High HIV-1 genetic diversity in Cuba

Abstract

HIV-1 subtype B is largely predominant in the Caribbean, although other subtypes have been recently identified in Cuba.
To examine HIV-1 genetic diversity in Cuba.
The study enrolled 105 HIV-1-infected individuals, 93 of whom had acquired the infection in Cuba. DNA from peripheral blood mononuclear cells was used for polymerase chain reaction amplification and sequencing of pol (protease-reverse transcriptase) and env (V3 region) segments. Phylogenetic trees were constructed using the neighbour-joining method. Intersubtype recombination was analysed by bootscanning.
Of the samples, 50 (48%) were of subtype B and 55 (52%) of diverse non-B subtypes and recombinant forms. Among non-B viruses, 12 were non-recombinant, belonging to six subtypes (C, D, F1, G, H and J), the most frequent of which was subtype G (n = 5). The remaining 43 (78%) non-B viruses were recombinant, with 14 different forms, the two most common of which were Dpol/Aenv (n = 21) and U(unknown)pol/Henv (n = 7), which grouped in respective monophyletic clusters. Twelve recombinant viruses were mosaics of different genetic forms circulating in Cuba. Overall, 21 genetic forms were identified, with all known HIV-1 group M subtypes present in Cuba, either as non-recombinant viruses or as segments of recombinant forms. Non-B subtype viruses were predominant among heterosexuals (72%) and B subtype viruses among homo- or bisexuals (63%).
An extraordinarily high diversity of HIV-1 genetic forms, unparalleled in the Americas and comparable to that found in Central Africa, is present in Cuba.

Full text

High HIV-1 genetic diversity in Cuba
Marı
´a Teresa Cuevasa, Ignacio Ruibalb, Marı
´a Luisa Villahermosaa,
He
´ctor Dı
´azb, Elena Delgadoa, Elena Va
´zquez-de Pargaa,
Lucı
´aPe
´rez-A
´lvareza, Madelı
´n Blanco de Armasb, Laureano Cuevasa,
Leandro Medranoa, Enrique Noab, Saladin Osmanovc, Rafael Na
´jeraa
and Michael M. Thomsona
Background: HIV-1 subtype B is largely predominant in the Caribbean, although other
subtypes have been recently identified in Cuba.
Objectives: To examine HIV-1 genetic diversity in Cuba.
Methods: The study enrolled 105 HIV-1-infected individuals, 93 of whom had
acquired the infection in Cuba. DNA from peripheral blood mononuclear cells was
used for polymerase chain reaction amplification and sequencing of pol (protease–
reverse transcriptase) and env (V3 region) segments. Phylogenetic trees were con-
structed using the neighbour-joining method. Intersubtype recombination was ana-
lysed by bootscanning.
Results: Of the samples, 50 (48%) were of subtype B and 55 (52%) of diverse non-B
subtypes and recombinant forms. Among non-B viruses, 12 were non-recombinant,
belonging to six subtypes (C, D, F1, G, H and J), the most frequent of which was
subtype G (n ¼5). The remaining 43 (78%) non-B viruses were recombinant, with 14
different forms, the two most common of which were Dpol/Aenv (n ¼21) and U(un-
known)pol/Henv (n ¼7), which grouped in respective monophyletic clusters. Twelve
recombinant viruses were mosaics of different genetic forms circulating in Cuba.
Overall, 21 genetic forms were identified, with all known HIV-1 group M subtypes
present in Cuba, either as non-recombinant viruses or as segments of recombinant
forms. Non-B subtype viruses were predominant among heterosexuals (72%) and B
subtype viruses among homo- or bisexuals (63%).
Conclusion: An extraordinarily high diversity of HIV-1 genetic forms, unparalleled in
the Americas and comparable to that found in Central Africa, is present in Cuba.
&2002 Lippincott Williams & Wilkins
AIDS 2002, 16:1643–1653
Keywords: HIV-1, Cuba, subtypes, recombinant forms, molecular epidemiology
Introduction
In the Americas, subtype B is largely predominant
among HIV-1 genetic forms, although in Brazil,
Argentina and Uruguay, substantial proportions of
infections are caused by F subtype or BF recombinant
viruses [1–4]. Apart from these three countries, infec-
tions with non-B genetic forms are unusual; however,
recently, it was reported that env subtypes A, C and H
had been identified in 5 of 11 samples in Cuba [5].
The identification of diverse non-B subtypes in Cuba is
not unexpected, considering that large numbers of
From the aInstituto de Salud Carlos III, Madrid, Spain, bLaboratorio de Investigacio
´n del SIDA, Havana, Cuba and cUNAIDS,
Geneva, Switzerland.
Requests for reprints to: Dr R. Na
´jera, Department of Viral Pathogenesis, Instituto de Salud Carlos III, Ctra. Majadahonda-
Pozuelo, Km. 2, 28220 Madrid, Spain.
Received: 10 August 2002; revised: 1 October 2001; accepted: 21 March 2002.
ISSN 0269-9370 &2002 Lippincott Williams & Wilkins 1643

Cuban military and civilian personnel had been sta-
tioned in the 1970s and 1980s in Angola [6], a country
neighbouring the Democratic Republic of Congo
(DRC), where the highest group M diversity is found
[7], and that many of the early cases of HIV infection
in Cuba were detected among these individuals [8].
The presence of numerous Cuban aid workers in
several sub-Saharan African countries might also have
contributed to the introduction of diverse HIV genetic
forms of African origin [8,9]. In a large-scale survey
carried out between 1986 and 1989, in which more
than 5 million individuals were tested, 122 (28%) of
434 HIV-seropositive infections detected were directly
attributable to the presence of Cuban military and
civilian personnel in Africa [10].
