Development of a chronically catheterized maternal-fetal macaque model to study in utero mother-to-fetus HIV transmission: a preliminary report

Four decades of leading-edge research in the reproductive and developmental sciences: the Infant Primate Research Laboratory at the University of Washington National Primate Research Center

The Infant Primate Research Laboratory (IPRL) was established in 1970 at the University of Washington as a visionary project of Dr. Gene (Jim) P. Sackett. Supported by a collaboration between the Washington National Primate Research Center and the Center on Human Development and Disability, the IPRL operates under the principle that learning more about the causes of abnormal development in macaque monkeys will provide important insights into the origins and treatment of childhood neurodevelopmental disabilities. Over the past 40 years, a broad range of research projects have been conducted at the IPRL. Some have described the expression of normative behaviors in nursery-reared macaques while others have focused on important biomedical themes in child health and development. This article details the unique scientific history of the IPRL and the contributions produced by research conducted in the laboratory. Past and present investigations have explored the topics of early rearing effects, low-birth-weight, prematurity, birth injury, epilepsy, prenatal neurotoxicant exposure, viral infection (pediatric HIV), diarrheal disease, vaccine safety, and assisted reproductive technologies. Data from these studies have helped advance our understanding of both risk and resiliency in primate development. New directions of research at the IPRL include the production of transgenic primate models using our embryonic stem cell-based technology to better understand and treat heritable forms of human intellectual disabilities such as fragile X.

Elucidation of the time course of adenosine deaminase APOBEC3G and viral infectivity factor vif in HIV-2(287) -infected infant macaques

BACKGROUND

Although the interactions of cellular cytidine deaminase A3G and viral infection factor (vif) of human immunodeficiency virus (HIV) were reported, regulation of A3G after in vivo HIV infection and disease progression is not known.

METHODS

Time courses of plasma virus, CD4(+) T lymphocyte Macaca levels, and concentrations of A3G and vif transcripts were determined in infant macaques infected with HIV-2(287) . These in vivo results were compared with those collected in vitro in HIV-2-infected T cells.

RESULTS

Human immunodeficiency virus-infected macaques exhibited plasma viremia (≥10(8) copies/ml) followed by a precipitous CD4(+) T-cell (from 40-70 to ≤5%) decline. An initial increase in A3G transcripts coincides with early increases in virus and vif RNA. As virus load continues to increase, A3G RNA decreases but recovers at a later phase as virus level stabilizes. Pearson correlation analysis revealed strong interactions of A3G-CD4, vif-CD4, and A3G-vif.

CONCLUSIONS

There is a time-dependent A3G and vif RNA interaction throughout the course of HIV infection.

Cognitive and motor deficits associated with HIV-2(287) infection in infant pigtailed macaques: a nonhuman primate model of pediatric neuro-AIDS

Lentivirus-infected nonhuman primates exhibit behavioral and neurological pathology similar to human immunodeficiency virus (HIV)-infected humans and offer a means to examine the effects of lentivirus infection while controlling for confounding factors inherent in human populations. The purpose of this study was to examine cognitive and motor development in infant macaques vertically infected with HIV-2287. Subjects were 20 infant pigtail macaques (Macaca nemestrina); 8 controls born to uninfected dams, and 12 infants whose dams had been inoculated and infected with HIV-2287 in the third trimester of pregnancy. Eight of these pregnancies had undergone surgical procedures in the form of maternal amniotic catheters or maternal amniotic and fetal carotid artery and jugular vein catheters. Data indicated that catheterization had little or no impact on behavioral development. Seven infants were vertically infected (as measured by polymerase chain reaction (PCR) at birth) and five were not infected (as measured by PCR and coculture on repeated testing). Infected infants attained cognitive and motor milestones at significantly later ages than controls. Uninfected infants, born to infected dams, attained developmental milestones at later ages than controls on all tasks, but this reached statistical significance only for the Fine Motor Task. Attainment of milestones was not correlated with viral dose, maternal CD4+ levels at parturition or infant viral RNA levels at birth. Attainment of milestones was negatively correlated with infants' proportions of CD4+ lymphocytes at birth and significantly correlated with proportions of CD4+ lymphocytes 2 weeks after birth, indicating poorer performance in those infants with a more rapid CD4+ depletion. These cognitive and motor deficits closely resemble those observed in human infants and children infected with HIV and indicate that HIV-2287-infected infant macaques represent an excellent model of pediatric neuro-acquired immunodeficiency syndrome (neuroAIDS).

HIV in central nervous system and behavioral development: an HIV-2287 macaque model of AIDS

OBJECTIVE

To determine which route of inoculation produced consistent and frequent HIV infection in the central nervous system (CNS) and alterations in cognitive and motor development in infant macaques.

