AIDS RESEARCH AND HUMAN RETROVIRUSESVolume 17, Number 12, 2001, pp. 1143–1154Mary Ann Liebert, Inc.

Natural Infection of Wild-Born Mandrills (Mandrillus sphinx)with Two Different Types of Simian Immunodeficiency Virus

JUN TAKEHISA,1 YOSUKE HARADA,2 NICAISE NDEMBI,1,3 INNOCENT MBOUDJEKA,2,3

YUKO TANIGUCHI,2 CHARLOTTE NGANSOP,3 SERAPHIN KUATE,3 LÉOPOLD ZEKENG,4

KENTARO IBUKI,2 TOSHIHIDE SHIMADA,2 BLAISE BIKANDOU,2 YUMI YAMAGUCHI-KABATA,2

TOMOYUKI MIURA,2 MIKIO IKEDA,5 HIROSHI ICHIMURA,1 LAZARE KAPTUÉ,3

and MASANORI HAYAMI2

ABSTRACT

We found a novel primate lentivirus in mandrill (Mandrillus sphinx). To clarify the evolutionary relationshipsand transmission patterns of human/simian immunodeficiency virus (HIV/SIV), we screened blood samplesfrom 30 wild-born healthy Cameroonian mandrills. Five (16.7%) of them were seropositive for SIV. ThreeSIV strains were isolated from the five seropositive mandrills by cocultivation of their peripheral bloodmononuclear cells (PBMCs) with PBMCs of rhesus macaques, a human T cell line (M8166), and/or a cynomol-gus macaque T cell line (HSC-F). One of the newly isolated SIV strains was intravenously inoculated into tworhesus macaques and resulted in chronic infection. In the SIV-infected macaques at 45 weeks after inocula-tion, we observed a mild decline in the number of peripheral CD41 lymphocytes, lymphadenopathy, and blas-tic follicular dendritic cells with mild follicular hyperplasia in the peripheral lymph nodes. A phylogeneticanalysis based on the pol sequence showed that the newly found SIVs from Cameroonian mandrills did notcluster with SIVmndGB1, which is the former representative strain of SIVmnd. The SIVmnds from Cameroonformed a new, independent lineage that branched before the root of the HIV-1/SIVcpz lineage with 996 of1000 bootstrap replications. They clustered host specifically, and exhibited about 16.9% diversity at the levelof nucleotide sequence among Cameroonian SIVmnd strains. These results indicate that the SIVmnds isolatedin Cameroon are a novel type of SIVmnd and have infected Cameroonian mandrills for a long time. We there-fore designated the Cameroonian SIVmnd as SIVmnd type 2 and redesignated SIVmndGB1 as SIVmnd type1. To date, M. sphinx is the only primate species other than humans that is naturally infected with two dif-ferent types of SIV.

1143

INTRODUCTION

SIMIAN IMMUNODEFICIENCY VIRUSES (SIVs) naturally infect anumber of species of nonhuman African primates. Along

with human immunodeficiency virus types 1 and 2 (HIV-1 andHIV-2), they are members of the primate lentivirus family. Onthe basis of full-length genomic sequences, the primatelentiviruses are classified into six major, approximately equidis-tant, phylogenetic lineages,1–3 namely, (1) SIV cpz4–7 in chim-panzees, which clusters with HIV-1; (2) SIVsm8–10 in sootymangabeys, which clusters with HIV-2; (3) SIVagm11–15 in four

species of African green monkey (Cercopithecus aethiops), (4)SIVmnd16 in mandrills (Mandrillus sphinx), SIVlhoest17,18 inl’Hoest monkeys (Cercopithecus lhoesti lhoesti) and SIVsun19

in a sun-tailed monkey (Cercopithecus lhoesti solatus), (5)SIVsyk20 in a Sykes monkey (Cercopithecus mitis albogularis),and (6) SIVcol21 in colobus monkeys (Colobus guereza). Ingeneral, SIVs cluster host specifically within an evolutionarytree, implying that they have infected their respective primatehosts for a long period of time,22 and have undergone “host-dependent” evolution.6,13,19

However, some infections have been reported to be the re-

1Department of Viral Infection and International Health, Graduate School of Medical Science, Kanazawa University, Kanazawa 920-8640,Japan.

2Laboratory of Viral Pathogenesis, Institute for Virus Research, Kyoto University, Sakyo-ku, Kyoto 606-8507, Japan.3Laboratoire d’Hematologie et d’Immunologie, Centre Hospitalier Universitaire, Yaoundé, Cameroon.4Laboratoire de Santé Hygiène Mobile, Ministère de la Santé, Yaoundé, Cameroon.5Fujirebio Incorporation, Chuo-Ku, Tokyo 103-0007, Japan.

sult of cross-species transmissions. For example, patas mon-keys, yellow baboons, and chacma baboons are “naturally” in-fected with SIVagm.23–25 SIVcpz and SIVsm appear to havebeen introduced to humans on multiple occasions, resulting inHIV-1 and HIV-2, respectively, which cause acquired immu-nodeficiency syndrome (AIDS) in humans,2,22,26–28 althoughSIVs have been found not to cause any diseases in their naturalhosts. It is also noted that SIVmnd in Mandrillus sphinx is clas-sified into the same lentivirus lineage as SIVlhoest and SIVsunin the same species, Cercopithecus lhoesti. Since the generaMandrillus and Cercopithecus are not closely related, eitherSIVmnd or SIVlhoest/SIVsun appears to have arisen throughcross-species transmission.