Cuba has the lowest HIV prevalence in the Americas,
with an estimated number of infections of 1950 at the
end of 1999 for a population of 11.2 million inhabitants
[11]. This corresponds to a 0.03% prevalence in adults,
which contrasts with a prevalence of 2% in the
Caribbean area considered globally (ranging from 0.7%
in Jamaica to 5.2% in Haiti), which is only second to
that of sub-Saharan Africa [12]. However, a sharp
increase in HIV infections in Cuba has recently been
reported [13]. Transmission in most cases is by either
hetero- or homosexual exposure [11]. Earlier in the
epidemic, the controversial policy of mandatory quar-
antine of all HIV-infected individuals in sanatoria was a
matter of lively debate in medical journals [10,14,15].
This policy was modified in 1993, when staying in a
sanatorium became voluntary.
To examine the distribution of HIV genetic forms, as
well as the prevalence of drug resistance-associated
mutations in Cuba [16], a study was conducted under
the provisions of UNAIDS, in which segments of pol
(protease–reverse transcriptase) and env of 105 HIV-
infected individuals were analysed. The results revealed
the presence of a high proportion of non-B subtype
viruses, and a high diversity of genetic forms, unprece-
dented in the Americas and comparable to that of
Central African countries.
Methods
Study subjects
The number of HIV-1-infected individuals enrolled
was 105, of whom 74 were men, 30 were women and
one was a perinatally infected child. Risk categories
were: 60 homo- or bisexual, 43 heterosexual, one
accidental exposure and one perinatal transmission.
Places of residence were 54 Havana City, 21 Villa
Clara province, 19 Matanzas province and 10 were
distributed among six other provinces (Granma, Cama-
gu
¨ey, Pinar del Rı
´o, Ciego de A
´vila, Cienfuegos and
Sancti Spiritus). 93 individuals were reported to having
contracted HIV in Cuba, two in North America (one
in the United States and one in Canada) and 10 in
Africa (four in DRC, three in Angola, two in Ethiopia
and one in Zambia). All samples were collected in
1999.
Sample preparation, amplification and
sequencing
Peripheral blood mononuclear cells were separated by
centrifugation on Ficoll–Hypaque gradient. Samples
were prepared for polymerase chain reaction (PCR) by
cell lysis and digestion with proteinase K, as described
[17]. A lysate of 2 3105cells was used for each PCR.
Amplification of pol and env (C2–V3–C3) segments
was done by nested PCR, using primers and thermo-
cycling profiles as described [18,19]. Amplification was
checked by electrophoresis in an agarose gel with
ethidium bromide staining. After enzymatic removal of
dNTP and primers remaining in solution [20], purified
PCR products were directly sequenced using ABI
Prism BigDye Terminator Cycle Sequencing kit and
ABI 377 sequencer (Applied Biosystems, Foster City,
California, USA). Electrophoretogram sequences were
corrected with BioEdit (Tom Hall, http://www.
mbio.ncsu.edu/BioEdit/bioedit.html).
Analysis of sequences
Sequences were aligned with reference sequences using
Clustal X [21], with manual adjustments, considering
protein sequences. Neighbour-joining phylogenetic
trees, based on Kimura’s two-parameter distances, with
consistency of tree topologies assessed by bootstrapping,
were constructed using Clustal X and viewed with
Treeview (Rod Page, http://taxonomy.zoology.gla.ac.
uk/rod/treeview.html). Intersubtype recombination in
pol was analysed by bootscanning using Simplot 2.5
(Stuart Ray, http://www.med.jhu.edu/deptmed/sray/
download/). Bootstrap support for sequence clusters of
70% or higher was considered significant [22]. Homo-
logies with GenBank sequences were searched using
BLAST Search (NCBI, http://www.ncbi.nlm.nih.gov/
BLAST). To exclude the possibility of PCR-mediated
artefacts, intersubtype breakpoints were confirmed in
duplicate PCR carried out separately.
Statistical analysis
Significance of differences in prevalences of B and
non-B subtype infections among groups with different
epidemiologic characteristics was analysed with the ÷2
test with Yate’s correction using Sigma software.
Results
Phylogenetic analysis of
pol
sequences
In the phylogenetic neighbour-joining tree of pol
AIDS 2002, Vol 16 No 121644

sequences (Fig. 1a,c), 50 sequences grouped with
subtype B reference viruses and 55 were of non-B
subtypes. Of these, 28 grouped with subtype D, seven
with subtype G, three with subtype C, one each with
subtypes F1, H and J and 14 branched apart from
subtype reference sequences. Of the last 14 samples,
nine formed a monophyletic cluster supported by 100%
bootstrap value. Bootscan analysis using Simplot soft-
ware (Fig. 2a) indicated that these nine sequences did
not appear to be recombinants of known subtypes,
although phylogenetic trees of partial pol segments
suggested that different segments might be distantly
related to G and F subtype viruses. Consequently, pol
sequences of this cluster are referred to as U, meaning
unclassified or unknown subtype. The remaining five
sequences not grouping with subtype references were
intersubtype recombinant (see below). pol sequences of
Cuba branching with subtype D, except CU20, clus-
CU53
CU47
(a) Pol, B subtype (b) V3, B subtype
SIVcpzUS
100
100
100
100
100
100
100
100
100
99
91
98
80
0.1
B
A.92UG037
A.U455
C.92BR025
C.ETH2220
G.92NG083
G.SE6165
H.V1997
H.90CF056
K.EQTB11C
K.MP535 J.SE7887
J.SE7022
F1. VI850
F1.93BR020.1
D.ELI
D.NDK
CU101
CU5
CU2
CU52
CU90
CU86
CU96
CU12
CU97
CU98
CU26
CU24
CU18
CU28
CU22
CU25
CU1
CU92
CU77
CU53
CU48
CU41
CU44
CU89
CU91
CU47
B.WEAU160
CU99
CU50
CU59
CU58
CU63
CU6
CU3
CU21
CU19
CU84
CU79
CU43
CU42
CU93
B.HXB2
B.JRFL
CU49
CU82
CU32
CU83
CU102
CU35
CU27
CU11
CU36
CU37
SIVcpzUS
100
100
96
100
100
100
92
92
99
86
90
0.1
G.92NG083
G.SE6165
A.U455
A.92UG037
H.V1997
H.90CF056
J.SE7022
J.SE7887
C.92BR025
C.ETH2220
F1.V1850
F1.93BR020.1
K.EQTB11C
K.MP535
D.83.ELI
D.NDK
B.WEAU160
CU37
CU36
CU83
CU35
CU102
CU92
CU1
CU52
CU2
CU86
CU28
CU90
CU25
CU24
CU26
CU96
CU22
CU97
CU12
CU98
CU18
B.HXB2
CU11
CU32
CU82
CU91
CU41
CU5
CU101
CU50
CU3
CU59
CU58
CU63
CU44
CU19
CU21
CU84
CU43
CU42
CU79
CU49
CU93
CU99
B.JRFL
CU77
CU6
CU89
CU48
B
Fig 1. Phylogenetic neighbour-joining trees of pol and V3 region sequences of B subtype (a,b) and non-B subtype (c,d) viruses of
Cuba. Viruses of Cuba are shown in boldface. Asterisks denote pol sequences that were identified as recombinant upon
bootscan analysis. Bootstrap values 70% or higher of key nodes are shown. Relevant clusters are signalled with brackets.
