Immunodeficiency in the Absence of High Viral Load in Pig-Tailed Macaques Infected with Simian Immunodeficiency Virus SIVsun or SIVlhoest

Abstract

Simian immunodeficiency virus (SIV) is known to result in an asymptomatic infection of its natural African monkey host. However, some SIV strains are capable of inducing AIDS-like symptoms and death upon experimental infection of Asian macaques. To further investigate the virulence of natural SIV isolates from African monkeys, pig-tailed (PT) macaques were inoculated intravenously with either of two recently discovered novel lentiviruses, SIVlhoest and SIVsun. Both viruses were apparently apathogenic in their natural hosts but caused immunodeficiency in PT macaques. Infection was characterized by a progressive loss of CD4⁺ lymphocytes in the peripheral blood and lymph nodes, generalized lymphoid depletion, a wasting syndrome, and opportunistic infections, such as Mycobacterium avium or Pneumocystis carinii infections. However, unlike SIVsm/mac infection of macaques, SIVlhoest and SIVsun infections in PT macaques were not accompanied by high viral loads during the chronic disease stage. In addition, no significant correlation between the viral load at set point (12 weeks postinfection) and survival could be found. Five out of eight SIVlhoest-infected and three out of four SIVsun-infected macaques succumbed to AIDS during the first 5 years of infection. Thus, the survival of SIVsun- and SIVlhoest-infected animals was significantly longer than that of SIVagm- or SIVsm-infected macaques. All PT macaques maintained strong SIV antibody responses despite progression to SIV-induced AIDS. The development of immunodeficiency in the face of low viremia suggests that SIVlhoest and SIVsun infections of macaques may model unique aspects of the pathogenesis of human immunodeficiency virus infection in humans.

Simian immunodeficiency viruses (SIV) are the origin of the human immunodeficiency viruses (HIV) (2, 26) and are found naturally in feral African primates. They are a polymorphic family of viruses that share approximately 50% amino acid identity in gag or pol genes between lineages (50, 58). As recently described (15), seven SIV lineages have been characterized to date: (i) the SIVcpz lineage from chimpanzees (Pan troglodytes) (26, 51, 52), (ii) the SIVsm lineage from sooty mangabeys (Cercocebus atys) (12, 13), (iii) the SIVagm lineage from African green monkeys (Chlorocebus aethiops superspecies) (1, 31), (iv) the SIVsyk lineage from Sykes' monkeys (Cercopithecus mitis) (22, 34), (v) the SIVlhoest lineage from L'Hoest monkeys (Cercopithecus lhoesti) (32, 56), (vi) the SIVcol lineage from a guereza colobus (Colobus guereza) (16), and (vi) the SIVgsn lineage from greater spot-nosed monkeys (Cercopithecus nictitans) (17). Some of these groups can be further subdivided; thus, the SIVcpz lineage also includes HIV type 1 (HIV-1) and is the origin of the HIV-1 epidemic (30, 57). The SIVsm group also includes SIVmac, isolated from captive macaques (19), and HIV-2 (14). Similarly, the SIVagm group consists of at least four related subtypes isolated from the tantalus (SIVtan), vervet (SIVver), grivet (SIVgri), and sabaeus (SIVsab) species of African green monkeys (1, 24, 33, 36, 39, 62). The SIVlhoest group comprises SIVsun, SIVlhoest, and SIVmnd-1 (7, 32, 56, 63, 64, 66). SIVcol is represented by a single isolate (16), and the SIVgsn lineage consists of SIVgsn from greater-spot-nosed monkeys (Cercopithecus nictitans), SIVmon from mona monkeys (Ceropithecus mona), and SIVmus from mustached monkeys (Cercopithecus cephus) (4, 15, 17). In addition to viruses that can be classified into lineages, there are also unclassified recombinant viruses, such as SIVrcm from red-capped mangabeys (Cercocebus torquatus) (3, 8, 27), SIVdrl from drill monkeys (Mandrillus leucophaeus) (38), SIVmnd-2 from mandrills (Mandrillus sphinx), and SIVdeb from DeBrazza's monkeys (Cercopithecus neglectus) (9).

Natural SIV infection is highly prevalent (10 to 60% seroprevalence) in feral African monkey populations (1, 31). Despite this high seroprevalence, there is little evidence that natural infection results in clinical disease, specifically immunodeficiency, in these animals (10, 45, 47), although an AIDS-like disease has been observed infrequently in naturally infected monkeys after long periods of infection (41). This apparent lack of disease association is characteristic of each of the African species naturally infected in the wild. However, passage of SIVsm to Asian species, such as macaques, resulted in an AIDS-like syndrome similar in many respects to human AIDS (18, 61). SIVsm is pathogenic for cynomolgus, rhesus, stump-tailed, and pig-tailed (PT) macaques (18, 35, 43, 46). Therefore, SIVsm infection of macaques has become a useful animal model for the study of AIDS pathogenesis and vaccine development (20). Macaques infected with SIVsm experience a period of acute plasma- and cell-associated viremia, followed by a progressive decline in the number of circulating CD4⁺ lymphocytes which precedes the onset of clinical symptoms (42). Early signs of disease include progressive weight loss, failure to thrive, chronic diarrhea, and lymphadenopathy (11). Infected macaques eventually develop opportunistic infections, such as Pneumocystis carinii pneumonia, Mycobacterium avium lymphadenitis and enteritis, candidiasis, adenovirus pancreatitis, and disseminated cytomegalovirus (CMV) infections, within months or years of infection (5, 42, 61). A significant number of animals also develop symptoms of central nervous system pathology (68) similar to lesions observed in AIDS dementia patients. The majority of SIV pathogenesis and vaccine studies have utilized SIVsm or SIVmac infection of macaques. In addition, SIVagm infection has been extensively studied in macaques (31, 33). SIVagm is pathogenic in PT macaques but not in rhesus and cynomolgus macaques; SIVagm-infected PT macaques show a pathogenesis pattern similar to that of those infected with SIVsm. In contrast, macaques (rhesus, cynomolgus, and PT) inoculated with SIVsyk became persistently infected but remained clinically healthy (34). Isolates from the other two SIV lineages (SIVcol and SIVgsn) have not yet been characterized for their pathogenesis in heterologous hosts.