METHODS

Infant macaques (Macaca nemestrina) were inoculated with the highly pathogenic strain HIV-2287 intravenously (n = 3) or intrathecally (n = 3). Uninfected infants were evaluated as controls. Disease progression was evaluated by virological assessment of blood and cerebral spinal fluid (CSF), CD4 T cell count in blood, and quinolinic acid levels in CSF (a surrogate marker of neuronal cell damage). The effect of HIV infection on cognitive and motor development in infants was monitored during the 6-month study.

RESULTS

Either route of HIV-2287 inoculation produced detectable viral RNA in CSF and productive infection in blood. Detection of virus in CSF paralleled a rise in quinolinic acid levels. All HIV-infected infants experienced a severe and rapid decline in CD4 T cell counts by 10 weeks after viral infection. HIV-infected infants, particularly those infected by the intravenous route, exhibited delays in reaching cognitive and motor milestones, which paralleled neuropathological changes.

CONCLUSIONS

The HIV-2287 infant model produced a high incidence of viral infection in the CNS regardless of the route of inoculation. Significant alteration in neurobehavioral development was observed in HIV-infected infants, and this measure was significantly impaired particularly in infants infected by the intravenous route. These data, coupled with the ability to detect viral RNA and changes in quinolinic acid levels in CSF, may allow quantitative evaluation of drug and immune candidates for treating neurological effects of AIDS.

Lipid–Drug Association Enhanced HIV-1 Protease Inhibitor Indinavir Localization in Lymphoid Tissues and Viral Load Reduction: A Proof of Concept Study in HIV-2287-Infected Macaques

Analysis of indinavir levels in HIV-positive patients indicated that drug concentrations in lymph node mononuclear cells (LNMCs) were about 25-35% of mononuclear cells in blood. To enhance lymphatic delivery of anti-HIV drugs, a novel drug delivery strategy was designed consisting of lipid-associated indinavir (50-80 nm in diameter) complexes in suspension for subcutaneous (SC) injection. Due to the pH-dependent lipophilicity of indinavir, practically all the drug molecules are incorporated into lipid phase when formulated at pH 7.4 and 5:1 lipid-to-drug (m/m) ratio. At pH 5.5, about 20% of drugs were found in lipid-drug complexes. Effects of lipid association on the time course of plasma indinavir concentrations were determined in macaques (Macaca nemestrina) administered with either soluble or lipid-associated formulation of indinavir (10 mg/kg, SC). Results yielded about a 10-fold reduction in peak plasma concentration and a 6-fold enhancement in terminal half-life (t1/2beta = 12 vs. 2 hours). In addition, indinavir concentrations in both peripheral and visceral lymph nodes were 250-2270% higher than plasma (compared with <35% with soluble lipid-free drug administration in humans). Administration of lipid-associated indinavir (20 mg/kg daily) to HIV-2287-infected macaques (at 30-33 weeks after infection) resulted in significantly reduced viral RNA load and increased CD4 T cell number concentrations. Collectively, these data indicate that lipid association greatly enhances delivery of the anti-HIV drug indinavir to lymph nodes at levels that cannot be achieved with soluble drug, provides significant virus load reduction, and could potentially reverse CD4 T cell depletion due to HIV infection.

Systemic and intestinal immune responses to HIV-2287 infection in Macaca nemestrina

Nonhuman primate models of human AIDS have been used successfully to evaluate candidate vaccines and infection intervention therapies. Successes of pathogenicity studies in primate models have been limited because of the varied infection outcomes and characteristic low number of study animals. The acutely pathogenic HIV-2(287)--Macaca nemestrina model has shown promise both in antiviral drug evaluation and in pathogenicity studies. Here we describe virus replication, spread, and host responses during the first 28 days of HIV-2(287) infection. Focusing on 18 macaques from a larger 27-macaque study, we report changing virus loads, CD4(+) cell depletions, and antibody responses both systemically and in the mucosa of the small intestine. After intravenous inoculation, blood and intestinal tissue were collected from pairs of macaques at 12 hr and 1, 2, 4, 6, 10, 14, 21, and 28 days postinfection. Specimens were examined for evidence of infection by quantitative cultures, in situ hybridization, lymphocyte subset monitoring, and antibody production. The data were presented serially as though all samples were collected from a single macaque. The highest blood virus loads were detected between days 10 and 14 and subsequently decreased through day 28. This coincided with a significant increase in ileum mucosa virus loads on day 10, which became undetectable by day 28. The lowest levels of CD4(+) cells were observed on days 21 and 28 in blood and ileum mucosa. CD4(+):CD8(+) cell ratios in blood and ileum dropped dramatically after day 10 to lowest levels by day 28. Intestinal virus loads were inversely correlated with CD4(+) cell and virus-specific antibody levels in the ileum after day 6. These results underscore the suitability of this model for pathogenicity studies as well as the importance of the intestinal lymphoid tissues as an initial site of virus replication and cell destruction during the acute, asymptomatic stage of AIDS development.