Subtypes or CRF with which viruses of Cuba cluster are indicated on the right of the corresponding brackets. U denotes a cluster
of viruses of unknown subtype in pol. MAL-like denotes viruses with pol sequences clustering with the African complex (ADKU)
recombinant isolate MAL.
HIV-1 diversity in Cuba Cuevas et al. 1645

tered with each other in the phylogenetic tree,
although bootstrap support of this group did not reach
significant values. Further analysis by bootscanning
revealed that five of the sequences of this cluster
(CU17, CU45, CU54, CU57 and CU70) were inter-
subtype recombinant (see below). When these recom-
binants were excluded from the analysis, the bootstrap
value supporting the cluster formed by the remaining
sequences of the group was 92%. The bootstrap value
supporting branching of this cluster with subtype D
references increased to 76% after excluding CU20 and
the reference isolate 84ZR085.
To examine intersubtype recombination, pol sequences
were analysed by bootscanning, which revealed that 12
were recombinant (Fig. 2d–l), with subtypes (in 59–39
order) as follows (U denotes unknown subtype): three
UK (CU33, CU34 and CU55, which grouped in the
phylogenetic tree with the African MAL isolate [23–
25]), two GBGB (CU100 and CU103), one BU
(CU13), one UB (CU66), two GD (CU45 and CU57,
which exhibit different crossover points), one DB
(CU54), one BD (CU17) and one DUD (CU70). In
all cases, grouping of different segments with different
reference sequences was supported by significant boot-
(c) Pol, non-B subtype (d) V3, non-B subtype
SIVcpzUS SIVcpzUS
CU54
CU13
CU46
CU66
CU78
CU81
CU39
CU56
CU85
CU87
CU74
CU100
CU103
CU75
CU20
CU9
CU4
CU45
CU88
CU14
CU68
CU76
CU69
CU15
CU70
CU16
CU33
CU34
CU55
CU7
CU62
CU17
CU67
CU61
CU104
CU57
CU38
CU30
CU105
CU73
CU65
CU60
CU64
CU51
CU40
CU94
CU95
CU71
CU80
CU31
CU29
CU23
A
H
J
D
01_AE
G
C
F1
B
B.HXB2
B.JRFL
F1.93BR020.1
F1.VI850
C.ETH2220
C.95IN21068
G.DRCBL
G.SE6165
CRF01_AE.90FC402
CRF01_AE.CM240
D.NDK
D.8ZR085
J.SE7887
J.SE7022
F2.MP255
F2.MP257
H.VI997
H.90CF056
A.U455
A.92UG037
A.SE7253
0.1
93
99
100
100
87
92
99 100
100 97
96 100
88 100
70
96
91
100
99
97
89
75
MAL-like
C
H
J
F1
U
G
D
0.1
100
82
100
100 100
100
100 100
99
97
93
100
100
95 83
99
75
100
93
CU33
CU55
CU34
CU39
CU81
CU10
CU4
CU9
CU13*
CU46
CU75
CU69
CU76
CU88
CU14
CU16
CU15
CU68
CU66*
CU100*
CU103*
CU72
CU74
CU85
CU87
CU56
CU20
CU70*
CU54*
CU67
CU17*
CU51
CU30
CU105
CU7
CU62
CU38
CU104
CU60
CU45*
CU57*
CU8
CU61
CU64
CU73
CU65
CU95
CU71
CU40
CU94
CU80
CU31
CU23
CU29
A.U455
A.92UG037
ADKU.MAL
C.95IN21068
C.92BR025
C.ETH2220
H.VI997
H.90CF056
K.EQTB11C
K.MP535
J.SE7887
J.SE7022
F2.MP255
F2.MP257
CU78
F1.VI850
F1.93BR020
G.SE6165
G.DRCBL
G.92NG083
B.HXB2
B.JRFL
D.84ZR085
D.ELI
D.NDK
Fig 1. (continued ).
AIDS 2002, Vol 16 No 121646

strap values, and results were consistent independently
of reference sequences used. Bootscan analysis and
phylogenetic trees of partial segments of the amplified
sequences indicated that D subtype and U segments of
recombinant pol sequences of CU13, CU66, CU45,
CU57, CU54, CU17 and CU70 (but not unclassified
segments of CU33, CU34 and CU55) grouped with
viruses of the D subtype and U pol clusters of Cuba,
respectively, described above. Similarly, the G subtype
segments of CU57 GD recombinant sequence clustered
% Bootstrap value
(a) CU68
Position (bp)
0 100 200 300 400 500 600 700 800 900
110
100
90
80
70
60
50
40
30
20
10
0
⫺10
A
B
C
D
F1
G
J
K
F2
H
Fig. 2. Bootscan plots of pol sequences of Cuba. Horizontal axis represents the position of the midpoint of the window from
nucleotide (nt) 1 of protease, and the vertical axis represents bootstrap values supporting branching with reference sequences.