In the present study, we sought to infect macaques with viruses of our recently discovered novel L'Hoest lineage, SIVlhoest and SIVsun. Since PT macaques seem to be more susceptible to infection with certain SIVs than rhesus and cynomolgus macaques (25, 33, 55), we infected PT macaques with SIVlhoest (n = 8) and SIVsun (n = 4) and monitored their viral loads (RNA and live virus), antibody responses, lymphocyte subsets, and clinical and pathological features over a period of 5 (SIVsun) to 7 (SIVlhoest) years.

MATERIALS AND METHODS

Animals and virus inoculations.

All animals used in this study were PT macaques (Macaca nemestrina) between the ages of 2 and 6 years housed in accordance with American Association for Accreditation of Laboratory Animal Care standards. When necessary, animals were immobilized with 10 mg of ketamine HCl (Parke-Davis, Morris Plains, N.J.) per kg of body weight, injected intramuscularly. The investigators adhered to the Guide for the Care and Use of Laboratory Animals prepared by the Committee on Care and Use of Laboratory Animals of the Institute of Laboratory Resources, National Resource Council, and experimental protocols were approved by the NIAID Animal Care and Use Committee. Prior to use, animals were negative for antibodies to HIV-2, SIV, type D retrovirus, and simian T-cell leukemia virus type 1.

SIVlhoest was isolated from peripheral blood mononuclear cells (PBMC) of a L'Hoest monkey by coculture of phytohemagglutinin-stimulated PBMC with Molt4clone8 cells (32). Virus stocks were prepared from these infected cells by filtration through a 0.45-μm-pore-size filter and cryopreserved in the vapor phase of liquid nitrogen for use in subsequent infectivity studies. These culture supernatants were used to infect Molt4clone8 and CEMss cells for preparation of cell-free virus stocks for infectivity studies with macaques (32). Four PT macaques (Macaca nemestrina) were inoculated intravenously with 1 ml of uncloned SIVlhoest virus stock, designated SIVlhoest-P (macaques 622, 623,626, and 627). Since this virus had been passaged twice through human T-cell lines, which might produce attenuation, virus was also isolated from a homogenate of cryopreserved spleen from the same L'Hoest monkey by short-term coculture with Molt4clone8 cells (4 days) and subsequent infection of macaque PBMC. This virus was designated SIVlhoest-S to distinguish it from the PBMC isolate; four additional PT macaques were inoculated intravenously with 1 ml of this primary virus isolate (macaques 633, 634, 635, and 636). SIVsun was isolated by short-term coculture of infected sun-tailed monkey (Cercopithecus solatus) PBMC with Molt4clone 8 cells, and a virus stock was grown in PT macaque PBMC and titered in CEMss cells. Four PT macaques were infected with 1 ml (6.6 × 10⁵ 50% tissue culture infective doses [TCID₅₀]) uncloned SIVsun stock grown in PT macaque PBMC. SIVlhoest-infected PT macaques were monitored for 7 years after infection, and SIVsun-infected PT macaques were monitored for 5 years after infection.

Virus isolations.

PBMC were stimulated with 5 μg phytohemagglutinin per ml of medium and interleukin-2 (10% in RPMI) for 3 days. Subsequently, 5 × 10⁶ PBMC were cocultivated with CEMss cells for 6 weeks, and infection was monitored by a [³²P]TTP-based reverse transcription (RT) assay (69). The cell-associated and cell-free viral loads were determined by limiting-dilution coculture. For the SIVlhoest-infected animals, log₁₀ dilutions of lymph node mononuclear cells (LNMC) (10⁶ to 10¹) were cocultivated with 10⁶ CEMss cells in quadruplicate and monitored weekly for RT. For the SIVsun-infected macaques, fivefold dilutions (10⁶ to 12.8) of PBMC or LNMC were cocultured with 10⁶ CEMss cells in triplicate and monitored weekly for RT. Cultures were maintained for 3 weeks and screened for RT activity. The infectious titer was calculated using the Spearman-Kaerber method (23).

Western blot analysis.

In order to detect SIVsun- and SIVlhoest-specific antibodies, SIVlhoest antigen was prepared by generating a large culture of SIVlhoest in CEMss cells. Virus was filtered through a 0.45-μm-pore-size filter and pelleted by ultracentrifugation. Antigen was separated on a 10% sodium dodecyl sulfate-acrylamide gel and blotted passively for 3 days onto a nitrocellulose filter. The nitrocellulose membrane was blocked for 1 h at 37°C with 3% bovine serum albumin (BSA) in phosphate-buffered saline (PBS), washed with 0.2% Tween 20 in PBS, and cut into strips. Strips were incubated with a 1:100 dilution of SIV-specific serum in Western blot (WB) diluent solution (1% BSA-0.4% Tween 20 in PBS) containing 1% nonfat milk powder for 1 h at room temperature. Subsequently, strips were washed with PBS-0.2% Tween 20 and incubated with a 1:200 dilution of the secondary antibody (biotin-conjugated anti-human immunoglobulin G; Amersham, Piscataway, NJ) in WB diluent and 1% nonfat milk powder for 1 h at room temperature. After a wash with 0.2% Tween 20 in PBS, 1:5,000 streptavidin alkaline phosphate (Invitrogen, Carlsbad, CA) in WB diluent and 1% nonfat milk powder were added for 1 h at room temperature. The strips were rinsed with distilled water and developed in a premixed solution of nitroblue tetrazolium (NBT) concentrate-5-bromo-4-chloro-3-indolylphosphate (BCIP) concentrate-0.1 M Tris, pH 9.5 (1:1:10) (BCIP/NBT phosphate substrate system; Kirkegaard & Perry Laboratories, Inc.).