In 1988, we isolated two SIV strains (SIVmndGB1 and GB2)from mandrills in a semi-free-ranging colony in Franceville,Gabon,29 constructed an infectious molecular clone of SIVmndGB1, and sequenced it.16 Although several SIVmndstrains have been isolated from other mandrills within thecolony, all of the viruses have been found to be highly related,indicating intracolonial transmission of the same SIVmndstrain.30,31 Little information is available on SIVs from wild-living mandrills inhabiting other areas.

In the present study, to further clarify the evolutionary rela-tionships and transmission patterns of HIV and SIV, especiallySIVmnd, we attempted to isolate and characterize additionalSIV strains from mandrills in Cameroon. We found a novel typeof SIVmnd, and showed that a primate species other than hu-mans is naturally infected with two different types of SIVs.

MATERIALS AND METHODS

Monkeys

Blood samples were obtained from 30 wild-born pet man-drills in 1998 and 1999. All the mandrills were caught in thewild and housed in villages near the town of Ebolowa in southCameroon. Five- to 10-ml samples of peripheral blood werecollected from these mandrills on site under ketamine anesthe-sia (100 mg/kg), using acid–citrate–dextrose (ACD) anticoag-ulant solution, with permission of the owners.

Serological tests

Plasma samples collected from the mandrills were initiallyscreened for anti-SIV antibodies by a commercial particle ag-glutination (PA) test kit (Serodia-HIV, Fujirebio, Tokyo,Japan), which was found to detect most anti-SIV antibodies.All reactive specimens were confirmed by a Western blotting(WB) assay (New Lavblot HIV-1 and HIV-2, and Peptilav HIV-1 and HIV-2; Sanofi Diagnostic Pasteur, Marnes-la-Co-quette, France).

PCR amplification and sequence analysis of mandrillmitochondrial DNA

The Mandrillus genus has two species: Mandrillus sphinx(mandrill) and Mandrillus leucophaeus (drill). To confirm thespecies of the SIV-infected monkeys identified in this study,we amplified and sequenced the 424-bp fragment of mito-chondrial (mt) cytochrome b gene, which can be used to dis-

tinguish subspecies of old-world primates.23,33 The fragment(corresponding to nucleotides [nt] 14725–15148 of human mi-tochondrial DNA) was amplified in the DNA extracted fromperipheral blood mononuclear cells (PBMCs) of SIV-positivemonkeys, using a single-round polymerase chain reaction(PCR). The PCR was performed as previously described withprimer pair L14724 (59-CGAAGCTTGATATGAAAAAC-CATCGTTG-39) and H15149 (59-AAACTGCAGCCCCTCA-GAATGATATTTGTCCTCA-3 9).32,33

Virus isolation

PBMCs from the mandrills were separated by Ficoll gradi-ent sedimentation, and cocultivated with concanavalin A(ConA)-stimulated PBMCs of HIV/HTLV-negative humans(Asian and Caucasian), rhesus macaques, and African greenmonkeys. All cultures were maintained for 4 weeks, and the su-pernatants collected at various time points were examined forreverse transcriptase (RT) activity as previously described.34 Inaddition, to determine the cell tropism of the virus isolates, sev-eral cell lines were infected with the isolates, and RT activityin the culture supernatant was monitored as described above.The cell lines used were human T-lymphoid cell lines M8166,35

molt4#8,36 CEMx174,37 MT-238 U937, THP-1,39 and A3.01,40

a chimpanzee T-lymphoid cell line (ChM114), and a cynomol-gus macaque T-lymphoid cell line (HSC-F).

Amplification and sequencing of SIVmnd pol gene

Genomic DNA was extracted from PBMCs by using glassmilk (Prep-A-Gene DNA purification kit; Bio-Rad, Hercules,CA). A part of the pol region covering the integrase gene ofproviral DNA (corresponding to nt 3766–4482 in HIV-1LAI)was amplified by nested PCR with the primers unipol7 (59-AAAGTAAATATAGTGACAGATTCACAGTATG-3 9) andHIV-1pol4 (59-TCACCTTTCCAGAGGAGCTTTGCTGGTC-CTTTCC-39) in the first round, and unipol5 (59-TG-GGTACCAGCACACAAAGGAATAGGAGGAAA-3 9) andunipol6 (59-CCACAGCTGATCTCTGCCTTCTCTGTAATA-GACC-39) in the second round. Amplification was done withone cycle of 94°C for 3 min, and 35 cycles of 94°C for 30 sec,50°C for 30 sec, and 72°C for 1 min with a final extension of72°C for 10 min. A longer fragment containing the gag and polregions (corresponding to nt 974–4514 in HIV-1LAI) was am-plified by extra-long PCR using an XL-PCR kit (Perkin-Elmer,Foster City, CA) with the primers unigag (59-ATCAATGAG-GAAGCTGCAGAATGGGA-3 9) and HIV-1pol4. Cycling con-ditions included a hot start (2 min, 95°C), 20 cycles of denat-uration (95°C) for 30 sec and extension (68°C) for 10 min, andthen 17 cycles of denaturation (95°C) for 30 sec and extension(68°C) for 10 min with 15-sec increments per cycle. Cloningand DNA sequencing were performed as previously de-scribed.41

Genetic distance and phylogenetic analysis

DNA sequences were aligned by using CLUSTAL W (ver-sion 1.73),42 with subsequent inspection and manual modifica-tion. The frequency of nucleotide substitution in each base pairof the sequences was estimated by the two-parameter method.43

A phylogenetic tree was constructed by the neighbor-joining

TAKEHISA ET AL.1144

(NJ) method,44 and its reliability was estimated by 1000 boot-strap replications.45 The tree was visualized by using TreeView(version 1.53).46

Distance plots

The query sequences were first aligned with a set of refer-ence sequences representing all the established genetic lineagesof primate lentiviruses. The reference strains used were HIV-1groups M (LAI), N (YBF30), and O (MVP5180), SIVcpz (GABand ANT), SIVsm (H4), SIVagmVer (Tyol), SIVagmSab (1C),SIVmnd-1 (GB1), SIVsyk (SYK173), SIVcol (CGU1) and

SIVrcm (GAB1). The genetic distance (two-parameter) be-tween selected pairs of sequences was determined by movinga window of 300 bp along the genome alignment in 10-bp in-crements. The distance was plotted at the midpoint of the 300-bp segment.

