Download Becker-Pergola_2000_AIDS Res Hum Retroviruses

AIDS RESEARCH AND HUMAN RETROVIRUSES
Volume 16, Number 8, 2000, pp. 807–813
Mary Ann Liebert, Inc.
Sequence Note
Analysis of HIV Type 1 Protease and Reverse Transcriptase
in Antiretroviral Drug-Naive Ugandan Adults
G. BECKER-PERGOLA,1 P. KATAAHA,2 L. JOHNSTON-DOW,3 S. FUNG,3 J.B. JACKSON,1
and S.H. ESHLEMAN 1
ABSTRACT
We analyzed plasma HIV-1 from 27 antiretroviral drug-naive Ugandan adults. Previous subtype analysis of
env and gag sequences from these samples identified subtypes A, C, D, and recombinant HIV-1. Sequences of
HIV-1 protease and reverse transcriptase (RT) were obtained with a commercial HIV-1 genotyping system.
Subtypes based on protease sequences differed from gag subtypes for 5 of 27 samples, demonstrating a high
rate of recombination between the gag and pol regions. Protease and RT sequences were analyzed for the
presence of amino acid polym orphisms at positions that are sites of previously characterized drug resistance
mutations. At those sites, frequent polym orphisms were detected at positions 36 and 69 in protease and positions 179, 211, and 214 in RT. Subtype-specific amino acid motifs were identified in protease. Most of the subtype A sequences had the amino acids DKKM at positions 35, 57, 69, and 89, whereas most subtype D sequences had the amino acids ERHL at those positions. Detection of those polym orphisms may provide a useful
approach for rapid identification of subtype A and D isolates in Uganda. This analysis significantly increases
the number of Ugandan protease and RT sequences characterized to date and demonstrates successful use of
a commercial HIV-1 genotyping system for analysis of diverse non-B HIV-1 subtypes.
C
antiretroviral drugs target HIV-1 protease and reverse transcriptase (RT). Resistance to antiretroviral drugs is frequently associated with selection of HIV1 variants with mutations in these enzymes. To date, most
studies of HIV-1 protease and RT have been performed for subtype (clade) B HIV-1, which is the most comm on subtype in
the United States and Europe. In contrast, there are few studies characterizing these enzymes in non-B HIV-1, which accounts for the majority of HIV-1 infections worldwide. Characterization of these enzym es in non-B HIV-1 is becoming
increasingly important, as the prevalence of non-B HIV-1 increases in countries where antiretroviral drugs are widely used,
and the availability of antiretroviral drugs increases in developing countries.
Diverse HIV-1 subtypes are found in Uganda, where the
U R R EN T LY L IC EN S ED
prevalence of HIV-1 infection is high. While general use of antiretroviral drugs in Uganda is limited, Uganda has been the site
of several clinical trials using antiretroviral drugs to treat and
prevent HIV-1 infection. Most HIV-1 infections in Uganda are
caused by subtype A and D HIV-1, 1,2 although infection with
subtypes C 1,2 and G 3,4 has also been reported. These subtypes
have been found in diverse regions throughout the world. In the
United States, individuals have been identified with subtype A,
C, and D HIV-1. 5–9 One study found subtype A HIV-1 in 2 of
22 individuals, 8 and another found 3 of 91 individuals to have
non-B HIV-1, including 1 individual with subtype A who had
no history of foreign travel or contact. 9 An important step in
understanding the genetic correlates of drug resistance in nonB HIV-1 is to characterize the baseline sequences of HIV-1 protease and RT from individuals who have never received anti-
1 Department
of Pathology, Johns Hopkins Medical Institutions, Baltimore, Maryland 21205.
Blood Bank, Kampala, Uganda.
Biosystems, Foster City, California 94404.
2 Nakasero
3 PE
807
808
retroviral therapy. In this sequence note we analyze HIV-1 protease and RT sequences from 27 antiretroviral drug-naive Ugandan adults.