Viral RNA load in plasma.

The viral RNA load in plasma was measured by real-time RT-PCR. For this purpose, a SIVlhoest and SIVsun RNA transcript standard was created. One-kilobase fragments of the envelope gene were PCR amplified from pSIVlhoest7 (AF075269) and pSIVsun8/20 (AF131870) using primer pairs SIVlhoest-envF (7472-7492) (5′ ATG CCA TGG GAA TGC AGA AGC CAT GAA TTA 3′)-SIVlhoest-envR (8451-8471) (5′ ATC GGA TCC CCA AAA GTT AAA TTC TTG TAA 3′) and SIVsun-envF (7522-7542) (5′ ATG CCA TGG GAG GCA GGT GGA ACC TTT AGC 3′)-SIVsun-envR (8501-8521) (5′ ATC GGA TCC TAG CTG TTT GAA CTT GTT TTG 3′). The PCR products were cloned into pTRI-19 in which a 30-bp poly(A) stretch was inserted using restriction enzymes NcoI and BamHI (a generous gift from Jeffrey Lifson). Plasmid DNA was linearized with EcoRI, in vitro transcribed, and purified with oligo(dT) beads. The DNA templates were removed by DNase I and phenol-chloroform extraction, and the RNA transcripts were purified with Sephadex spin columns and quantified by measurements of A₂₆₀ based on the calculated extinction coefficient for the transcript sequence. A serial fivefold dilution series of the standard RNA templates was assayed in duplicate to generate the standard curve for each assay.

Real-time PCR was performed using primers F-lhoe/sun (5′ AAT GGG CTT AGT ATC GAC GAT 3′) and R-lhoe/sun (5′ CGT GCT TGG AGA TTC TTC AC 3′) and probe P-lhoe/sun (5′ 6-carboxyfluorescein-ATA TTG CAG CAG CAG AAG CAG TTG-6-carboxytetramethylrhodamine 3′). The primers span a highly conserved 150-bp sequence in the transmembrane protein region of the envelope. Plasma samples were stored at −70°C prior to processing. From the plasma of the SIV-infected animals, virions were pelleted by ultracentrifugation and RNA was extracted using the QIAamp Viral RNA Mini kit (QIAGEN, Valencia, CA). RT reactions were performed using TAQMAN reverse transcription reagents (Applied Biosystems, Foster City, CA) according to the manufacturer's instructions. The TAQMAN PCR Core Reagent kit (Applied Biosystems, Foster City, CA) was used for the DNA PCR. The sequence detection reactions were as follows: 50°C for 2 min, 95°C for 10 min, and 45 cycles of 95°C for 15 s and 60°C for 60 s. Assays were performed in triplicate, and results were normalized to the volume of plasma extracted and expressed as SIV RNA copy equivalents per ml of plasma as previously described (29), with a lower limit of detection of 100 copy equivalents per ml. A virus culture supernatant was assayed with each run to ensure reproducibility between assays. The interassay variation coefficient was less than 25%.

Lymphocyte immunophenotyping.

Lymphocyte subsets (CD2, CD3, CD20, CD4, CD8) were analyzed sequentially throughout the infection by fluorescence-activated cell sorter analysis using a Coulter Epics 753 (assays were performed by FASTSYSTEMS, Inc., Gaithersburg, Md.). Heparinized whole-blood samples were incubated for 20 min in the dark at 4°C in the presence of sodium azide with the appropriate monoclonal antibody conjugate. Following staining, erythrocytes were lysed, and the leukocytes were fixed in 1% paraformaldehyde and analyzed with a flow cytometer. The monoclonal antibodies used were CD4 clone M-T477 (conjugated with phycoerythrin; BD Pharmingen) to identify CD4⁺ lymphocytes, Leu2A clone SK1 (conjugated with peridinin chlorophyll protein; BD Pharmingen) to identify CD8⁺ lymphocytes, T11 clone SFCI3PT2H9 (conjugated with RD1; Beckman Coulter) to identify CD2⁺ T lymphocytes, CD3e clone SP34 (conjugated with fluorescein isothiocyanate; BD Pharmingen) to identify CD3⁺ T cells, and Leu16 clone L27 (conjugated with fluorescein isothiocyanate; BD Pharmingen) to identify CD20 expressed on B lymphocytes.

Immunohistochemistry and in situ hybridization.

Formalin-fixed, paraffin-embedded tissues were stained for SIV RNA as previously described (18) using SIVlhoest-specific probes. Briefly, the tissue sections were deparaffinized, rehydrated with water, pretreated with 0.2 N hydrochloric acid and proteinase K, prehybridized, and hybridized overnight at 51°C with either the antisense or the sense riboprobe. The riboprobe consisted of a mixture of probes encompassing 90% of the SIV genome conjugated with digoxigenin-UTP (Lofstrand Labs Ltd., Gaithersburg, Md.) at a final concentration of 1.75 ng/μl. The hybridized sections were washed in standard posthybridization buffers and RNase solutions (RNase A from Sigma, St. Louis, MO, and RNase T₁ from Roche Molecular Biochemicals, Indianapolis, IN). The sections were blocked in 3% normal sheep and horse serum in 0.1 M Tris (pH 7.4) and then incubated with a 1:500 dilution of alkaline phosphatase-conjugated sheep anti-digoxigenin (Roche Molecular Biochemicals) for 1 h. Sections were then rinsed in Tris buffer, reacted with NBT/BCIP (Vector Laboratories, Ltd., Burlingame, Calif.) for 10 h, and visualized with a Zeiss Axiophot microscope (Carl Zeiss Inc., Thornwood, NY).