Pathogenicity of SIVmnd in vivo

Two juvenile male rhesus macaques (MM208 and MM213)negative for SIV and simian T cell lymphotropic virus type I(STLV-I) were inoculated intravenously with 1 3 105 TCID50

(median tissue culture infective dose) of the virus isolate from

NOVEL TYPE OF SIVmnd-2 FROM MANDRILLS 1145

TABLE 1. SEROLOGICAL SURVEY OF MANDRILLS

Result of:

Subject HIV-1 PAa HIV-2 PA HIV-1 Western blotb HIV-2 Western blot HIV-1/2 Western blotc

cm16 ,32 1024 6 (p68, p52, p34, p25) 6 (gp140, gp105, p34, p26) 2cm31 512 512 6 (gp160, gp110/120, p18) 6 (gp140, gp105) 2cm32 ,32 512 6 (p25) 6 (gp105, p26) 2cm33 1024 512 6 (gp160, gp110/120, p25) 6 (gp105, p26) 2cm35 256 1024 6 (gp160, gp110/120) 6 (gp105) 2

aParticle agglutination (PA) titer against HIV-1 or HIV-2. The antibody titer was measured according to manufacturer instructions, and a titer ,32 was considered not reactive (Serodia-HIV; Fujirebio, Tokyo, Japan).

bNew Lavblot HIV-1 and HIV-2 (Sanofi Diagnostic Pasteur, Marnes-la-Coquette, France).cPeptilav HIV-1 and HIV-2 (Sanofi Diagnostic Pasteur).

FIG. 1. Western blots showing anti-SIV antibodies that cross-reacted with HIV-1 and HIV-2 antigens. HIV-1- and HIV-2-pos-itive (1) and HIV-1/2-negative (2) sera were supplied as controls by the manufacturer.

Cameroonian mandrills (SIVmnd). Peripheral blood sampleswere obtained from the macaques inoculated with the virus iso-lates at various time points before and after infection, and thenumber of CD41 T cells in PBMCs, the presence of the virus,the existence of proviral DNA, and the ability to produce anti-body to SIV were monitored longitudinally in the ma-caques.47,48 Virus isolation was done by coculture of the mon-key PBMCs with M8166 cells for 5 weeks, and virus recoverywas judged by RT activity in the culture supernatant as de-scribed above. Proviral DNA was detected by nested PCR asdescribed above, using the pol primers HIV-1pol3 (59-TAAAAGGAGAAGCCATGCATGGACAAGTAGA-3 9) andHIV-1pol4 in the first round and unipol1 (59-AGTGGA-TTCATAGAAGCAGAAGT-3 9) and HIV-1pol2 (59-CCC-CAATCCCCCCTTTTCTTTTAAAA-3 9) in the second round.Antibody responses were monitored by using a PA test kit asdescribed above. These monkeys were housed throughout theexperimental period, and autopsies were carried out under the

regulations approved by the Institutional Animal Care and UseCommittee of the Institute for Virus Research (Kyoto Univer-sity, Kyoto, Japan).

RESULTS

Identification of SIV-seropositive mandrills

Plasma samples collected from household mandrills inCameroon were tested for the presence of anti-SIV antibodies,using the PA test kits. Of 30 Cameroonian mandrills, 5 (16.7%)had antibodies that cross-reacted to commercially available HIVantigens (Ag). Of these five monkeys, two (cm16 and cm32)had antibodies that cross-reacted with HIV-2 Ag, and three(cm31, cm33, and cm35) had antibodies that cross-reacted withboth HIV-1 and HIV-2 Ag (Table 1). None of the plasma sam-ples cross-reacted with HIV-1 Ag only. The five PA-positive

TAKEHISA ET AL.1146

FIG. 2. A phylogenetic tree of mitochondrial DNA sequences. The tree was constructed on the basis of a part of the cytochromeb (396 bp) sequence. Mandrillus mitochondrial DNA sequences previously reported were included. SIV-infected Cameroonianmandrills (cm16, cm31, cm32, cm33, and cm35) identified in this study are boxed. Brackets indicate the species classificationspreviously defined. Bootstrap values of key nodes in 1000 replications are also shown.

plasma samples were then examined by WB assay (Fig. 1 andTable 1), which showed several bands that faintly reacted withHIV-1 and/or HIV-2 Ag. These data indicated that the five mon-keys (cm16, cm31, cm32, cm33, and cm35) were infected withSIV. These five SIV-infected monkeys showed no evidence oflymphadenopathy, wasting, or other signs of illness.

Identification of mandrill species

Phylogenetic analysis of the mtDNA sequences of the fivemonkeys infected with SIV showed that they all were Man-drillus sphinx (Fig. 2).

