Susan F. Lyons

A Predominantly HIV Type 1 Subtype C-Restricted Epidemic in South African Urban Populations

395 SOUTH AFRICA is now reaching the advanced stages of the HIV-1 epidemic, and in 1997 more than 16% of pregnant women attending government clinics were HIV seropositive.1 There is geographical variation in the timing and severity of the epidemic within South Africa, with the northeastern region of the country (KwaZulu-Natal Province) having a prevalence of 26.92% compared with the most southern province (Western Cape), having a prevalence of only 6.29%. In a study in Cape Town, located in Western Cape Province, subtype B was associated with homosexual transmission and subtype C with heterosexual transmission.2 On the basis of the age of the epidemic and the genetic distance between gag sequences, this study suggested multiple introductions of subtype C strains into this region of South Africa. HIV subtype C has also been identified in mine workers from the north central Gauteng Province of South Africa,3 as well as in women of Durban (KwaZulu-Natal Province).4 Owing to differences in the timing of the epidemic in different regions of the country, ethnic diversity within South Africa, as well as the presence of emigrants and refugees from other African countries, it is possible that other subtypes may have been independently introduced into different regions of the country. To assess the distribution of HIV-1 subtypes in South Africa, blood samples were obtained from 107 infected individuals who originated from five major cities situated in four geographically distinct regions of the country. HIV-1 provirus was subtyped by heteroduplex mobility assay (n 5 107); seven of these samples were sequenced in the V3±V5 region and the complete gp120 was sequenced for one subtype C isolate. In addition, to assess the relationship among HIV-1 subtype C sequences from the southern African subcontinent, these sequences were compared phylogenetically with 140 published V3 (env) sequences, 105 of which were subtype C sequences originating from the southern African region. EDTA blood was obtained from a total of 107 infected individuals attending urban clinics located in four provinces of South Africa: Johannesburg (n 5 34) and Pretoria (n 5 5) (Gauteng Province); Durban (n 5 20) (KwaZulu-Natal Province); Bloemfontein (n 5 28) (Free State Province); and Cape Town (n 5 20) (Western Cape Province) (see Table 1). Samples were collected between 1994 and 1996. DNA was extracted, the 700-bp V3±V5 region was amplified by nested polymerase chain reaction (PCR), and PCR fragments were subtyped by heteroduplex mobility assays (HMA) as described.2 The 700-bp V3±V5 fragment of seven samples, and the complete gp120 region of the env gene (1600 bp) from one of these samples (ZA347TS), were cloned and sequenced as described.2

High Human Immunodeficiency Virus Type 1 RNA Load in the Cerebrospinal Fluid from Patients with Lymphocytic Meningitis

Thirty-seven matched cerebrospinal fluid (CSF) and plasma samples from 34 human immunodeficiency virus type 1 (HIV-1)-infected patients with suspected meningitis were analyzed for levels of HIV-1 RNA and markers of inflammation. Patients with tuberculous (n = 9) or cryptococcal (n = 6) meningitis had the highest CSF virus loads, which in many cases exceeded the levels in plasma, compared with patients with meningococcal meningitis (n = 3), aseptic meningitis (n = 8), tuberculoma (n = 2), or AIDS dementia complex (n = 4) or with normal lumbar punctures (n = 3). CSF virus load correlated significantly with the number of infiltrating lymphocytes (r = .60, P < .001) but not with plasma virus load, the levels of beta2-microglobulin in the CSF, or the integrity of the blood-brain barrier. These data suggest significant intrathecal HIV-1 replication in patients with lymphocytic meningeal infections such as tuberculous and cryptococcal meningitis.

Restriction fragment length polymorphism analysis for rapid gag subtype determination of human immunodeficiency virus Type 1 in South Africa