Formalin-fixed, paraffin-embedded tissue sections were stained with an anti-human CD4 mouse monoclonal antibody, clone 1F6 (Novacastra Laboratories, Newcastle, United Kingdom). Sections were rehydrated and processed for 6 to 8 min in a Presto pressure cooker (National Presto Industries, Eau Claire, WI) in 1 mM EDTA (pH 8.0) or Tris to unmask antigens. The samples were sequentially treated with PBS, aqueous hydrogen peroxide, serum block (3% normal goat serum-1% nonfat milk-0.5% BSA), and the specific monoclonal antibody for 1 h. The reaction was visualized by using the Vectastain mouse-immunoglobulin G peroxidase ABC kit (Vector Laboratories, Burlingame, Calif.) and diaminobenzidine followed by 10 s of treatment in DAB enhancing solution (Vector Laboratories). Samples were then rinsed in distilled water and counterstained with hematoxylin.

Statistical analysis.

The set point for viral RNA was defined operationally as 12 weeks postinoculation, the first time point at which viral RNA levels had achieved a semi-steady-state plateau. Arithmetic means of log₁₀ viral RNA copies at set point and peak log₁₀ RNA copies were compared using Student's t test. Log₁₀ viral RNA copies at set point, the percentage of decline in CD4⁺ lymphocyte counts at 6 months and 1 year of infection, the CD4⁺ lymphocyte slope in the first year, and survival rates were compared using Pearson's correlation coefficient. The significance of r was calculated by the two-tailed test. The slope of the CD4⁺ lymphocyte decline was calculated using the SLOPE function in Microsoft Excel. The Kaplan-Meier plot was generated using GraphPad Prism.

RESULTS

SIVlhoest- and SIVsun-infected macaques develop opportunistic infections.

In the present study, eight macaques were infected with uncloned SIVlhoest and four with uncloned SIVsun, and the animals were monitored for 5 (SIVsun) to 7 (SIVlhoest) years. The median survival time was 3.8 years for the SIVlhoest-S-infected animals and >5.5 years for the SIVlhoest-P-infected macaques. Exact median survival could not be determined for the latter animals, since two out of four animals were still alive after the 7-year observation period. The median survival time for the SIVsun-infected animals was 4.3 years (Table 1). The animals that progressed to disease during the observation period showed classic opportunistic infections, e.g., Mycobacterium avium enteritis and lymphadenitis (PT622, PT633, PT634, PT636), vegetative bacterial endocarditis (PT622, PT528), Cryptosporidium infection (PT636), and CMV infection in the lung, spleen, adrenals, and pancreas (PT636). Other significant findings included hepatic and splenic amyloidosis (PT626), intestinal leiomyosarcoma (PT527), multifocal encephalitis (probably due to simian virus 40), and pyogranulomatous pneumonia.

TABLE 1.

Pathology of SIVlhoest- and SIVsun-infected macaques at time of death

Virus	Macaque no.	Time of death (postinfection)	Pathological finding(s)a
SIVIhoest (PBMC)	622	148 wk	M. avium enteritis, lymphadenitis, vegetative endocarditis
	623	Alive (>7 yr)	N/A
	626	216 wk	Hepatic and splenic amyloidosis, enteritis
	627	Alive (>7 yr)	N/A
SIVIhoest (spleen)	633	259 wk	Cryptosporidia, M. avium enteritis
	634	98 wk	M. avium enteritis
	635	Alive (>7 yr)	N/A
	636	137 wk	CMV (lung, spleen, adrenals, pancreas), M. avium enteritis
SIVsun (PBMC)	527	181 wk	Leiomyosarcoma, intussusception
	528	231 wk	Vegatative endocarditis, severe diffuse granulomatous parasitic pneumonia (lung mites), multifocal encephalitis (probably due to SV40)
	530	212 wk	Pyogranulomatous pneumonia, microabscesses with bacterial colonies
	605	Alive (>5 yr)	N/A

^aN/A, not applicable; SV40, simian virus 40.

Superficial examination of the disease course suggested a prolonged survival time relative to the median survival time (1 to 2 years) reported for SIVsm/mac infection of macaques (28, 29, 40). For detailed comparison of the disease outcomes of SIVlhoest- and SIVsun-infected macaques, the survival curves of the SIVlhoest- and SIVsun-infected macaques were graphed in a Kaplan-Meier plot and compared to those of cohorts of SIVagm- and SIVsm-infected macaques. As shown in Fig. 1, the survival curves of the SIVlhoest- and SIVsun-infected macaques were similar to one another. Their survival times were significantly longer than those of macaques infected with SIVsmE543-3 (n = 16; median survival, 2 years). The latter group was composed of rhesus macaques inoculated with the same dose and stock of SIVsmE543-3 in three separate animal studies, two of which have been published (28, 40). A cohort of PT macaques infected with SIVagm90 and SIVagm9063 (n = 8) also progressed to disease faster than SIVlhoest- or SIVsun-infected animals (median survival time, 1.1 years) (29). SIVagm, SIVlhoest, and SIVsun are each primary isolates that have not been passaged in macaques, whereas SIVsmE543 is the result of 2 macaque passages. Therefore, the accelerated disease outcome for macaques inoculated with SIVsmE543 relative to those inoculated with SIVlhoest or SIVsun could be due to prior adaptation and selection in macaques.

FIG. 1.

Kaplan-Meier survival plot of macaques infected with SIVlhoest (n = 8), SIVsun (n = 4), SIVsmE543-3 (n = 16), or SIVagm90 (n = 8). The plot was generated with GraphPad Prism 3.0. Survival of SIVlhoest- and SIVsun-infected animals is prolonged compared to that of SIVagm- and SIVsm-infected macaques.

High primary viremia in SIVlhoest- and SIVsun-infected PT macaques.