Plasma samples used for analysis were obtained from asymptomatic volunteer blood donors at the Nakasero Blood Bank in
Kampala, Uganda in 1993 and 1994. 1 Units from donors that
were repeated reactive for HIV in donor-screening tests were
not used for transfusion. Plasma samples from 27 of those units
have been analyzed in this article. HIV-1 from the same sam ples was subtyped in a previous report by phylogenetic analysis of env gp41 and gag p24 sequences. 1 Most of the sam ples
(24 of 27) had the sam e subtype in both the env and gag regions. Those included 15 subtype A, 2 subtype C, and 7 subtype D samples. Three samples contained recombinant HIV-1,
one with subtype A in env and D in gag (A/D) and two with
subtype D in env and subtype A in gag (D/A) (Table 1).
Analysis of HIV-1 protease and RT sequences was performed with the Perkin-Elmer/ Applied Biosystems (PE Biosys-
BECKER-PERGOLA ET AL.
tems) HIV genotyping system (HIV genotyping system
Prt/5 9 RT; PE Biosystems, Foster City, CA). In this system ,
HIV-1 RNA is isolated from plasma and reverse transcribed
with random hexamer prim ers and Moloney murine leukemia
virus RT. An HIV-1 DNA fragment including the regions encoding protease (amino acids 1–99) and RT (amino acids
1–310) is amplified by polymerase chain reaction (PCR), using
TaqGold in a single 40-cycle reaction. The amplified DNA is
then purified and sequenced with six or seven primers and
BigDye terminator reagents. Sequences are edited, aligned,
translated into amino acids, and analyzed for the presence of
amino acid polymorphism s. Controls to monitor for and prevent DNA contam ination in sample preparation and DNA amplification have been described previously. 10
Protease sequences from the 27 Ugandan isolates are shown
in Fig. 1. Phylogenetic methods were used to determine the subtype of each sample in the protease-coding region (Fig. 2). Of
the 27 samples, 17 had subtype A protease regions, and 10 had
FIG. 1. Ugandan HIV-1 protease sequences. Ugandan plasma samples were analyzed with the PE Biosystem s HIV genotyping system as described. Protease sequences were aligned by the CLUSTAL method (MegA lign, DNASTAR, Lasergene). HIV1 subtypes were defined for the protease region by using the corresponding nucleotide sequences (see Fig. 2); the protease subtype of each isolate is indicated in parentheses. Dashes indicate identity with the consensus sequence (Majority, top). Because
the majority of samples are subtype A in protease, the consensus sequence resembles the Ugandan subtype A protease sequences.
Dashes indicate identity with the consensus sequence. X indicates the presence of an amino acid mixture. Amino acid mixtures
in the protease sequences are as follows: 135.544: 63L/P; 161.287: 57K/R; 194.793: 14K/R; 225.710: 63P/S; 326.675:
14K/R,15I/V; 108.448: 36I/M; 134.463: 13I/ V; 326.570: 33L/V; 326.662: 37D/N; 501.045: 15I/V, 69H/Y, 72I/M: 503.083: 14K/R,
60D/E, 63P/S, 64I/V. The GenBank accession numbers of the corresponding protease nucleotide sequences are as follows: 129.733,
135.544, 140.223, 161.287, 171.005, 184.574, 190.574, 194.793, 204.987, 225.706, 230.580, 326.636, 326.642, 326.662, 326.675:
GenBank numbers AF177347–AF177361, respectively. 108.448, 225.745: GenBank numbers AF176037–AF225.745, respectively.
134.463, 151.940, 42.877, 99.237, 326.570, 501.045, 503.083: GenBank numbers AF216993, AF216994, AF216996, AF216999,
AF216995, AF216997, AF216998, respectively. 225.710, 602.174, 230.298: GenBank num bers AF216992, AF216991,
AF216990, respectively.