Isolation and cell tropism of the virus

Virus was successfully isolated from three of the five seropos-itive mandrills (cm16, cm33, and cm35) by cocultivation of theirPBMCs with those of rhesus macaques (Table 2). The newly iso-lated viral strains were designated as 98CM16, 99CM33, and99CM35. To determine their host range, their growth in variousCD41 T cell lines was assayed by RT activity of the supernatant.At 5 weeks after infection, each of these strains were able to growrelatively well in M8166 cells, a human T cell line, moderatelywell in HSC-F cells, a cynomolgus macaque T cell line, and notat all in the other cell lines (Table 2). No clear cytopathic effectwas observed in either cell line.

Phylogenetic analysis of Cameroonian SIVmnd

To clarify the evolutionary relationships between Cameroon-ian SIVs and other primate lentiviruses, a phylogenetic treebased on the 717-bp-long DNA fragment of the integrase genewas constructed by the NJ method (Fig. 3A). Five Cameroon-ian virus strains (98CM16, 99CM31, 99CM32, 99CM33, and99CM35) did not cluster with SIVmndGB1, which until nowhas been considered the representative strain of SIVmnd.

SIVmndGB1 was isolated in Gabon16 and forms a phylogeneticlineage with SIVlhoest/SIVsun. These five strains formed anew, independent lineage that branched before the root of theHIV-1/SIVcpz lineage with 996/1000 bootstrap replications.These strains exhibited relatively high diversities (16.9% on av-erage with a range from 13.9 to 21.8%) at the level of nucleo-tide sequence. The nucleotide homology between 98CM16 andSIVmndGB1 in the integrase region was only 67.0%. In viewof the fact that the genetic homology between HIV-1 and HIV-2 is 65.4%, the 67.0% homology between 98CM16 andSIVmndGB1 indicates that these strains are different types ofSIVmnd. Therefore, we designated the Cameroonian SIVmndas SIVmnd type 2 (SIVmnd-2), and redesignated SIVmndGB1as SIVmnd type 1 (SIVmnd-1GB1). Over the integrase region,SIVmnd-298CM16 was most closely related to SIVrcm from ared-capped mangabey isolated in Gabon,49 with 79.5% nucle-otide sequence identity, and were also closely related to HIV-1LAI, with 74.2% nucleotide sequence identity.

To further clarify the evolutionary relationship betweenSIVmnd-2 and SIVdrl obtained from a drill (Mandrillus leu-cophaeus) in Nigeria,50 a phylogenetic tree based on the 787-bp-long DNA fragment of the RT gene was constructed.SIVmnd-298CM16 was more closely related to SIVdrl than toSIVrcm (Fig. 3B), and the levels of nucleotide homology be-tween SIVmnd-298CM16 and SIVdrl and SIVrcm were 81.8 and76.5%, respectively.

Distance plots of the gag–pol region

Distance plots were constructed to investigate the extent ofsequence difference between SIVmnd-298CM16 and other pri-mate lentiviruses across nearly 3550 bp of the gag and pol re-gions. Figure 4 shows the distance values between SIVmnd-298CM16 and 12 HIV/SIV reference strains. SIVmnd-2 wasclearly differentiated from SIVmnd-1GB1. SIVrcm basically ex-

NOVEL TYPE OF SIVmnd-2 FROM MANDRILLS 1147

TABLE 2. HOST RANGES OF THE NEWLY ISOLATED

VIRUSES FROM CAMEROONIAN MANDRILLS

RT activity (cpm/ml)a

Source 98CM16 98CM33 99CM35

PBMCsHuman (Asian) — — —Human (Caucasian) — — —Rhesus macaque 377 256 312African green monkey — — —

Cell line (human)M8166 3048 856 1861Molt4#8 — — —CEMx174 — — —MT-2 — — —U937 — — —THP-1 — — —A3.01 — — —

Cell line (chimpanzee)ChM114 — — —

Cell line (cynomolgus macaque)HSC-F 1588 452 1125

aRT activities represent peak values.

hibited the lowest distance throughout the gag and pol regions,although the pattern of distance between SIVmnd-298CM16 andSIVrcm was not continuously uniform.

Infectivity of SIVmnd-2 in vivo

To investigate the potential infectivity of SIVmnd-2, two ju-venile rhesus macaques, MM208 and MM213, were infectedintravenously with SIVmnd-298CM16, and were monitored forup to 45 weeks after infection with the virus. The virologicaland immunological states of the macaques before and after virusinfection are shown in Table 3 and Fig. 5.

In the SIVmnd-infected macaques, the number of peripheral

CD41 lymphocytes decreased at 1 week postinfection, recov-ered to the level before infection by 3 weeks postinfection andgradually decreased to the level of 1 week postinfection by theend of the follow-up period. The virus was recovered on threeoccasions from both MM208 (at 1, 2, and 4 week postinfec-tion) and from MM213 (at 1, 2, and 6 week postinfection).Proviral DNA was detected by PCR at 1 week postinfection inboth MM208 and MM213. The anti-SIVmnd antibody was de-tected starting at 2 weeks postinfection in MM208 and startingat 3 weeks postinfection in MM213, and continued to be de-tected in both macaques until the end of the follow-up period.However, MM208 had a higher titer of the antibody thanMM213 did.

TAKEHISA ET AL.1148

FIG. 3. Phylogenetic trees of the SIVmnd-2 strains based on a part of the pol sequence covering the integrase gene (717 bp)(A) and a part of the pol sequence covering the 59 RT gene (787 bp) (B). The sequences of the SIVmnd-2 strains are boxed andthe sequence SIVmnd-1GB1 is circled. Six primate lentiviral lineages are indicated by brackets. Bootstrap values of key nodes in1000 replications are also shown.