Windows of 300 nt were used, except for CU45 (180 nt) and CU103 (200 nt), advancing in steps of 20 nt. Trees were constructed
with the neighbour-joining algorithm using Kimura’s two-parameter distances, with transversion to transition ratio set to 2.
Subtype reference isolates used were U455 (A), JRFL (B), ETH2220 (C), NDK (D), 93BR020.1 (F1), MP255 (F2), DRCBL (G),
90CF056 (H), SE7022 (J), and MP535 (K). D and G subtype references used for analysis of recombinant sequences were D and G
subtype pol sequences of viruses from Cuban isolates CU87 and CU40, respectively, which clustered uniformly with database D
and G subtype reference isolates in the analysed segment, as shown in (b) and (c). In bootscans of CU13, CU66, and CU70, the
pol sequence of CU68 from the U cluster of Cuba was used as reference.
CU40(b)
B
D
C
H
Position (bp)
0 100 200 300 400 500 600 700 800 900
% Bootstrap value
110
100
90
80
70
60
50
40
30
20
10
0
⫺10
G
C
H
CU87(c)
110
100
90
80
70
60
50
40
30
20
10
0
⫺10
% Bootstrap value
Position (bp)
0 100 200 300 400 500 600 700 800 900
CU55(d)
A
B
C
D
F1
F2
G
H
J
K
Position (bp)
0 100 200 300 400 500 600 700 800 9001.000
% Bootstrap value
110
100
90
80
70
60
50
40
30
20
10
0
⫺10
(e) CU103
B
G
C
H
Position (bp)
% Bootstrap value
110
100
90
80
70
60
50
40
30
20
10
0
⫺10
0 100 200 300 400 500 600 700 800 900
CU13
B
U (CU68)
C
H
Position (bp)
(f)
% Bootstrap value
110
100
90
80
70
60
50
40
30
20
10
0
⫺10
0 100 200 300 400 500 600 700 800 900
HIV-1 diversity in Cuba Cuevas et al. 1647

with G subtype pol sequences of CU74, CU85 and
CU87 (which form a monophyletic group). This
indicates that these recombinant were probably gener-
ated by recombination of genetic forms circulating in
Cuba.
Phylogenetic analysis of V3 sequences
In phylogenetic trees (Fig. 1b,d), sequences of the V3
region grouped with the following subtypes or CRF:
A(n¼26), B (n ¼54), C (n ¼2), D (n ¼1), F1
(n ¼1), G (n ¼6), H (n ¼11), J (n ¼1) and CRF01-
AE (n ¼1). In two samples, CU10 and CU72, which
are of subtypes C and G, respectively, in pol, the V3
region could not be amplified.
Subtypes in V3 were coincident with those in pol for
all viruses that were of non-recombinant subtypes B,
C, F1, G, H and J in pol. Viruses of the Cuban D
subtype pol cluster, except CU8, were of subtype A in
V3, forming in this segment a monophyletic group,
which also included viruses with recombinant pol
sequences CU17 (BD in pol) and CU57 (GD in pol).
Viruses forming the U cluster in pol, except CU46 and
CU75, were of subtype H in V3, forming in this
segment a monophyletic group, which also included
CU66(g)
B
U (CU68)
C
H
Position (bp)
% Bootstrap value
110
100
90
80
70
60
50
40
30
20
10
0 100 200 500 600 700 800 900
0
⫺10
300 400
Fig. 2. (continued)
CU45(h)
D (CU40)
G (CU87)
C
H
Position (bp)
% Bootstrap value
110
100
90
80
70
60
50
40
20
10
0
⫺10
0 100 200 300 400 500 600 700 800 900
30
CU57(i)
D (CU40)
G (CU87)
C
H
Position (bp)
% Bootstrap value
110
100
90
80
70
60
50
40
30
20
10
0
⫺10
0 100 200 300 400 500 600 700 800 900
CU54(j)
B
D (CU40)
C
H
Position (bp)
% Bootstrap value
110
100
90
80
70
60
50
40
30
20
10
0
⫺10
0 100 200 300 600 700400 500 800 900
CU17(k)
B
D (CU40)
C
H
Position (bp)
% Bootstrap value
110
100
90
80
70
60
50
40
30
20
10
0
⫺10
0 100 200 300 400 500 600 700 800 900
CU70(l)
D (CU40)
U (CU68)
C
H
Position (bp)
% Bootstrap value
110
100
90
80
70
60
50
40
30
20
10
0
⫺10
0 100 200 300 400 500 600 700 800 9001.000
AIDS 2002, Vol 16 No 121648

CU8 (D in pol), CU45 (GD in pol) and CU70 (DUD
in pol). CU46 grouped with subtype B, and CU75
with CRF01-AE in V3. The three viruses that were
MAL-like in pol grouped with subtype A reference
sequences in V3, similarly to the isolate X327 of Spain
[26], but different from both MAL [23–25] and the
partly MAL-like virus NOGIL3 of Norway [27],
which are of subtypes D and H, respectively, in env.
Subtypes in pol and V3 of all viruses are shown in
Tables 1 and 2, with distribution of genetic forms in
Cuba represented graphically in Fig. 3. Overall, 43
(78%) of the 55 non-B subtype viruses were recombi-
nant. The number of genetic forms identified was 21,
including seven non-recombinant subtypes and 14
recombinant forms.