UGANDAN HIV-1 PROTEASE AND REVERSE TRANSCRIPTASE
subtype D protease regions. In 22 of 27 samples, the subtype
of the gag and protease regions was the same (Table 1). However, 5 of 27 samples had different gag and protease subtypes,
suggesting recom bination between these regions. Interestingly,
809
those included both samples that subtyped as C in the env and
gag regions (225.745 and 108.448). One of those samples was
a C/A recombinant and the other was a C/D recombinant. Comparison of the subtypes in all three regions (env, gag, and pro-
FIG. 2. Phylogenetic analysis of HIV-1 protease sequences. Protease sequences (297 nucleotides) from the 27 Ugandan isolates were aligned with a set of 27 reference sequences from subtypes A–J recomm ended for HIV-1 subtyping. 11 Alignments
were performed by the CLUSTAL method in the program Seqpup (Don Gilbert/ftp.bio.indiana.ed u). A neighbor-joining tree was
constructed with 500 bootstrap replications of a Kimura two-param eter matrix to evaluate the robustness of the phylogenetic relationship between the sequences. The programs DNAdist and NEIGHBOR were used for this analysis. Bootstrap values were
obtained from a consensus tree, using the program CONSENSE of the software package PHYLIP v3.572.0. The tree was rooted
with the chimpanzee strain CPZGAB. Branch lengths are proportional to the amount of sequence divergence between taxa in the
tree. The Ugandan isolates are indicated in boldface.
810
BECKER-PERGOLA ET AL.
T A B LE 1. Env, Gag, A N D P R O TEA SE
S U B TY PES O F U G A N D A N I SO LA TE S a
Sample
Env
Gag
Pro
99.237
129.733
135.544
140.223
161.287
171.005
184.574
190.574
194.793
204.987
225.706
225.710
225.745
230.580
326.636
326.642
326.675
42.877
108.448
134.463
151.940
230.298
326.570
326.662
501.045
503.083
602.174
D
A
A
A
A
A
A
A
A
A
A
D
C
A
A
A
A
D
C
D
D
A
D
A
D
D
D
D
A
A
A
A
A
A
A
A
A
A
A
C
A
A
A
A
D
C
D
D
D
D
A
D
D
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
D
D
D
D
D
D
D
D
D
D
a HIV-1 from 27 Ugandan plasma samples was amplified,
sequenced, and subtyped by phylogenetic analysis. Subtyping
of the env gp41 region and the gag p24 region was performed
in a previous study. 1 Subtyping of the protease region is
described in Fig. 2. The subtype assignments for each region
are shown.
T A B LE 2.
A M IN O A C ID P O LY M O R PHISM S
IN
tease) revealed a high rate of recombination, involving 7 of 27
(26% ) of the sam ples; one sample (602.174) subtyped as D/A/D
in env/gag/protease, suggesting that this variant was generated
by multiple recombination events.
Protease sequences shown in Fig. 1 were analyzed for the
presence of amino acid polymorphism s at positions of previously characterized drug resistance mutations. 1 1 Polym orphisms were detected at nine of those positions (Table 2). In
subtype A, the polymorphisms M36I and H69K were present
in all isolates. Comparison of the protease sequences in Fig. 1
also revealed polymorphism s at four positions that were different in subtype A versus D HIV-1. Most subtype A sequences
had the amino acids D, K, K, and M at positions 35, 57, 69,
and 89, whereas most subtype D sequences had the amino acids
E, R, H, and L at those positions. For comparison, we analyzed
the amino acid sequences at those positions in other HIV-1 subtypes (B, C, F, G, H, and J).11 The DKKM pattern seen in Ugandan subtype A isolates was not seen in sequences from other
subtypes. The ERHL pattern seen in Ugandan subtype D isolates was also prevalent among subtype B sequences, but was
not seen in other subtypes. Rayfield et al. described a rapid
method for identifying subtype A and D HIV-1 in Ugandan
samples. 2 That method is based on nucleotide differences in the
C2–V3 region of these subtypes. Nucleotide differences that
encode amino acid polymorphism s in protease may also be exploited for rapid screening of Ugandan HIV-1 isolates to identify those that are likely to have subtype A or D protease-coding regions. Direct analysis of polymorphisms in protease may
be desirable for studies of HIV-1 drug resistance, since our
analysis reveals a high rate of intersubtype recombination between pol and other regions, such as env and gag.
