviruses 10 (1995) 1197-1201 17. Kasper, P., R. Kaiser, j.-P. Kleim, ]. Oldenburg, H. H. Brackmann, ]. Rockstroh, and K. E. Schneweis: Diversification of HIV-1 strains after infection from a unique source. AIDS Res. Human Retroviruses 9 (1993) 153-157 18. Kleim, ].-P., A. Ackermann, H. H. Brackmann, M. Gahr, and K. E. Schneweis: Epidemiologically closely related viruses from hemophilia B patients display high homology in two hypervaria ble regions of the HIV-l env gene. AIDS Res. Human Retroviruses 7 (1991) 417-421 19. Kostrikis, L G., E. Bagades, Y. Cao, L. Zhang, D. Dimitriou, and D. D. Ho: Genetic analysis of HTV-l strains from patients in Cyprus: Identification of a new subtype designated subtype 1..J. Virol. 69 (1995) 6122-6130 20. Lamers, S. L., J. W. Slesman, ].x. She, K. A. Barrie, S. M. Pomeroy, D.]. Barrett, and M. Goodenow: Independent variation and positive selection in env VI and V2 domains within maternal-infant strains of HIV-l in vivo . .J. Virol. 67 (1993) 3951-3960 21. Leitner, T., E. Halapi, G. Scarletti, P. Rossi,]. Albert, E. M. Fenyo, and M. Uhlen: Analysis of heterogenous viral populations by direct sequencing. BioTechniques 15 (1993) 120-127 22. Louwagie,j., J-: E. McCutchan, M. Peeters, T. P. Brennan, E. Sanders-Buell, G. A. Eddy, G. van der Groen, G. M. Gershy-Damet, R. Deleys, and D. S. Burke: Phylogenetic analysis of gag genes from 70 international HIV-1 isolates provides evidence for multiple genotypes. AIDS 7 (1993) 769-780 23. Mascola, ]. R.o ]. Louwagie, F. E. McCutchan, C L. Fischer, P. A. Hegerich, K. F. Wagner, A. K. Fowler, .f. G. McNeil, and D. S. Burke: Two antigetically distinct subtypes of HIV-l: Viral genotype predicts neutralization serotype,]. Inf. Dis. 169 (1994) 48-54 24. Myers, G., B. Korber, S. Wain-Hobson, K.- T. jeang, L. E. Henderson, and G. N. Pavlakis, eds: Human Retroviruses and AIDS 1994. Los Alamos National Laboratory, Los Alamos, New Mexico (1994) 25. Pathak, v: K. and H. M. Temin: Broad spectrum of in vitro forward mutations, hypermutations, and mutational hotspots in a retroviral shuttle vector after single replication

HIV Transmission from Mother to Child

317

cycle: Deletions and deletions with insertions. Proc. Natl. Acad. Sci. USA 87 (1990) 6024-6028 26. Robertson, D. L., P. M. Sharp, F. E. McCutchan, and B. H. Hahn: Recombination in HIV-1. Nature 374 (1995) 124-126 27. Rossi, P. and V. Moschese: Mother-to-child transmission of HIV. FASEB J. 5 (1991) 2419-2426 28. Sala, M., E. Pelletier, and S. Wain-Hobson: HIV-l gp120 sequences from a doubly infected drug user. AIDS Res. Human Retroviruses 11 (1995) 653-655 29. Scarlatti, G., T. Leitner, E. Halapi, j. Wahlberg, P. Marchisio, M. A. Clerici-Schoeller, H. Wigzell, E. M. Fenyo, j. Albert, M. Uhien, and P. Rossi: Comparison of V3 sequences of HIV-l from infected children with the RNA and DNA sequences of the virus populations of their mothers. Proc. Nat!' Acad. Sci. USA 90 (1993) 1721-1725 30. Shugars, D. c., M. S. Smith, D. H. Glueck, P. V. Nantermet, F. Seillier-Moseiwitsch, and R. Swanstrom: Analysis of HIV-1 nef gene sequences present in vivo. J. Virol. 67 (1993) 4639-4650 31. Simmonds, P., P. Balfe, j. F. Peutherer, C. A. Ludlam, j. O. Bishop, and A.j. Leigh Brown: HIV-infected individuals contain provirus in small numbers of peripheral mononuclear cells and at low copy number. J. Viro!. 64 (1990) 864-872 32. Wieland, U., j. Hartmann, H. Suhr, B. Salzberger, H.j. Eggers, and j. E. Kuhn: In vivo genetic variability ofthe HIV-l vif gene. Virology 203 (1994) 43-51 33. Wolfs, T. F., G. Zwart, M. Bakker, and!. Goudsmit: HIV-l genomic RNA diversification following sexual and parenteral virus transmission. Virology 189 (1992) 103-11 0 34. Wolinsky, S. M., C. M. Wike. B. T. M. Korber, C. Hutto, W. P. Parks, L. L. Rosenblum, K. j. Kunstman, M. R. Furtado, and.f. L. Munoz: Selective transmission of HIV-l variants from mothers to infants. Science 255 (1992) 1134-1136 35. Zhang, L. Q., P. MacKenzie, A. Cleland, E. C. Holmes, A.j. Leigh Brown, and P. Simmonds: Selection for specific sequences in the external envelope protein of HIV-l upon primary infection.]. Virol. 67 (1993) 3345-3356 36. Zhu, T., N. Wang, A. Carr. S. Wolinsky, and D. D. Ho: Evidence for coinfection by multiple strains of HIV-l subtype B in an acute seroconverter. J. Virol. 69 (1995) 1324-1327 37. Zwart, G., T. F. W. Wolfs, R. Bookelman, S. Hartman, M. Bakker, C. A. B. Boucher, C. Kuiken, and j. Goudsmith: Greater diversity of the V3 neutralization domain in Tanzania compared with The Netherlands: serological and genetic analysis. AIDS 7 (1993) 467-474 Dr. rer. nat. Pia Kasper, Boehringer Mannheim GmbH, D-82377 Penzberg, Germany, Tel.: 08856/60-3307, Fax: 08856/60-3131