Phylogenetic relations with database sequences
Homologies of non-B subtype sequences with se-
quences deposited in GenBank were searched using the
BLAST algorithm. No sequences homologous to those
of the U pol cluster grouping with these in phyloge-
netic trees were found in the database. There were 10
database H subtype sequences that grouped with the
Cuban U/H viruses in V3, with significant bootstrap
values (80%), seven of which were from the DRC
(Fig. 4). One subtype A sequence from the Republic
of Congo grouped with the Cuban D/A recombinants
in V3. Pol sequences of these African viruses are not
available; therefore, it is not known if they are
phylogenetically related to the U/H and D/A recombi-
nants of Cuba, respectively, in pol.
Pol sequences of CU33, CU34 and CU55 clustered
with MAL isolate and with MAL-like pol sequences of
NOGIL3 of Norway [27] and X327 of Spain [26]. In
V3, CU33 and CU34 grouped with X327 and with
two other subtype A sequences of the Republic of
Congo (Fig. 4). Pol sequences of these two African
viruses are not available and, therefore, it is not known
if they are also MAL-like in this segment.
Viruses of subtypes A and H in env, previously reported
in Cuba [5], grouped with D/A and U/H recombinant
viruses, respectively, of our study (Fig. 4).
Genetic–epidemiological correlations
There were differences in the prevalence of non-B
subtype viruses between groups defined by gender, risk
category and place of infection. Non-B subtype viruses
(including recombinants) were more frequent among
heterosexuals (72%) than among homo/bisexuals (37%)
(P,0.001), and among women (73%) than among
men (44%) (P,0.05). However, the prevalence of
non-B subtype viruses among heterosexual men (69%)
did not differ significantly from that in women.
Nine of ten infections (90%) reported to having been
acquired in Africa were with non-B viruses. Notably,
seven were non-recombinant. Subtypes of non-B
viruses acquired in Africa were: three C (two acquired
in Ethiopia and one in Zambia), one F1 (Angola), one
G (DRC), one H (Angola), one J (Angola) and two
MAL-like in pol and subtype A in env (both acquired in
DRC). Among infections acquired in Cuba, 46 (50%)
of 92 were with viruses of non-B subtypes.
Within subtype B and D/A recombinant clusters, there
were some subclusters, supported by high bootstrap
values, of viruses found in individuals sharing epi-
demiological features. One of these subclusters com-
prised 14 subtype B viruses (signalled with brackets in
Fig. 1), all of which corresponded to homo- or bisexual
men, 12 of them from Havana City, representing 46%
of subtype B viruses among individuals of this risk
Table 2. Distribution of recombinant HIV-1 genetic forms in Cuba.
Subtypes
Number of
samples pol (59–39)V3
21 D A
7UH
1BDA
1DBB
1 DUD H
1DH
1UB
1BUB
1UBB
1 U CRF01_AE
1GDA
1GDH
2 GBGB G
3UKA
Subtypes in pol (protease– reverse transcriptase) and env V3 region
of HIV-1 viruses from Cuba are shown, with the number of samples
of each genetic form, from a total of 105 analysed, shown on the left
column. (Two viruses, with pol subtypes G and C, respectively, that
could not be amplified in V3 are included in Table 1. Alternating
subtype segments in pol of recombinant viruses are placed in 59–39
order. U, unclassified segment.
Table 1. Distribution of non-recombinant HIV-1 genetic forms in
Cuba.
Number of samples Subtypes (pol and env V3)
50 B
5G
3C
1D
1F1
1H
1J
Subtypes in pol (protease– reverse transcriptase) and env V3 region
of HIV-1 viruses from Cuba are shown, with the number of samples
of each genetic form, from a total of 105 analysed, shown on the left
column. Two viruses, with pol subtypes G and C, respectively, that
could not be amplified in V3 are included here with non-recombi-
nant viruses.
HIV-1 diversity in Cuba Cuevas et al. 1649

category in this city. Within the D/A recombinant
cluster, five of the six viruses found in homo- or
bisexual men from Havana City formed a subcluster
supported by high bootstrap values (signalled with
brackets in Fig. 1). Genetic distances in phylogenetic
trees (Fig. 1) and dates of diagnosis (1996 or later in all
but one of the B subtype subcluster, and 1997 or 1998
in the D/A recombinant subcluster) were consistent
with relatively recent outbreaks, each originating from
a common source.
Discussion
Here we report the most extensive survey to date on
HIV genetic diversity in Cuba, with 105 samples
analysed from an estimated HIV-infected population of
approximately 2000 individuals. Previous studies were
limited to 15 individuals in 1995 [28] and a recently
published study of 11 individuals [5]. In both studies,
only the V3 region was analysed. In the first study, it
was stated that the predicted V3 amino acid sequences
were similar to those of subtype B reference isolates. In
the second study, V3 sequences were of subtypes B
(n ¼6), A (n ¼2), C (n ¼2) and H (n ¼1). In our
study, two segments of the viral genome, pol and env,
were analysed phylogenetically using neighbour-joining
trees and also by bootscanning to detect possible
intersubtype recombination. The results show that
approximately half (52%) of HIV-1 infections are
caused by non-B subtype or recombinant viruses, with
a high diversity of genetic forms, including non-
recombinant viruses of seven subtypes (B, C, D, F1, G,
H and J) and recombinant viruses of 14 different
genetic forms in 78% of the non-B samples, containing
segments of subtypes A, B, D, G, H and K and
CRF01-AE, as well as unclassified segments. Only one
of the recombinant forms, a virus related to a MAL-
like isolate reported in Spain (but probably acquired in
Africa) [26], has been reported previously. Neverthe-
less, in spite of this diversity, our results provide
phylogenetic and epidemiological evidence of only four
genetic forms (B and G subtypes and Dpol/Aenv and
Upol/Henv recombinants) currently circulating in Cuba.