RT sequences from the 27 Ugandan isolates were also analyzed (Fig. 3). The corresponding nucleotide sequences were
U G A N D A N HIV-1 P R O TEA SE
AND
RT SE Q U EN C ES a
Polymorphism s detected
Enzyme
Position
Protease
20
33
36
45
60
63
69
77
82
139
141
179
211
214
RT
Subtype A
(n 5 17)
Subtype D
(n 5 10)
I (1), R (2)
F (1)
I (17)
R (1)
R (1)
V/L (1)
I/M (1), I (5)
V (1), I (1), L/P (1), P/S (1) T (2)
K (17)
E/D (1), E (2)
P/S (1)
H/Y (1), Y (2)
I (2)
I (1)
M (1)
I (13)
N (1), K (3), S (12)
F/L (1), F (14)
E (1)
I/V (1), I (1)
S (1), K/R (1), K (5)
F (10)
Previously
characterized
resistance
mutation
K20RM
L33F
M36I
K45I
D60E
L63P
H69Y
V77I
V82AFIST
T139I
G141E
V179D E
R211K
L214F
a
Subtype A and D protease sequences were analyzed for the presence of amino acid polym orphism s at positions of previously
characterized drug resistance mutations (see text). The polymorphism s detected at each position are shown for subtype A and D
isolates. Polymorphism s corresponding to known drug resistance mutations are shown in boldface. The number of isolates with
each polymorphism is shown in parentheses. Amino acid mixtures are indicated (e.g., a mixture of proline [P] and leucine [L] at
position 63 is indicated as L/P).
FIG. 3. Ugandan HIV-1 RT sequences. Ugandan plasma samples were analyzed as described in Fig. 1. RT sequences are shown. Amino acid mixtures are as follows: 99.237:
6D/N; 129.733: 43K/R, 48A/S, 251N/S; 135.544: 123N/S, 214F/L; 190.574: 173L/S; 225.706: 6D/N, 178I/L, 292I/V; 225.710: 6D/E, 300E/Q; 225.745: 291D/E; 230.580: 292I/V;
326.636: 174K/Q, 200A/T, 207E/G; 326.642: 142I/V; 326.675: 122E/K; 42.877: 211K/R; 108.448: 50I/V, 173A/S, 288A/S, 292I/V; 134.463: 6E/K, 49K/R, 165A/T; 230.298: 27P/T,
48L/S, 121D/Y; 326.570: 123E/G; 326.662: 178M/V, 179I/V, 216A/T, 292I/V; 501.045 178I/M, 224D/E; 503.083 39P/T, 121D/H, 122E/K; 602.174: 40D/E, 292I/V. GenBank accession numbers of the corresponding RT nucleotide sequences are as follows: 129.733, 135.544, 140.223, 161.287, 171.005, 184.574, 190.574, 194.793, 204.987, 225.706, 230.580,
326.636, 326.642, 326.662, 326.675: GenBank numbers AF177362–AF177376, respectively. 230.298, 602.174, 225.710, 134.463, 151.940, 326.570, 42.877, 501.045, 503.083, 99.237:
GenBank numbers AF188494–AF188503, respectively. 108.448 and 225.745: GenBank numbers AF177543–AF177544, respectively.
UGANDAN HIV-1 PROTEASE AND REVERSE TRANSCRIPTASE
811
812
used for phylogenetic subtyping, using the reference sequences
and methods described in Fig. 2 (data not shown). The subtype
assignments based on protease sequences and RT sequences
were the same for 25 of 27 samples. For two of the sam ples
(225.710 and 108.448), subtypes could not be definitively assigned on the basis of sequences from the RT region.
The Ugandan RT sequences in Fig. 3 were examined for the
presence of amino acid polymorphism s at positions in RT that
are sites of previously characterized drug resistance mutations. 11 Amino acid polymorphism s were detected at five of
those positions (Table 2). Polymorphism s at the three positions
(179, 211, and 214) were observed in multiple isolates. The
polymorphism s V179I, R211S, and L214F were present in the
majority of subtype A isolates. In contrast, most of the subtype
D isolates had valine (V) at position 179 and lysine (K) at position 211; all the subtype D isolates had the L214F polymorphism. The significance of these polymorphism s in non-B HIV1 is not known.