Autopsies of the SIVmnd-infected macaques were carriedout at 45 weeks postinfection, when both macaques showed nosigns of illness other than lymphadenopathy. In hema-toxylin–eosin staining of the lymph node sections from the ma-caques, a mild follicular hyperplasia was observed, but involu-tion was not. Blastic follicular dendritic cells were also foundin germinal centers (data not shown).

DISCUSSION

In the present study, we found a novel primate lentivirus,SIVmnd type 2 (SIVmnd-2), that naturally infected wild-bornmandrills in Cameroon. SIVmnd-2 was found to be phyloge-

netically different from SIVmnd-1GB1,16 which is the formerrepresentative strain of SIVmnd. SIVmnd-2 thus constitutes anew group of primate lentiviruses. This is the first report of anonhuman primate that is naturally infected with two differenttypes of SIVs.22

The extent of sequence diversity among the SIVmnd-2strains was about 16.9% in the pol region, which encodes in-tegrase (Fig. 3A). This region is known to be the most con-served region across the viral genome.51,52 The diversity amongthe SIVmnd-2 strains was almost as high as the diversity amongthe strains of three other viruses: SIVsm,9,10 SIVagmVer ofvervet monkeys,13 and SIVlhoest.18 Furthermore, the diversityamong the SIVmnd-2 strains was even greater than that amongSIVcpz strains of Pan troglodytes troglodytes.6,7 In addition,

NOVEL TYPE OF SIVmnd-2 FROM MANDRILLS 1149

FIG. 3. Continued.

TAKEHISA ET AL.1150

FIG

. 4.

Dis

tanc

e pl

ots

com

pari

ng S

IVm

nd-2

98C

M16

with

rep

rese

ntat

ives

of

prim

ate

lent

ivir

uses

(H

IV-1

gro

ups

M, N

, and

O, S

IVcp

zGA

B, S

IVcp

zAN

T, S

IVsm

, SIV

agm

Ver

,S

IVag

mS

ab,

SIV

mnd

-1, S

IVsy

k, S

IVco

l, an

d S

IVrc

m).

The

ali

gnm

ent

was

sec

tione

d in

to 3

00-n

t se

gmen

ts t

hat

wer

e m

oved

in

step

s of

10

nucl

eotid

es.

The

gen

etic

dis

tanc

esbe

twee

n SI

Vm

nd-2

98C

M16

and

each

of

the

resp

ecti

ve r

efer

ence

str

ains

wer

e ca

lcul

ated

by

the

two-

para

met

er m

etho

d. T

he d

ista

nce

valu

e fo

r ea

ch s

egm

ent

was

plo

tted

at t

hem

idpo

int

of t

he s

egm

ent.

The

seg

men

ts o

f th

e ph

ylog

enet

ic t

ress

in

Fig

. 3

are

indi

cate

d w

ith

bars

(de

sign

ated

as

3Aan

d 3B

).

the seroprevalence of SIVmnd-2 was 16.7% among the man-drills inhabiting Cameroon, which is located in the northern partof their habitat. These results suggest that Cameroonian man-drills have been infected with SIVmnd-2 for a long time, andthat SIVmnd-2 has evolved “host dependently,” indicating thatCameroonian mandrills are a natural reservoir for SIVmnd-2.

Phylogenetic analysis showed that SIVmnd-2 clustered to-gether with SIVdrl from a drill (Fig. 3B), which belongs to thesame genus as mandrill, Mandrillus. This suggests thatSIVmnd-2 and SIVdrl have evolved “genus dependently.” Thegenus-dependent evolution may suggest that the ancestors ofmandrill and drill had been infected with SIV by the time ofthe division into the two distinct species, and that the SIV hasevolved “species dependently” and become SIVmnd-2 and SIVdrl, respectively. However, molecular clock extrapolationsestimated the date of emergence of the common ancestor of sixmajor lineages of primate lentiviruses to be about 200 years

ago,53 although variations in the rate of evolution among dif-ferent HIV and SIV could contribute to the error in the molec-ular clock estimation. Furthermore, mandrills and drills inhabitgeographically close areas, the west coastal regions of CentralAfrica. It may, therefore, be more likely that the cross-speciestransmission of SIV between mandrills and drills occurred after the division into the two distinct species and then SIVmnd-2 and SIVdrl have evolved “species dependently.”

The SIVmnd-2 lineage was found to branch before the rootof the HIV-1/SIVcpz lineage (Fig. 3). In fact, SIVmnd-2showed relatively high sequence homology with SIVrcm49 inthe integrase (79.5%) and RT (76.5%) genes. Across thegag–pol region, the distance between SIVmnd-298CM16 andSIVrcm was much smaller than those between SIVmnd-298CM16

and the other HIV/SIV reference strains, although the patternand extent of nucleotide sequence divergence between SIVmnd-298CM16 and other primate lentiviruses were not uni-

NOVEL TYPE OF SIVmnd-2 FROM MANDRILLS 1151

TABLE 3. VIROLOGICAL STATUS AND ANTIBODY RESPONSE OF SIVMND-298CM16-INFECTED MONKEYS

Weeks postinoculation

Macaque Method 1 2 3 4 6 8 12 16 20 24 45

MM208 Virus isolation 1 1 2 1 2 2 2 2 2 2 2Proviral DNAa 1 1 1 1 1 1 1 1 1 1 1Antibody to HIV-2b ,32 64 1,024 1,024 2,048 2,048 4,096 4,096 .16,384 4,096 8,192

MM213 Virus isolation 1 1 2 2 1 2 2 2 2 2 2Proviral DNA 1 1 1 1 1 1 1 1 1 1 1Antibody to HIV-2 ,32 ,32 64 64 256 256 256 512 256 128 256

a1, Positivity of DNA PCR for detecting proviral SIVmnd-2.bPA titer against HIV-2 (Serodia-HIV, Fujirebio, Tokyo, Japan). Two macaques never developed a reaction to HIV-1 by PA.