To monitor the viral loads in the infected animals, we performed virus isolations, limiting-dilution coculture, and quantitative real-time RT-PCR. Virus was consistently isolated from all of the SIVlhoest- and SIVsun-infected animals in the first 4 weeks of infection. Virus isolation became intermittent in the subsequent weeks following inoculation (Table 2). Virus could only occasionally be isolated from SIVlhoest-inoculated macaques by week 120, and attempts to reisolate virus were terminated at week 252. The SIVsun-infected animals were intermittently virus isolation positive between weeks 9 and 32 and then became negative (Table 3). Virus isolation attempts were terminated at week 180. This pattern of virus isolation is consistent with low numbers of infected PBMC in peripheral blood.

TABLE 2.

Virus isolation from peripheral blood mononuclear cells of SIVIhoest-inoculated pig-tailed macaques

Wk	Statusa of:
	PT622	PT623	PT626	PT627	PT633	PT634	PT635b	PT636
1	+	+	+	+	+	+	+	+
2	+	+	+	+	+	+	+	+
4	+	+	+	+	+	+	+	+
8	+	−	+	+	+	+	+	+
12	+	−	−	−	+	+	+	+
18	+	+	+	+	+	+	+	+
20	+	−	−	+	−	+	+	−
24	−	−	−	+	+	+	+	−
28	−	−	−	+	+	+	+	−
36	+	+	−	−	+	−	+	−
44	−	+	+	+	+	+	+	+
48	−	−	+	+	−	−	+	−
52	−	−	−	+	−	−	+	−
60	−	−	+	+	−	−	−	−
68	−	−	+	+	+	−	−	−
76	−	−	−	−	−	−	−	−
90	−	−	−	+	−	+	−	−
104	−	−	+	−	−	† (98)	−	−
120	−	−	−	−	−		−	−
136	−	−	−	−	−		−	† (137)
144	−	−	−	−	−		−
152	† (148)	−	−	−	−		−
160		−	−	+	−		+
168		−	−	−	−		−
			† (216)		† (259)

^a+, virus was isolated; −, no virus was isolated; †, death. Numbers in parentheses indicate the week of death.

^bVirus was isolated from PT635 (week 228). Virus isolation was terminated in week 252.

TABLE 3.

Virus isolation from peripheral blood mononuclear cells and lymph node mononuclear cells of SIVsun-inoculated pig-tailed macaques

Wk	Statusa of PBMC (LNMC) in:
	PT527	PT528	PT530	PT605
0	−	−	−	−
1	+ (+)	+ (+)	+ (+)	+ (+)
2	+ (+)	+ (+)	+ (+)	+ (+)
3	+	+	+	+
4	+ (+)	+ (+)	+ (+)	+ (+)
6	+	+	+	+
9	−	+	+	−
12	−	+	+	−
16	+ (+)	+ (+)	+ (+)	− (+)
20	−	+	+	−
24	+	+	+	−
28	−	−	−	−
32	−	−	+	−
36b	−	−	−	−

^a+, virus was isolated; −, no virus was isolated.

^bAll animals were virus isolation negative after week 36; virus isolation was terminated in week 180.

SIV-specific in situ hybridization of lymph node biopsy samples collected from SIVlhoest-infected macaques during the first 4 weeks of infection demonstrated high levels of expression at 1 week with declining numbers by 2 and 4 weeks, trapping of virus in germinal centers by 4 weeks postinfection, and rapid development of lymphadenopathy (36). Similar kinetics of expression were observed in SIVsun-infected PT macaques (data not shown). Cell-associated viral loads of PBMC and LNMC were assayed in SIVsun-infected animals by limiting-dilution coculture. As shown in Table 4, the cell-associated viral load peaked in week 1 or 2 in the peripheral blood or lymph nodes, in agreement with the RNA copy numbers. Cell-associated viral loads were generally higher in the lymph nodes than in the peripheral blood. Similar results were obtained in limiting-dilution cocultures of LNMC of SIVlhoest-infected macaques at 1 and 2 weeks postinoculation (mean, 742 and 44 infected cells/10⁶ cells).

TABLE 4.

Cell-associated viral load in PBMC and LNMC in pig-tailed macaques infected with SIVsun

Wk	No. of infected PBMC (LNMC) in:
	PT527	PT528	PT530	PT605
0	0	0	0	0
1	164 (481)	164 (164)	33 (56)	281 (281)
2	56 (96)	481 (822)	19 (96)	2 (7)
3	1	96	2	11
4	1 (11)	164 (56)	11 (33)	1 (11)
6	1	164	56	1
9	<1	19	11	<1
12	<1	1	1	<1
16	1 (ND)a	1 (11)	1 (4)	<1 (ND)
20	<1	<1	1	<1
24	<1	1	<1	<1

^aND, not done.

Real-time RT-PCR was used to more accurately quantify the extent of viremia in these animals. As shown in Fig. 2, peak viral loads were reached by week 1 in SIVlhoest-infected animals (average, 2.2 × 10⁸/ml). This corresponded to an infectious titer of approximately 1,000 TCID₅₀. Plasma viral RNA loads peaked slightly later, between weeks 1 and 2, in SIVsun-infected animals (average, 5.8 × 10⁶/ml). The log₁₀ peak viral load was statistically significantly higher in animals inoculated with SIVlhoest than in those inoculated with SIVsun (P = 0.0003 [Table 5]). However, there was no statistically significant difference in peak viremia between animals that received SIVlhoest derived from PBMC and those that received SIVlhoest derived from spleen (P = 0.4984 [Table 5]). In the SIVsun-inoculated PT macaques, the cell-associated viral load in PBMC correlated strongly with the viral RNA load in plasma in the first 24 weeks of infection (r = 0.753; P = < 0.0001).

FIG. 2.

Viral RNA loads of PT macaques infected with SIVsun (A), PBMC-derived SIVlhoest (B), or spleen-derived SIVlhoest (C). Viral RNA loads were determined by quantitative TAQMAN RT-PCR. Viral loads peaked at 1 to 2 weeks after infection and then remained below or around 10⁴ RNA copies per ml for the remainder of the observation period. Gray bars indicate viral loads at set point.