Suppression of maternal virus load with zidovudine, didanosine, and indinavir combination therapy prevents mother-to-fetus HIV transmission in macaques

Recently, we developed a maternal-fetal macaque model using a highly pathogenic HIV-2 strain, HIV-2287, to study the time course of HIV transmission in utero. Most pregnant macaques (Macaca nemestrina) infected with HIV-2287 (10-103 infective doses) transmitted HIV to their fetuses, as verified by positive identification of virus-infected mononuclear cells and free viral RNA in fetal blood. To determine whether an antiretroviral drug combination therapy composed of two dideoxynucleosides, azidothymidine (15 mg/kg) and dideoxyinosine (15 mg/kg), and a protease inhibitor, indinavir (25 mg/kg), could completely inhibit mother-to-fetus HIV transmission, we administered these drugs orally through gastric catheters to five pregnant macaques infected with 10 infective doses of HIV-2287. Beginning 30 minutes after HIV inoculation, the dams were given the combination antiviral therapy three times daily until delivery by cesarean section. Drug treatment reduced the maternal virus load to a minimally detectable level but did not prevent primary HIV-2287 infection. All fetal and infant blood samples were virus negative by internally controlled RNA polymerase chain reaction (QC-RNA-PCR) and virus coculture assays. Fetal and infant CD4+ T-cell levels remained normal throughout the experiment. These findings strongly suggest that combination chemotherapy with azidothymidine, dideoxyinosine, and indinavir can suppress maternal viral load enough to prevent mother-to-fetus transmission of HIV.

Antigen-specific cytokine responses in vaccinated Macaca nemestrina

We describe a new surrogate assay for CD8 + T lymphocyte activity that has the capability of discriminating between cytotoxic T lymphocyte (CTL) activity and cytokine-mediated suppressive activity. We applied this approach to two groups of Macaca nemestrina vaccinated with a minimally pathogenic strain of human immunodeficiency virus type 2 [HIV-2 (HIV-2(KR))] as a model of an attenuated virus vaccine. Group 1 was then inoculated with a non-infectious stock of a pathogenic strain, HIV-2287. Both groups 1 and 2 were subsequently challenged with an infectious stock of HIV-2287. Five out of six group 1 animals were protected against CD4 decline, whereas three out of six animals in group 2 were protected. Analysis of CTL responses demonstrated strong activity against HIV-2(KR)-Gag in group 1. It was determined that strong CTL responses correlate with antigen-specific T-helper (Th) type 1 responses. This antigen-specific cytokine assay has the potential to better elucidate the functional mechanisms of CD8 + T-cell-mediated protection than traditional methods to date.

A minimally replicative HIV-2 live-virus vaccine protects M. nemestrina from disease after HIV-2(287) challenge

M. nemestrina immunized with an apathogenic HIV-2 molecular clone (HIV-2KR) were protected from CD4 decline and disease upon challenge with HIV-2(287), after any immunizing virus could be detected. Higher but not lower inocula of HIV-2KR were protective against intravenous inoculation of either 10(5) or 10(1) TCID50 of HIV-2(287). Protected animals displayed substantial reductions in PBMC proviral burden (1-3 logs), viral titers (1-2 logs), and plasma viral RNA (2-4 logs) compared to unprotected or naive animals as early as 1 week postinfection. Plasma viral RNA became undetectable after 24 weeks in protected animals, but remained high in unprotected animals. No viral RNA was present in the spleen of the protected animal necropsied more than a year after challenge (though viral DNA was still present). No neutralizing responses could be demonstrated, but CTL activity was detected sooner and at higher levels after challenge in protected than in unprotected macaques. In this novel HIV-2 vaccine model, protection was clearly dose-dependent, and clearance of challenge virus RNA from the plasma did not require detectable ongoing replication of the immunizing virus at the time of challenge.

Characterization of a molecular clone of HIV type 2 infectious for Macaca nemestrina

A lambda phage clone containing a full-length HIV-2 provirus, designated HIV-2KR, was obtained from the genomic DNA of Molt4 clone 8 (Molt4/8) lymphoblastic cells infected with the HIV-2PEI2 strain. HIV-2KR is genetically distinct from known HIV-2 isolates, possessing both a unique deletion in the LTR promoter region, and a long rev reading frame. It is replication competent in vitro after transfection into Molt4/8 cells, replicates in a variety of established human T lymphoblastic (Molt-3, Molt4/8, SupT1, H9, C8166) and myelomonocytic (U937) cell lines, and displays prominent cytopathic effects on infection of Molt4/8 cells, reflecting usage of both CCR5 and CXCR4 coreceptors. In addition, HIV-2KR was found to be infectious for human and Macaca nemestrina peripheral blood lymphocytes, and primary human monocyte-macrophage cultures. Intravenous inoculation of cell-free virus into M. nemestrina resulted in infection characterized by transient, low-level viremia and modest temporary decline in CD4 lymphocyte numbers, making HIV-2KR the first HIV-2 molecular clone reported to be infectious for this primate species.