The remaining viruses either were acquired in Africa
or were detected in only one or two individuals. D/A
recombinants (n ¼21) and U/H viruses (n ¼7)
grouped in respective monophyletic clusters in both
pol and env, suggesting that they are candidates for
recognition as circulating recombinant forms, pending
sequencing of full-length genomes [29]. These two
recombinant forms might be of African origin, since
the parental viruses were not detected in Cuba. In
GenBank, no sequences phylogenetically related to the
pol sequences of the U/H or D/A recombinants of
Cuba are found, but there are African viruses with env
sequences related to these recombinants (Fig. 4). pol
sequences of these African viruses are not available,
which does not allow us to confirm if recombinant
viruses related to those circulating in Cuba are present
in Africa.
The HIV-1 genetic diversity found in Cuba has no
parallel in the Americas. In other countries of the
Caribbean area, infections are almost uniformly with
B D/A U/H G UK/A C Others
47.6%
20% 6.7%
4.8%
2.8%
2.8%
15.3%
Other nonrecombinant
– 1 D
– 1 F1
– 1H
– 1 J
Other recombinant (pol/env)
– 1 BD/A
– 1 DB/B
– 1 DUD/H
– 1 D/H
– 1 U/B
– 1 BU/B
– 1 UB/B
– 1 U/01_AE
– 1 GD/A
– 1GD/H
– 2 GBGB/G
Fig. 3. Graphic depiction of the distribution of HIV-1 genetic forms in Cuba.
AIDS 2002, Vol 16 No 121650

subtype B, although occasional cases of infections with
other subtypes have been reported [30–32]. In the
American continent, non-B subtype infections are
relatively common only in Brazil, Argentina and
Uruguay, although these are limited to F and C
subtypes and BF recombinants in Brazil [1,2,33,34], BF
recombinants in Argentina [3] and F subtype (possibly
BF recombinants) and imported CRF01-AE viruses in
Uruguay [4,35].
A diversity of HIV-1 genetic forms of a degree
comparable to that found in Cuba has only been
reported in Central African countries, with the highest
group M diversity found in the DRC [8,36]. However,
considering that the estimated number of HIV infec-
tions in Cuba is only approximately 2000, the HIV
genetic diversity found in Cuba, in relative terms, has
no parallel in any other country.
Several factors may have contributed to the high HIV
genetic diversity in Cuba. Large numbers of Cuban
soldiers and civilian personnel were stationed in the
1970s and 1980s in Angola [6,14]. Although there are
no published studies on HIV-1 genetic diversity in
Angola, it would not be unexpected to find a high
HIV-1 genetic diversity in this country, since it is
bordering the DRC (in fact, the three infections that
were acquired in Angola were of three different
subtypes, F1, H and J). Additionally, large numbers of
‘internationalist’ Cuban aid workers, serving in several
sub-Saharan African countries, may have also contrib-
uted to the introduction of non-B genetic forms
[6,10,14]. In our study, only 10 of 105 infections were
acquired in Africa, indicating that non-B viruses intro-
duced from Africa have started to circulate in Cuba. A
superimposed factor contributing to the generation of
HIV genetic diversity in Cuba is recombination [37].
Phylogenetic analyses indicate that 12 infections are
with 11 different viruses generated by recombination
between the four genetic forms circulating in Cuba.
Considering the low prevalence of HIV infections in
Cuba, the diversity of recombinants generated locally is
disproportionately high. Although there is no direct
evidence of this, it could be suspected that recombina-
tion of HIV genetic forms would have been facilitated
by the former policy of prolonged quarantine of all
HIV-infected individuals in selected sanatoria [10,14].
This, together with the scarcity of condoms in Cuba in
earlier years [38], could have set conditions favourable
for coinfections with multiple genetic forms, with
subsequent generation of recombinants.
The finding of a high HIV diversity in Cuba may have
implications for vaccine design [39–42] and for the use
of tests for viral load determination [43,44] and for
detection of drug resistance mutations [45–47].
Although the prevalence of HIV infections in Cuba is
the lowest in the Americas, and travel of Cubans
outside the country is presently limited, the possibility
of exporting some of the multiple genetic forms of
Cuba to other countries may not be negligible, consid-
ering the expansion of tourism in Cuba in recent years,
with a concomitant increase in casual sexual contacts,
in relation with declining economic conditions
[14,38,48,49].
A recently reported HIV-1 complex recombinant virus
from Cameroon (CM53379) sequenced in the full-length
SIVcpzUS
CRF01_AE.90CF402
B.HXB2
D.NDK
K.MP535
C.ETH2220
F1.93BR020.1
G.DRCBL
F2.MP255
J.SE7022
A.SE7253
A.92UG037
A.Q23
A.U455
93CUPL10(CU)
97SE-1181(SN)
97CG282.24(CG)
X327(ES)
96NG-LUTBD083(NG)
H.V1991
H.V1997
H.90CF056
KTB52(CD)
KCD2(CD)
97SE-1212(SN)
BCB79(CD)
BCB80(CD)
KS38(CD)
KTB32(CD)
CA13(CM)
93CUPL08(CU)
93CUPL16(CU)
CU80
CU31
CU65
CU30
CU105
CU38
CU61
CU33
CU34
CU14
CU69
CU15
CU68
0.1
H
A
99
97
86
98
71
99
70
87
76
77
Fig. 4. Neighbour-joining tree of env V3 sequences of
viruses phylogenetically related to viruses of Cuba. Se-
quences of Cuba of our study are in boldface, and related
database sequences are underlined, identified by the name
of the isolate followed by the country two letter code in
parentheses. Bootstrap values 70% or greater of key nodes
are shown. Relevant clusters are signalled with brackets.
Subtypes with which the sequences of Cuba cluster are
indicated on the right of corresponding brackets. CD, Demo-
cratic Republic of Congo; CG, Republic of Congo; SN,
Senegal; NG, Nigeria; ES, Spain; CU, Cuba.