TABLE 5.

Comparison of viral loads of SIVsun- and SIVlhoest-infected macaques

Virus	Log10 viral RNA copies/ml (mean ± SD)
	At set pointa	Peakb
SIVsunc	3.25 ± 0.32	6.75 ± 0.11
SIVlhoest-P	2.72 ± 0.50	7.99 ± 0.61
SIVlhoest-S	3.95 ± 0.51	8.25 ± 0.36

^aP = 0.8587 for comparison of SIVsun versus SIVlhoest-P and SIVlhoest-S; P = 0.1205 for comparison of SIVsun versus SIVlhoest-P; P = 0.0137 for comparison of SIVlhoest-P versus SIVlhoest-S.

^bP = 0.0003 for comparison of SIVsun versus SIVlhoest-P and SIVlhoest-S; P = 0.0072 for comparison of SIVsun versus SIVlhoest-P; P = 0.4984 for comparison of SIVlhoest-P versus SIVlhoest-S.

^cFrom PBMC.

Low to moderate viral set points in SIVlhoest- and SIVsun-infected macaques.

Previous studies of SIV-infected macaques and HIV-infected humans have shown that the plateau level of plasma viremia (set point) following primary infection is predictive of disease outcome. Thus, higher levels of plasma viremia are associated with more-rapid disease progression. We therefore chose a period of 12 weeks postinfection as the point where the viral load had reached a fairly stable plateau level in the majority of the animals. Set point viral RNA levels were similar in macaques inoculated with SIVlhoest (7.1 × 10³ copies/ml) and SIVsun (2.3 × 10³ copies/ml). In comparison, 7.6 × 10⁴ viral RNA copies per ml of plasma were measured in plasma collected from the sun-tailed monkey that was the source of SIVsun; this is at the top of the range of chronic viremia in infected PT macaques.

There was no statistically significant difference in viral set point between the SIVlhoest- and SIVsun-infected animals. However, set point viral RNA levels were significantly higher in animals inoculated with SIVlhoest-S than in those inoculated with SIVlhoest-P (P = 0.0137 [Table 5]). Subsequently, the viral loads remained stable between 10² and 10³ copies per ml during the course of infection. A 10- to 100-fold increase in viremia shortly before death was observed for all of the animals except PT530 (SIVsun infected), in which the viral load remained steady until death. To confirm the low level of viral replication in these animals, SIVlhoest-specific in situ hybridization was performed on tissues of four of the PT macaques inoculated with SIVlhoest. Virus-expressing cells were observed at a low frequency in the majority of tissues, with the greatest frequency observed in PT622 (Fig. 3).

FIG. 3.

SIV-specific hybridization showed a low number of SIV-expressing cells in tissues of a SIVlhoest-infected macaque (PT622) collected terminally. (A) Four SIV-positive cells in the mesenteric lymph node. Magnification, ×10. (B) A SIV-expressing intraepithelial lymphocyte in the ileum. Magnification, ×40. (C) A SIV-expressing cell in the lung. Magnification, ×10.

Decline of CD4⁺ T cells in the absence of high viral loads.

Despite strong control of viremia in SIVlhoest- and SIVsun-infected macaques, a progressive loss of CD4⁺ T cells was observed in all of the animals. The steepest decline in CD4⁺ cells occurred during the first year of infection, with a loss of CD4⁺ cells between 43 and 94% (Fig. 4). The majority of animals exhibited CD4⁺ T-cell levels that met the criteria for clinical AIDS (200 per μl) for several years (PT622, PT623, PT633, PT635, PT528, PT530, PT605) prior to development of AIDS. While all of these animals exhibited lymphadenopathy and intermittent diarrhea, they did not meet our clinical criteria for euthanasia (>20% loss in body weight, diarrhea unresponsive to supportive treatment). Thus, the diagnosis of symptomatic AIDS was dissociated from the degree of CD4⁺ cell depletion. Interestingly, CD4⁺ lymphocyte counts in a number of animals, for example, PT605, continued to decrease despite relatively low plasma viral RNA loads (between 10² and 10³ copies per ml). Peripheral CD4⁺ T-cell numbers were less than 100/μl in all but one animal (PT626) at the time of euthanasia.

FIG. 4.

Absolute numbers of CD4⁺ lymphocytes in PT macaques infected with SIVsun (A), PBMC-derived SIVlhoest (B), or spleen-derived SIVlhoest (C). CD4⁺ lymphocyte numbers decreased continuously during the course of infection, in particular during the first 2 years.

As shown for a representative lymph node in Fig. 5A, infection with SIVlhoest or SIVsun caused severe paracortical and follicular depletion by the time of euthanasia. Severe disruption of lymphoid architecture with scarring and involution of germinal centers was also evident (Fig. 5A). In order to confirm the extent of terminal cellular depletion, we used immunohistochemistry to detect CD4⁺ T cells in lymphoid tissues. The spleen and peripheral lymph nodes collected from four SIVlhoest-infected PT macaques at the time of euthanasia (PT622, PT633, PT634, PT636) were stained for CD4 protein expression. As expected from the CD4⁺ lymphocyte depletion in the blood of these animals, lymph nodes were also severely depleted of CD4⁺ lymphocytes compared to the lymph node of an uninfected healthy macaque (Fig. 5B and C). Since similar degrees of peripheral depletion were observed in the SIVsun-infected animals, representative tissues of these animals were not examined.

FIG.5.

SIVlhoest infection leads to severe lymphoid depletion. (A) H&E-stained section of the mesenteric lymph node at the time of death of PT 626, demonstrating severe paracortical and follicular depletion. (B) Marked depletion of CD4⁺ T lymphocytes in the lymph node of PT 626 demonstrated by immunohistochemical staining. (C) Normal distribution of CD4⁺ T lymphocytes in the lymph node of an uninfected macaque.