FIG. 5. Change in the number of peripheral (CD31) CD41 T lymphocytes in two macaques before and after SIVmnd-2 in-fection.

form (Fig. 4). These results suggest that cross-species trans-mission of SIVs between mandrills and red-capped mangabeysoccurred in the past, and that SIVmnd-2 could have originatedfrom the same ancestor of the SIVrcm and HIV-1/SIVcpz lin-eages. To further clarify the evolutionary relationships andtransmission patterns between SIVmnd-2 and other primatelentiviruses including HIVs, we are now sequencing the fullgenomes of the SIVmnd-2 strains. We are also trying to isolatemore strains of SIVmnd from mandrills in other parts of theirhabitat in western Central Africa.

SIVs appear to result in an asymptomatic infection in theirnatural host.2,26,54 Such SIVs include SIVsm in sootymangabeys and SIVagm in African green monkeys. Simi-larly, mandrills infected with SIVmnd-2 do not appear to develop any clinical symptoms. However, SIVsm andSIVagm have been reported to induce AIDS-like symptomsin unnatural primate hosts, such as Asian macaques,17,55–58

after experimental infection. In our study, a persistent in-fection with SIVmnd-298CM16 was established in rhesus ma-caques after intravenous virus inoculation. This was expectedon the basis of the in vitro susceptibility of macaque PBMCsto SIVmnd-2. At autopsy, 45 weeks after inoculation, a milddecline in the number of peripheral CD41 lymphocytes, lymphadenopathy, and blastic follicular dendritic cells withmild follicular hyperplasia in peripheral lymph nodes were observed in both of the chronically infected macaques.After additional in vivo passages, SIVmnd-2 might be ableto induce a typical AIDS-like disease in macaques, whichwould provide some clues to clarify the pathogenesis of SIVs in unnatural primate hosts after cross-species trans-mission.

Studies have revealed that SIVcpz has been introduced intohumans at least three times, resulting in HIV-1 groups M, N,and O.22 The cross-species transmission of SIVcpz is nowthought to have occurred through humans being exposed to theblood of chimpanzees infected with SIVcpz during hunting andbutchering of chimpanzees in western Central Africa. In thisregion, mandrills, drills, mangabeys, and other monkeys are reg-ularly hunted along with chimpanzees. Thus, cross-speciestransmission of SIVs to humans continues to be a major con-cern as a potential source of generating new human viruses. Toprevent further introduction of SIVs into humans and the ap-pearance of a new “HIV” as well as to clarify the evolutionaryrelationships among primate lentiviruses, further study of hu-man and simian lentiviral infections is needed, specially in west-ern Central Africa, where there is frequent contact between hu-mans and other primates.

ACKNOWLEDGMENTS

This work was supported in part by International ScientificResearch Program grant 12376002 from Monbu-kagaku-sho(Ministry of Education) and by the Research Fellowships ofJapan Society for the Promotion of Science for Young Scien-tists. We are grateful to Dr. Preston A. Marx of Tulane Uni-versity and Dr. Beatrice H. Hahn of the University of Alabamaat Birmingham for providing the unpublished SIVrcm sequenceand for helpful suggestions.

SEQUENCE DATA

GeneBank accession numbers AF301607 to AF301611,AF301612 to AF301611, and AF367411 have been assigned tothe sequences for SIVmnd-2 (integrase), mtDNA, and SIVmnd-2 (gag–pol), respectively, obtained in this study.

REFERENCES

1. Sharp PM, Robertson DL, Gao F, and Hahn BH: Origins and di-versity of human immunodeficiency viruses. AIDS 1994;8:S27–S42.

2. Hayami M, Ido E, and Miura T: Survey of simian immunodefi-ciency virus among nonhuman primate populations. Microbiol Im-munol 1994;188:1–20.

3. Kuiken C, Foley B, Hahn BH, et al.: Human Retroviruses andAIDS: A Compilation and Analysis of Nucleic Acid and Amino AcidSequences. Los Alamos National Laboratory, Los Alamos, NewMexico, 1999.

4. Huet T, Cheynier R, Meyerhans A, Roelants G, and Wain-HobsonS: Genetic organization of a chimpanzee lentivirus related to HIV-1. Nature 1990;345:356–359.

5. Vanden Haesevelde MM, Peeters M, Jannes G, et al.: Sequenceanalysis of a highly divergent HIV-1-related lentivirus isolatedfrom a wild captured chimpanzee. Virology 1996;221:346–350.

6. Gao F, Bailes E, Robertson DL, et al.: Origin of HIV-1 in the chim-panzee Pan troglodytes troglodytes. Nature 1999;397:436–441.

7. Corbet S, Muller-Trutwin MC, Versmisse P, et al.: env sequencesof simian immunodeficiency viruses from chimpanzees inCameroon are strongly related to those of human immunodefi-ciency virus group N from the same geographic area. J Virol 2000;74:529–534.

8. Hirsch VM, Olmsted RA, Murphey-Corb M, Purcell RH, and John-son PR: An African primate lentivirus (SIVsm) closely related toHIV-2. Nature 1989;339:389–392.

9. Chen Z, Telfer P, Gettie A, et al.: Genetic characterization of newWest African simian immunodeficiency virus SIVsm: Geographicclustering of household-derived SIV strains with human immuno-deficiency virus type 2 subtypes and genetically diverse virusesfrom a single feral sooty mangabey troop. J Virol. 1996;70:3617–3627.