The low prevalence of prim ary drug resistance mutations in
this population is consistent with the lack of widespread use of
antiretroviral drugs in Uganda. It also demonstrates that primary drug resistance mutations identified in subtype B HIV-1
do not occur as common polymorphism s in HIV-1 from
Uganda. It is interesting to note that many of the isolates contained polym orphism s at positions of previously characterized
drug resistance mutations. Furthermore, many of the isolates
contained amino acid mixtures at those positions (e.g., protease
60D/E, 63L/P, 63P/S, and 69H/Y and RT 179I/V, 211K/ R, and
214F/L). This suggests that HIV-1 variants with polym orphisms
at those positions arise frequently in viral populations without
drug selection.
It is not known whether drug resistance mutations characterized in subtype B HIV-1 have the same phenotypic effects
in other HIV-1 subtypes. In theory, polymorphisms in HIV-1
protease and RT that vary among subtypes could influence the
natural resistance of isolates to antiretroviral drugs, modulate
the fitness of HIV-1 variants that acquire drug resistance mutations, or influence the probability that other mutations causing drug resistance would arise at those positions during drug
selection. Studies comparing the genetic correlates of drug resistance in different HIV-1 subtypes are clearly needed to address these issues.
Genetic polymorphism s can complicate drug resistance
analysis of HIV-1. Studies illustrate problem s that can arise by
using genotypic assays based on hybridization. 12–14 The PE
Biosystems HIV genotyping system used in this article was developed for analysis of subtype B HIV-1. In this system , sequence heterogeneity among HIV-1 subtypes can potentially interfere with reverse transcription, PCR amplification, or DNA
sequencing, if primers fail to bind to the HIV-1 template. An
abstract has described successful use of this system for analysis of cultured non-B HIV-1 reference isolates, using both kit
and nonkit primers. 15 Here, we demonstrate successful use of
this system for analysis of diverse HIV-1 subtypes in plasma,
without the use of culture, and using only primers supplied by
the manufacturer. Continued development and optimization of
genotyping assays will be important for evaluation of HIV-1
protease and RT in drug-naive and drug-experienced patients
with non-B HIV-1 infection. Such assays will play a key role
BECKER-PERGOLA ET AL.
in enhancing our understanding of drug resistance in non-subtype B HIV-1, and monitoring the emergence of drug-resistant
HIV-1 worldwide.
ACKNOWLEDGMENTS
The authors thank Catherine Brennan (Abbott Laboratories)
for helpful discussions regarding sample analysis. The authors
also thank Vernell Moutlon for technical assistance, and PE
Biosystems for providing reagents for this study.
This work was supported in part by the Pediatric AIDS Clinical Trials Group (PACTG), the Adult AIDS Clinical Trials
Group (AACTG), and NIH-CH/HD34348.
REFERENCES
1. Brennan CA, Lund JK, Golden A, Yamaguchi J, Vallari AS,
Phillips JF, Kataaha PK, Jackson JB, and Devare SG: Serologic
and phylogenetic characterization of HIV-1 subtypes in Uganda.
AIDS 1997;11:1823– 1832.
2. Rayfield MA, Downing RG, Baggs J, Hu DJ, Pieniazek D, Luo
CC, Biryahwaho B, Otten RA, Sempala SDK, and Dondero TJ: A
molecular epidemiologic survey of HIV in Uganda. HIV Variant
Working Group. AIDS 1998;12:521– 527.
3. Kaleebu P, Bobkov A, Cheingsong-Po pov R, Bieniasz P, Garaev
M, and Weber J: Identification of HIV-1 subtype G from Uganda.
AIDS Res Hum Retroviruses 1995;11:657.
4. Mellquist JL, Bowman B, Kasturi L, Guay L, Kataaha P, Mmiro
F, Ndguwa C, Jackson JB, and Eshleman SH: Characterization of
HIV type 1 gp120 V3 region sequences from Ugandan infants.
AIDS Res Hum Retroviruses 1998;14:1391– 1395.