Loss of peripheral CD4⁺ T cells by 1 year correlates with survival.

Clinical outcomes for the infected macaques were assessed by survival time, viral loads, and surrogate markers of disease progression, such as the percentage of CD4⁺ T-cell decline by 6 months or 1 year and the slope of the decline in counts of CD4⁺ T cells during the first year of infection (Table 6). The correlation between the plasma viral load at set point and survival time or CD4⁺ cell loss was evaluated using Pearson's correlation coefficient (r). The correlation between survival and the percentage of CD4⁺ lymphocytes retained at 6 months was not significant (r = 0.41; P > 0.05). However, survival correlated significantly with the slope of CD4⁺ lymphocyte decline during the first year (r = 0.76; P < 0.005), as well as with the percentage of CD4⁺ lymphocytes remaining after 1 year of infection (r = 0.83; P < 0.005). Thus, measures of CD4⁺ cell loss after 1 year of infection appeared to be a good surrogate marker for the rate of disease progression.

TABLE 6.

Pearson's correlation coefficient for comparisons between viral RNA load at set point, survival, and CD4⁺ lymphocyte counts for SIVsun- and SIVlhoest-infected macaques combined^a

Parameter	Pearson's r for comparison of the indicated parameter versus:
	Log10 viral RNA copies at set point	Survival
% CD4+ lymphocyte loss
At 6 mo	−0.26b	0.41b
At 1 yr	−0.51b	0.82c
Slope of CD4+ cell loss at 1 yr	−0.53b	0.76c
Log10 viral RNA copies at set point		−0.42b

^aSignificant values (P < 0.05) are boldfaced.

^bP > 0.05.

^cP < 0.005.

As mentioned earlier, plasma viral load at set point was defined operationally as the viral RNA load at 12 weeks postinoculation. Considering all of the animals, the correlation between the viral RNA load at set point and survival was not significant (r = −0.42; P > 0.05). In addition, set point viral load did not correlate with CD4⁺ T-cell loss at 6 months or 1 year (Table 6). Neither did the set point correlate with survival or CD4 loss within individual groups (data not shown). These statistical analyses suggest that the viral RNA load at set point was not significantly correlated with either survival or loss of CD4⁺ T lymphocytes. Since the designation of the set point is fairly arbitrary and is based on the stabilization of viremia, we also used the mean post-acute-phase viral load (8 to 24 weeks) as a measure of plateau levels of viremia. Correlations between the average or sum of viral loads from 8 and 24 weeks postinfection, or of peak viral loads, and survival or loss of CD4⁺ lymphocytes were also not significant (data not shown).

SIVlhoest and SIVsun infections result in declines in CD8⁺ and B lymphocytes.

As shown in Fig. 6 and 7, infections with SIVlhoest or SIVsun also resulted in declines in numbers of CD8⁺ T cells and B lymphocytes. CD8⁺ T-cell levels were variable at the time of death, but B-lymphocyte numbers had decreased in all animals that became ill during the observation period; 6 of 8 animals had B lymphocyte levels below 100 per μl, and 2 out of 8 had levels below 200 per μl, prior to death. There was a significant correlation between the percentage of B lymphocytes retained after 1 year of infection and survival (r = 0.59; P = 0.02), but not between the percentage of CD8⁺ lymphocytes retained after 1 year and survival. Presumably the decline in these lymphocyte populations is an indicator of the overall decline in immune function rather than a direct result of virus killing.

FIG. 6.

Absolute numbers of CD8⁺ lymphocytes in PT macaques infected with SIVsun (A), PBMC-derived SIVlhoest (B), or spleen-derived SIVlhoest (C). CD8⁺ lymphocyte numbers decreased continuously during the course of infection.

FIG. 7.

Absolute numbers of B lymphocytes in PT macaques infected with SIVsun (A), PBMC-derived SIVlhoest (B), or spleen-derived SIVlhoest (C). B-lymphocyte numbers decreased continuously during the course of infection.

Strong humoral immune responses in SIVlhoest- and SIVsun-infected macaques.

To investigate humoral immune responses, we performed SIVlhoest-specific Western blotting on sera of the SIVsun- and SIVlhoest-infected animals. Since SIVlhoest and SIVsun cross-react serologically, SIVlhoest antigen was used for both animals. As shown in Fig. 8, both SIVsun- and SIVlhoest-S-infected animals developed strong antibody responses by 4 weeks postinfection. The response of SIVlhoest-P-infected animals (not shown) was comparable to that of SIVlhoest-S-infected animals. None of the macaques lost reactivity to Gag or Env proteins during the course of infection.

FIG. 8.

Antibodies against SIVlhoest and SIVsun were detected by Western blotting. (A) SIVsun-infected PT macaques; (B) PT macaques infected with spleen-derived SIVlhoest. SIVlhoest virus lysates were separated by sodium dodecyl sulfate-acrylamide gel electrophoresis and blotted onto a nitrocellulose membrane. Sera from weeks 0 to 292 (SIVlhoest-infected macaques) and weeks 0 to 220 (SIVsun-infected macaques) were tested for SIVlhoest-specific antibodies. SIVsun serologically cross-reacts with SIVlhoest.

DISCUSSION

In the present study, we demonstrated the pathogenicity of SIVlhoest and SIVsun in a heterologous host, the PT macaque. Although these viruses did not induce AIDS in their natural hosts, the L'Hoest and sun-tailed monkeys, both viruses resulted in progressive loss of CD4⁺ T cells, severe lymphoid depletion, and the development of an AIDS-like syndrome in macaques characterized by opportunistic infections such as Mycobacterium avium, Cryptosporidium, and CMV infections. Therefore, SIVlhoest and SIVsun infections are comparable to SIVsm and SIVagm infections; they are apparently apathogenic for their natural host species but cause an AIDS-like syndrome in Asian macaques (1, 6, 10, 29, 31, 33, 40, 48, 60). Naturally infected African green monkeys, mandrills, and sooty mangabeys characteristically maintain moderate to high viremia. This also appears to be the case for SIV-infected sun-tailed monkeys, although our assessment is based on a single animal. Unfortunately, we did not have the opportunity to measure viral loads in the naturally infected L'Hoest monkey.