10. Chen Z, Luckay A, Sodora DL, et al.: Human immunodeficiencyvirus type 2 (HIV-2) seroprevalence and characterization of a dis-tinct HIV-2 genetic subtype from the natural range of simian im-munodeficiency virus-infected sooty mangabeys. J Virol 1997;71:3953–3960.

11. Fukasawa M, Miura T, Hasegawa A, et al.: Sequence of simianimmunodeficiency virus from African green monkey, a new mem-ber of the HIV/SIV group. Nature 1988;333:457–461.

12. Fomsgaard A, Hirsch VM, Allan JS, and Johnson PR: A highly di-vergent proviral DNA clone of SIV from a distinct species ofAfrican green monkey. Virology 1991;182:397–402.

13. Allan JS, Short M, Taylor ME, et al.: Species-specific diversityamong simian immunodeficiency viruses from African green mon-keys. J Virol 1991;65:2816–2828.

14. Jin MJ, Hui H, Robertson DL, et al.: Mosaic genome structure ofsimian immunodeficiency virus from West African green monkeys.EMBO J 1994;13:2935–2947.

15. Soares MA, Robertson DL, Hui H, Allan JS, Shaw GM, and Hahn BH: A full-length and replication-competent proviral cloneof SIVAGM from tantalus monkeys. Virology 1997;228:394–399.

16. Tsujimoto H, Hasegawa A, Maki N, et al.: Sequence of a novel

TAKEHISA ET AL.1152

simian immunodeficiency virus from a wild caught African man-drill. Nature 1989;341:539–541.

17. Hirsch VM, Campbell BJ, Bailes E, et al.: Characterization of anovel simian immunodeficiency virus (SIV) from l’Hoest monkeys(Cercopithecus l’hoesti): Implications for the origins of SIVmndand other primate lentiviruses. J Virol 1999;73:1036–1045.

18. Beer BE, Bailes E, Dapolito G, et al.: Patterns of genomic sequencediversity among their simian immunodeficiency viruses suggestthat l’Hoest monkeys (Cercopithecus lhoesti) are a naturallentivirus reservoir. J Virol 2000;74:3892–3898.

19. Beer BE, Bailes E, Goeken R, et al.: Simian immunodeficiencyvirus (SIV) from sun-tailed monkeys (Cercopithecus solatus): Ev-idence for host-dependent evolution of SIV within the C. lhoestisuperspecies. J Virol 1999;73:7734–7744.

20. Hirsch VM, Dapolito GA, Goldstein S, et al.: A distinct Africanlentivirus from Sykes’ monkeys. J Virol 1993;67:1517–1528.

21. Courgnaud V, Pourrut X, Bibollet-Ruche F, et al.: Characteriza-tion of a novel simian immunodeficiency virus from guerezacolobus monkeys (Colobus guereza) in Cameroon: A new lineagein the nonhuman primate lentivirus family. J Virol 2001;75:857–866.

22. Hahn BH, Shaw GM, De Cock KM, and Sharp PNM: AIDS as azoonosis: Scientific and public health implications. Science2000;287:607–614.

23. Bibollet-Ruche F, Galat-Luong A, Cuny G, et al.: Simian immu-nodeficiency virus infection in a patas monkey (Erythrocebuspatas): Evidence for cross-species transmission from African greenmonkeys (Cercopithecus aethiops sabaeus) in the wild. J Gen Vi-rol 1996;77:773–781.

24. Jin MJ, Rogers J, Phillips-Conroy JE, et al.: Infection of a yellowbaboon with simian immunodeficiency virus from African greenmonkeys: Evidence for cross-species transmission in the wild. JVirol 1994;68:8454–8460.

25. van Rensburg EJ, Engelbrecht S, Mwenda J, et al.: Simian immu-nodeficiency viruses (SIVs) from eastern and southern Africa: De-tection of a SIVagm variant from a chacma baboon. J Gen Virol1998;78:1809–1814.

26. Barré-Sinoussi F: HIV as the cause of AIDS. Lancet 1996;348:31–35.

27. Hillis DM: AIDS. Origins of HIV. Science 2000;288:1757–1759.28. Korber B, Muldoon M, Theiler J, et al.: Timing the ancestor of the

HIV-1 pandemic strains. Science 2000;288:1789–1796.29. Tsujimoto H, Cooper RW, Kodama T, et al.: Isolation and char-

acterization of simian immunodeficiency virus from mandrills inAfrica and its relationship to other human and simian immunode-ficiency viruses. J Virol 1988;62:4044–4050.

30. Cooper R, Feistner A, Evans S, Tsujimoto H, and Hayami M: Alack of evidence of sexual transmission of a simian immunodefi-ciency agent in a semifree-ranging group of mandrills. AIDS1989;3:764.

31. Nerrienet E, Amouretti X, Muller-Trutwin MC, et al.: Phyloge-netic analysis of SIV and STLV type I in mandrills (Mandrillussphinx): Indications that intracolony transmissions are predomi-nantly the result of male-to-male aggressive contacts. AIDS ResHum Retroviruses 1998;14:785–796.

32. Irwin DM, Kocher TD, and Wilson AC: Evolution of the cy-tochrome b gene of mammals. J Mol Evol 1991;32:128–144.

33. Zhang YP and Ryder OA: Mitochondrial cytochrome b gene se-quences of Old World monkeys: With special reference on evolu-tion of Asian colobines. Primates 1998;39:39–49.

34. Willey RL, Smith DH, Laskey LA, et al.: In vitro mutagenesisidentifies a region within the envelope gene of the human immu-nodeficiency virus that is critical for infectivity. J Virol 1988;62:139–147.