5. Brodine SK, Mascola JR, Weiss PJ, Ito SI, Porter KR, Artenstein
AW, Garland FC, McCutchan FE, and Burke DS: Detection of
diverse HIV-1 genetic subtypes in the USA. Lancet 1995;346:
1198– 1199.
6. Gao F, Yue L, Hill SC, Robertson DL, Graves AH, Saag MS, Shaw
GM, Sharp PM, and Hahn BH: HIV-1 sequence subtype D in the
United States. AIDS Res Hum Retroviruses 1994;10:625– 627.
7. Beatrice ST, Oleszko WR, Punsalang A, Chaisson MA, Torian LV,
Schable CA, and Weisfuse IB: HIV-2 and non-B subtypes of HIV1 group M in specimens submitted to the New York City (NYC)
Department of Health Retrovirology Laboratory from 1993– 1997.
In: 12th World AIDS Conference , Geneva, Switzerland, 1998. [Abstract 42116]
8. Irwin KL, Pau CP, Lupo D, Pienazek D, Luo CC, Olivo N, Rayfield M, Hu DJ, Weber JT, Respess RA, Janssen R, Minor P, and
Ernst J: Presence of human immunodeficie ncy virus (HIV) type 1
subtype A infection in a New York community with high HIV
prevalence: A sentinel site for monitoring HIV genetic diversity in
North America. Centers for Disease Control and Prevention—
Bronx Lebanon HIV Serosurvey Team. J Infect Dis 1997;
176:1629– 1633.
9. Weidle PJ, Ganea CE, Pienniazek D, Ramos A, Schable CA, Ernst
J, and McGow an J: Prevalence of HIV-1 group M, non-B subtypes
in a Bronx New York community: A sentinel site for monitoring
HIV genetic diversity in the United States. In: 12th World AIDS
Conference , Geneva, Switzerland, 1998. [Abstract 13225]
10. Mellquist JL, Becker-Pergola G, Gu J, Guay L, Himes L, Kataaha
P, Mmiro F, Ndugwa C, Jackson JB, and Eshleman SH: Dual transmission of subtype A and D HIV-1 viruses from a Ugandan woman
to her infant. AIDS Res Hum Retroviruses 1999;15:217– 221.
UGANDAN HIV-1 PROTEASE AND REVERSE TRANSCRIPTASE
11. Korber B, Kuiken C, Foley B, Hahn B, McCutchan F, Mellors J,
and Sodroski J: Human Retroviruses and AIDS: A Compilation and
Analysis of Nucleic Acid and Amino Acid Sequences . Theoretical
Biology and Biophysics Group, Los Alamos National Laboratory,
Los Alamos, New Mexico, 1998.
12. Koch N, Yahi N, Colson P, Fantini J, and Tamalet C: Genetic polymorphism near HIV-1 reverse transcriptase resistance-associated
codons is a major obstacle for the line probe assay as an alternative method to sequence analysis. J Virol Methods 1999;80:25– 31.
13. Puchhammer-Stockl E, Schmied B, Mandl CW, Vetter N, and
Heinz FX: Comparison of line probe assay (LIPA) and sequence
analysis for detection of HIV-1 drug resistance. J Med Virol
1999;57:283– 289.
14. Vahey M, Nau ME, Barrick S, Cooley JD, Sawyer R, Sleeker AA,
Vickerman P, Bloor S, Larder B, Michael NL, and Wegner SA:
Performance of the affymetrix GeneChip HIV PRT 440 platform
for antiretroviral drug resistance genotyping of human immuno-
813
deficiency virus type 1 clades and viral isolates with length polymorphisms. J Clin Microbiol 1999;37:2533– 2537.
15. Huang D, Polstra D, Geisler T, Scianna S, Bremer J, Robb M,
Michael N, Brun-Vezinet F, and Reichelderfer P: Characterization
of an international HIV-1 subtype panel for future use as standards
and reference strains. In: 6th Conf. on Retroviruses and Opportunistic Infections , Chicago, Illinois, 1999. [Abstract 719]
Address reprint requests to:
Susan Eshlem an
Department of Pathology
Johns Hopkins University School of Medicine
Ross Building 646
720 Rutland Ave.
Baltimore, Maryland 21205