HIV-1 diversity in Cuba Cuevas et al. 1651

genome [50] clusters with high boostrap values, both in
pol and env, with the U (pol)/H (env) recombinants from
Cuba. This supports the Central African ancestry of these
Cuban viruses.
In summary, in contrast to the almost uniform pre-
dominance of subtype B in other Caribbean countries,
a high diversity of HIV-1 genetic forms, unprecedented
in the Americas, has been found in Cuba. Some of
these genetic forms were imported from Africa and
others were generated locally by recombination. The
causes of this diversity are related to historical circum-
stances peculiar to Cuba, which have contributed to
make HIV diversity in this country a mosaic of the
West (represented by B subtype sequences, presumably
of North American origin) and of Africa (represented
by multiple non-B subtype and recombinant viruses)
grafted into the Caribbean.
Acknowledgments
We thank Jose
´Esparza for his contribution to organiz-
ing the UNAIDS program under which this study was
done, and Francisco Parras for his support of this study.
Sponsorship: This work was financed by Technical
Service Agreement HQ/98/457048, UNAIDS, and by
grant 98BVII236 of Plan Nacional del SIDA, Ministerio
de Sanidad y Consumo, Spain.
References
1. Louwagie J, Delwart EL, Mullins JI, McCutchan FE, Eddy G, Burke
DS. Genetic analysis of HIV-1 isolates from Brazil reveals
presence of two distinct genetic subtypes. AIDS Res Hum Retro-
viruses 1994, 10:561– 567.
2. Morgado MG, Sabino EC, Shpaer EG et al. V3 region polymorph-
isms in HIV-1 from Brazil: prevalence of subtype B strains
divergent from North American/European prototype and detec-
tion of subtype F. AIDS Res Hum Retroviruses 1994, 10:
569– 576.
3. Thomson MM, Villahermosa ML, Va
´zquez-de-Parga E et al.
Widespread circulation of a B/F intersubtype recombinant form
among HIV- 1-infected individuals in Buenos Aires, Argentina.
AIDS 2000, 14:897– 899.
4. Russell KL, Carcamo C, Watts DM et al. Emerging genetic
diversity of HIV-1 in South America. AIDS 2000, 14:1785– 1791.
5. Go
´mez CE, Iglesias E, Perdomo W et al. Isolates from four
different HIV type 1 clades circulating in Cuba identified by
DNA sequence of the C2–V3 region. AIDS Res Hum Retro-
viruses 2001, 17:55– 58.
6. Torres-Anjel MJ. Macroepidemiology of the HIVs-AIDS (HAIDS)
pandemic. Insufficiently considered zoological and geopolitical
aspects. Ann N Y Acad Sci 1992, 653:257 –273.
7. Vidal N, Peeters M, Mulanga-Kabeya C et al. Unprecedented
degree of human immunodeficiency virus type 1 (HIV-1) group
M genetic diversity in the Democratic Republic of Congo
suggests that the HIV-1 pandemic originated in Central Africa.
J Virol 2000, 74:10498– 10507.
8. Scheper-Hughes N. AIDS, public health, and human rights in
Cuba. Lancet 1993, 342:965– 967.
9. Ubell RN. High-tech medicine in the Caribbean. 25 years of
Cuban health care. N Engl J Med 1983, 309:1468 –1472.
10. Pe
´rez-Stable EJ. Cuba’s response to the HIV epidemic. Am J
Public Health 1991, 81:563– 567.
11. UNAIDS. Epidemiological fact sheet on HIV/AIDS and sexually
transmitted infections. 2000 update. http://www.unaids.org.
2001.
12. Voelker R. HIV/AIDS in the Caribbean: big problems among
small islands. JAMA 2001, 285:2961– 2963.
13. Hsieh YH, Chen CW, Lee SM, de Arazoza H. On the recent
sharp increase in HIV detections in Cuba. AIDS 2001, 15:
426– 428.
14. Santana S, Faas L, Wald K. Human immunodeficiency virus in
Cuba: the public health response of a Third World country. Int J
Health Serv 1991, 21:511– 537.
15. Santana S. Article on HIV in Cuba criticized. Am J Public Health
1992, 82:899– 900.
16. Ruibal-Brunet IJ, Cuevas MT, Dı´az-Torres H et al. Genotypic
resistance mutations to antiretroviral drugs in HIV-1 B and non-
B subtypes from Cuba. Pan Am J Public Health 2001, 10:
174– 179.
17. Tenorio A, Echevarrı´a JE, Casas I, Echevarrı´a JM, Tabare
´sE.
Detection and typing of human herpesviruses by multiplex
polymerase chain reaction. J Virol Meth 1993, 44:261– 269.
18. Villahermosa ML, Thomson M, Va
´zquez-de Parga E et al. Im-
proved conditions for extraction and amplification of human
immunodeficiency virus type 1 RNA from plasma samples with
low viral load. J Hum Virol 2000, 3:27–34.
19. Leitner T, Korovina G, Marquina S, Smolskaya T, Albert J.
Molecular epidemiology and MT-2 cell tropism of Russian HIV
type 1 variant. AIDS Res Hum Retroviruses 1996, 12:
1595– 1603.
20. Werle E, Schneider C, Renner M, Volker M, Fiehn W. Convenient
single-step, one tube purification of PCR products for direct
sequencing. Nucl Acids Res 1994, 22:4354– 4355.
21. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG.
The CLUSTAL_X windows interface: flexible strategies for multi-
ple sequence alignment aided by quality analysis tools. Nucl
Acids Res 1997, 25:4876– 4882.
22. Hillis DM, Bull JJ. An empirical test of bootstrapping as a
method for assessing confidence in phylogenetic analysis. Syst
Biol 1993, 42:182– 192.
23. Alizon M, Wain-Hobson S, Montagnier L, Sonigo P. Genetic
variability of the AIDS virus: nucleotide sequence analysis of
two isolates from African patients. Cell 1986, 46:63 –74.