The characteristics of primary SIVlhoest or SIVsun infection of PT macaques were similar in many aspects to those previously reported for SIVsm or SIVagm infection of macaques (28, 29, 40). Thus, we observed high primary viremia, high numbers of virus-expressing cells in lymphoid tissues, the development of lymphadenopathy, and subsequent scarring and involution of germinal centers in lymph nodes as well as a significant decline in circulating CD4⁺ T cells in both SIVsun- and SIVlhoest-infected PT macaques. Many of the other correlates associated with SIVsm/mac-induced AIDS, such as levels of activation and spontaneous apoptosis, were not assessed in this pilot study.

However, survival was prolonged for macaques infected with SIVlhoest or SIVsun compared to that of macaques infected with SIVsm and SIVagm (29). Therefore, SIVlhoest and SIVsun appear to be less pathogenic than SIVsm or SIVmac in macaques. This might be partially explained by significant containment of viremia during chronic infection. Although SIVlhoest and SIVsun actively replicated during the acute phase of infection (32), virus was contained at low levels throughout infection and remained low even at the time of death. This contrasts with the majority of SIVsm/mac-infected macaques, where active virus replication occurs throughout infection (28, 40). Since post-acute-phase viremia was low in the majority of SIVhoest- and SIVsun-infected macaques, viral RNA loads at set point did not correlate with survival or CD4⁺ T-cell loss. In contrast, plateau levels of viremia in SIVsm/mac infection differ significantly and are tightly linked to the rate of disease progression. Thus, animals with high persistent viremia progress rapidly, whereas those animals with better containment of viremia progress significantly more slowly, with a minor fraction of animals showing no evidence of progression for many years. The slow disease course observed in SIVlhoest- and SIVsun-infected macaques is similar to that of nonprogessors in SIVsm/mac or HIV infection and is consistent with a significant role of viremia and the cytopathic effects of SIV on CD4⁺ T cells. However, the animals in the present study all had comparable levels of viremia and yet showed different disease courses, differing in their survival time from 1.5 to >7 years. Consequently, survival did not correlate with the extent of viremia in this animal model. SIVsm/mac-infected animals that contain viremia to low levels generally remain clinically asymptomatic, with progression to disease occurring at a later time point and associated with rising plasma virus levels due to the development of immune escape variants (49). Indeed, ten Haaft et al. have postulated a rather arbitrary virus threshold of 10⁴ copies/ml of plasma, above which pathogenic infection and progression to AIDS are observed (65). The SIVlhoest- and SIVsun-infected PT macaques were unusual in that virus levels remained well below or close to this threshold even in the period just prior to death. The mechanistic basis for the restriction in viremia in these animals is not clear and could include immune control or depletion of target cells, presumably CCR5⁺ CD4⁺ T cells. The early depletion of CD4⁺ T cells is consistent with the latter explanation, but further studies of CD4⁺ T-cell subsets in the peripheral blood and mucosal sites will be required in the future to resolve these issues.

Despite strong containment of viremia, the animals still showed progressive loss of CD4⁺ T cells and development of AIDS-like symptoms. These data suggest that the active role of virus replication in AIDS may not be as incontrovertible as previously suspected. Indeed, many have speculated that the pathogenesis of AIDS is more complex than simply cytopathic destruction of CD4⁺ T-cell targets and may involve other aspects such as immune activation-induced cell death, failure of regenerative capacity of CD4⁺ T cells by the thymus and/or bone marrow, loss of immunologic function associated with failure of antigen presentation (follicular dendritic cells), and dysfunction of helper T cells (21, 37). Many of these parameters were beyond the scope of the present study. The data also suggest that our understanding of the pathogenesis of AIDS is incomplete. Clearly, AIDS is not entirely the result of virus-induced CD4⁺ T-cell death but rather of the perturbation of the balance between thymic output and regeneration of T cells, and activation and virus-induced cell death.

Another intriguing feature of SIVlhoest and SIVsun infection of macaques was that significant peripheral CD4⁺ T-cell depletion occurred long before the development of clinical AIDS (1 to 2 years postinfection). This is clearly different from SIVsm/mac infection of macaques and HIV infection of humans, where the onset of AIDS is associated with the decrease in CD4⁺ T cells below a critical threshold of approximately 250 CD4⁺ T cells/μl. The early CD4⁺ lymphocyte loss in this study indeed is not unlike the dissociation of CD4⁺ T-cell loss from AIDS observed in simian-human immunodeficiency virus-infected macaques that do not succumb rapidly within the first 3 to 4 months after challenge (54, 59). The eventual development of opportunistic infections in SIVlhoest- and SIVsun-infected PT macaques was not associated with another precipitous decline in the total CD4⁺ T-cell population. More-detailed subset analyses of memory/naïve subsets and of CD4⁺ T-cell function, including proliferative ability and cytokine production, are required to determine whether there was perhaps a selective loss of one of these subsets or functions, since such a selective loss of CD4⁺ memory T cells in the blood and mucosal sites is a hallmark of early SIVsm/mac infection of macaques (44, 53, 67).

In summary, we investigated the pathogenicity of SIVlhoest and SIVsun in PT macaques. Both viruses caused a continuous decline in CD4⁺ T-lymphocyte levels and the development of AIDS-like symptoms in the absence of high viral loads during chronic infection. The animals maintained strong humoral immune responses even prior to death. Further studies are warranted to investigate the immunopathology of SIVsun and SIVlhoest in macaques, since this virus infection may model unique aspects of AIDS pathogenesis.