35. Clapham RP, Weiss RA, Dalgleish AG, Exley M, Withby D, and

Hogg N: Human immunodeficiency virus infection of monocyticand T-lymphocytic cells: Receptor modulation and differentiationinduced by phorbol ester. Virology 1987;158:44–51.

36. Kikukawa R, Koyanagi Y, Harada S, Kobayashi N, Hatanaka M,and Yamamoto N: Differential susceptibility to the acquired im-munodeficiency syndrome retrovirus in cloned cells of human leu-kemic T-cell line Molt-4. J Virol 1986;57:1159–1162.

37. Salter RD, Howell DN, and Cresswell P: Genes regulating HLAclass I antigen expression in T–B lymphoblast hybrids. Immuno-genetics 1985;21:235–246.

38. Miyoshi I, Kubonishi I, Yoshimoto S, and Shiraishi Y: A T-cellline derived from normal human cord leukocytes by co-culturingwith human leukemic T-cells. Gann 1981;72:978–981.

39. Tsuchiya S, Yamabe M, Yamaguchi Y, Kobayashi Y, Konno T,and Tada K: Establishment and characterization of a human acutemonocytic leukemia cell line (THP-1). Int J Cancer 1980;26:171–176.

40. Folks T, Benn S, Rabson A, et al.: Characterization of a continu-ous T-cell line susceptible to the cytopathic effects of the acquiredimmunodeficiency syndrome (AIDS)-associated retrovirus. ProcNatl Acad Sci USA 1985;82:4539–4543.

41. Takehisa J, Zekeng L, Ido E, et al.: Human immunodeficiency virustype 1 intergroup (M/O) recombination in Cameroon. J Virol 1999;73:6810–6820.

42. Thompson JD, Higgins DG, and Gibson TJ: CLUSTAL W: Im-proving the sensitivity of progressive multiple sequence alignmentthrough sequence weighting, position-specific gap penalties andweight matrix choice. Nucleic Acids Res 1994;22:4673–4680.

43. Kimura M: A simple method for estimating evolutionary rates ofbase substitutions through comparative studies of nucleotide se-quences. J Mol Evol 1980;16:111–120.

44. Saito N and Nei M: The neighbor-joining method: A new methodfor reconstructing phylogenetic trees. Mol Biol Evol 1987;4;406–425.

45. Felsenstein J: Confidence limits on phylogenies: An approach us-ing the bootstrap. Evolution 1985;39:783–791.

46. Page RD: TreeView: An application to display phylogenetic treeson personal computers. Comput Appl Biosci 1996;12:357–358.

47. Igarashi T, Kuwata T, Takehisa J, et al.: Geonomic and biologicalalteration of a human immunodeficiency virus type 1 (HIV-1)–simian immunodeficiency virus strain mac chimera, with HIV-1Env, recovered from a long-term carrier monkey. J Gen Virol1996;77:1649–1658.

48. Steger KK, Dykhuizen M, Mitchen JL, et al.: CD41-T-cell andCD201-B-cell changes predict rapid disease progression aftersimian-human immunodeficiency virus infection in macaques. JVirol 1998;72:1600–1605.

49. Georges-Courbot MC, Lu CY, Makuwa M, et al.: Natural infec-tion of a household pet red-capped mangabey (Cercocebus torqua-tus torquatus) with a new simian immunodeficiency virus. J Virol1998;72:600–608.

50. Clewley J, Lewis J, Brown D, and Gadsby E: A novel simian im-munodeficiency virus (SIVdrl) pol sequence from the drill mon-key, Mandrillus leucophaeus. J Virol 1998;72:10305–10309.

51. Miura T, Sakuragi J, Kawamura M, et al.: Establishment of a phy-logenetic survey system for AIDs-related lentivirus and demon-stration of a new HIV-2 subgroup. AIDS 1990;4:1257–1261.

52. Takehisa J, Zekeng L, Ido E, et al.: Various types of HIV mixed-infections in Cameroon. Virology 1998;245:1–10.

53. Korber B, Theiler J, and Wolinsky S: Limitations of a molecularclock applied to considerations of the origins of HIV-1. Science1998;280:1868–1871.

54. Rey-Cuille MA, Berthier JL, Bomsel-Demontoy MC, et al.: Simianimmunodeficiency virus replicates to high levels in sootymangabeys without inducing disease. J Virol 1998;72:3872–3886.

NOVEL TYPE OF SIVmnd-2 FROM MANDRILLS 1153

55. Daniel MD, Letvin NL, King NW, et al.: Isolation of T-cell tropicHTLV-III-like retrovirus from macaques. Science 1985;228:1201–1204.

56. Letvin NL, Daniel MD, Sehgal PK, et al.: Induction of AIDS-likedisease in macaque monkeys with T-cell tropic retrovirus STLV-III. Science 1985;230:71–73.

57. Murphey-Corb M, Martin LN, Rangan SR, et al.: Isolation of anHTLV-III-related retrovirus from macaques with simian AIDS andits possible origin in asymptomatic mangabeys. Nature 1986;321:435–437.

58. Hirsch VM, Dapolito G, Johnson PR, et al.: Induction of AIDS bysimian immunodeficiency virus from an African green monkey:

Species-specific variation in pathogenicity correlates with the ex-tent of in vivo replication. J Virol 1995;69:955–967.

Address reprint requests to:Jun Takehisa

Department of Viral Infection and International HealthGraduate School of Medical Science

Kanazawa UniversityKanazawa 920-8640, Japan

E-mail: juntak@pop.m.kanazawa-u.ac.jp

TAKEHISA ET AL.1154