24. Robertson DL, Hahn BH, Sharp PM. Recombination in AIDS
viruses. J Mol Evol 1995, 40:249–259.
25. Gao F, Robertson DL, Carruthers CD et al. An isolate of human
immunodeficiency virus type 1 originally classified as subtype I
represents a complex mosaic comprising three different group
M subtypes (A, G, and I). J Virol 1998, 72:10234– 10241.
26. Thomson MM, Delgado E, Manjo
´nNet al. HIV-1 genetic
diversity in Galicia, Spain: BG intersubtype recombinant viruses
are circulating among injecting drug users. AIDS 2001, 15:
509– 516.
27. Jonassen TO, Grinde B, Asjo B, Hasle G, Hungnes O. Inter-
subtype recombinant HIV type 1 involving HIV-MAL-like and
subtype H-like sequence in four Norwegian cases. AIDS Res
Hum Retroviruses 2000, 16:49– 58.
28. Rolo FM, Miranda L, Wainberg MA et al. Envelope V3 region
sequences of Cuban HIV-1 isolates. J Acquir Immune Defic Syndr
Hum Retrovirol 1995, 9:123– 125.
29. Robertson DL, Anderson JP, Bradac JA et al. HIV-1 nomenclature
proposal. Science 2000, 288:55– 56.
30. Desgranges C, Fillon S, Audoly G et al. Presence of HIV-1
subtypes B and F and HTLV-I in HIV/HTLV coinfected indivi-
duals of Martinique. J Acquir Immune Defic Syndr Hum Retro-
virol 1996, 13:468– 470.
31. Ouka M, Menu E, Barin F et al. Introduction of HIV-1 subtype A
in the French West Indies. J Acquir Immune Defic Syndr Hum
Retrovirol 1998, 19:308– 310.
32. Flores I, Pieniazek D, Moran N et al. HIV-1 subtype F in single
and dual infections in Puerto Rico: a potential sentinel site for
monitoring novel genetic HIV variants in North America. Emerg
Infect Dis 1999, 5:481– 483.
33. Csillag C. HIV-1 subtype C in Brazil. Lancet 1994, 344:1354.
34. Ramos A, Tanuri A, Schechter M et al. Dual and recombinant
infections: an integral part of the HIV-1 epidemic in Brazil.
Emerg Infect Dis 1999, 5:65– 74.
AIDS 2002, Vol 16 No 121652

35. Artenstein AW, Coppola J, Brown AE et al. Multiple introductions
of HIV-1 subtype E into the western hemisphere. Lancet 1995,
346:1197– 1198.
36. Rambaut A, Robertson DL, Pybus OG, Peeters M, Holmes EC.
Human immunodeficiency virus. Phylogeny and the origin of
HIV-1. Nature 2001, 410:1047–1048.
37. Robertson DL, Sharp PM, McCutchan FE, Hahn BH. Recombina-
tion in HIV-1. Nature 1995, 374:124–126.
38. Veeken H. Cuba: plenty of care, few condoms, no corruption.
BMJ 1995, 311:935– 937.
39. Mascola JR, Louwagie J, McCutchan FE et al. Two antigenically
distinct subtypes of human immunodeficiency virus type 1: viral
genotype predicts neutralization serotype. J Infect Dis 1994,
169:48– 54.
40. Rowland-Jones SL, Dong T, Fowke KR et al. Cytotoxic T cell
responses to multiple conserved HIV epitopes in HIV- resistant
prostitutes in Nairobi. J Clin Invest 1998, 102:1758– 1765.
41. Dorrell L, Dong T, Ogg GS et al. Distinct recognition of non-
clade B human immunodeficiency virus type 1 epitopes by
cytotoxic T lymphocytes generated from donors infected in
Africa. J Virol 1999, 73:1708– 1714.
42. Cao H, Mani I, Vincent R et al. Cellular immunity to human
immunodeficiency virus type 1 (HIV-1) clades: relevance to HIV-
1 vaccine trials in Uganda. J Infect Dis 2000, 182:1350– 1356.
43. Burgisser P, Vernazza P, Flepp M et al. Performance of five
different assays for the quantification of viral load in persons
infected with various subtypes of HIV-1. Swiss HIV Cohort
Study. J Acquir Immune Defic Syndr 2000, 23:138– 144.
44. Jagodzinski LL, Wiggins DL, McManis JL et al. Use of calibrated
viral load standards for group M subtypes of human immuno-
deficiency virus type 1 to assess the performance of viral RNA
quantitation tests. J Clin Microbiol 2000, 38:1247–1249.
45. Vahey M, Nau ME, Barrick S et al. Performance of the
Affymetrix GeneChip HIV PRT 440 platform for antiretroviral
drug resistance genotyping of human immunodeficiency virus
type 1 clades and viral isolates with length polymorphisms.
J Clin Microbiol 1999, 37:2533–2537.
46. Lee CC, Thoe SY, Leo YS, Chan KP, Ling AE. Detection of
drug-selected mutations in HIV-1 subtype E reverse transcrip-
tase sequence using the line probe assay. AIDS 2000, 14:
1064– 1065.
47. Debyser Z, van Wijngaerden E, van Laethem K et al. Failure to
quantify viral load with two of the three commercial methods in
a pregnant woman harboring an HIV type 1 subtype G strain.
AIDS Res Hum Retroviruses 1998, 14:453– 459.
48. Burr C. Havana. Assessing Cuba’s approach to contain AIDS and
HIV. Lancet 1997, 350:647.
49. Barry M. Effect of the U.S. embargo and economic decline on
health in Cuba. Ann Intern Med 2000, 132:151– 154.
50. Carr JK, Torimiro JN, Wolfe ND et al. The AG recombinant
1bNG and novel strains of group M HIV-1 are common in
Cameroon. Virology 2001, 268:168– 181.
HIV-1 diversity in Cuba Cuevas et al. 1653