A rapid method for identification of human immunodeficiency virus Type 1 (HIV-1) gag subtypes was developed based on restriction fragment length polymorphism (RFLP) analysis of 400 or 650 bp long polymerase chain reaction (PCR) fragments encompassing the start of the p17 (400 bp) and part of the p24 (650bp) regions. The consensus sequences of subtypes A-D, the only subtypes identified in South Africa, were analyzed to detect restriction endonucleases which generate unique patterns for each subtype. Four restriction endonucleases were identified: AluI, AccI, SwaI and XmnI. Digestion of a 400 bp fragment with AluI allowed identification of subtype C. Samples not identified were then reamplified, and a 650 bp fragment digested with AccI to identify subtype B, followed by SwaI and XmnI to distinguish between subtypes A and D. This strategy was applied to 87 samples previously subtyped by either sequence analysis of the gag p17 region (n = 33); or heteroduplex mobility assay (HMA) based on the env gene (n = 75); or both (n = 21). Out of the 87 samples, RFLP identified two samples as subtype A, 28 as subtype B, 56 as subtype C and one as a subtype D virus. No discrepancies were found between RFLP gag subtypes and gag sequence subtypes demonstrating the reliability of this method. There was also no discordance between gag RFLP subtypes and env HMA subtypes, suggesting that there were no recombinant viruses detected relating to the genomic regions analyzed. RFLP is an effective technique for the rapid screening in an HIV epidemic of limited diversity, such as in South Africa.

Identification of cell subsets expressing intracytoplasmic cytokines within HIV-1-infected lymph nodes

Objective: To describe the endogenous cytokine profile of HIV‐1‐infected lymph nodes (LN) and to identify the phenotype of individual cells expressing intracytoplasmic cytokines. Design and methods: Whole LN biopsies were collected from three HIV‐seronegative controls and four HIV‐1‐positive individuals with persistent generalized lymphadenopathy. Three had established infection (Centers for Disease Control and Prevention 1993 criteria, stages A2, C1 and C3) and one was undergoing seroconversion illness. A combination of three methods was used to assess the impact of HIV‐1 on LN architecture and endogenous cytokine expression. Immunocytochemistry was used to locate follicular dendritic cells (FDC), interdigitating cells and T and B cells. Reverse transcriptase‐polymerase chain reaction was used to assess mRNA for interleukin (IL)‐1&bgr;, IL‐2, IL‐4, IL‐6, IL‐10, tumour necrosis factor (TNF)‐&agr; and interferon (IFN)‐&ggr; in collagenase‐digested LN cells. Three‐colour flow cytometry was used to identify intracytoplasmic cytokine expression within cell subsets. Results: Germinal center (GC) hyperplasia was observed in LN from two patients with established HIV‐1 infection, and the third, coinfected with Mycobacterium tuberculosis, showed extensive necrosis. In the patient undergoing seroconversion, there was an extensive FDC network within the expanded and confluent GC which covered expansive areas of the LN. There was varied expression of IL‐1&bgr;, IL‐4, IL‐6, IL‐10 and TNF‐&agr; mRNA from the four HIV‐1‐infected LN and the patient undergoing seroconversion showed evidence for a mixed cytokine profile, which also included IL‐2 and IFN‐&ggr;. Flow cytometry revealed intracytoplasmic IL‐1&bgr; protein restricted to cells expressing CD19, CD21 and CD38 antigens. Conclusions: Cytokines were detected in freshly isolated HIV‐1‐infected LN cells without requiring an exogenous stimulus. Seroconversion was associated with an expanded FDC network within enlarged GC, bounded by defined mantle zones containing B cells. There was diverse cytokine mRNA expression and IL‐1&bgr; protein was restricted to cells expressing B‐cell‐associated antigens.

Effect of interferon on Vero cells persistently infected with SSPE virus and lytically infected with measles virus

SummaryThe effect of lymphoblastoid interferon (IFN) on viral polypeptides synthesized in a Vero cell culture persistently infected with subacute sclerosing panencephalitis virus (SSPE-Vero) was compared to that of Vero cells infected with exogenous measles virus. After IFN treatment there was no significant decrease in the synthesis of SSPE viral proteins, but inhibition of synthesis of measles polypeptides was readily seen. In Vero and SSPE-Vero cells IFN was able to inhibit replication of Sindbis virus although the effect in Vero cells was significantly more sensitive. In both cell lines IFN was able to stimulate 2′–5′ oligoadenylate synthetase (E enzyme) but not the protein kinase system. The SSPE-Vero cells showed a lower basal level of E enzyme activity than the Vero cells.

Effect of interferon on Vero cells persistently infected with Sendai virus compared to Vero cells persistently infected with SSPE virus

SummaryPersistent infections with Sendai and SSPE virus were established in Vero cells. Sequential passages of these cells were monitored by immunofluorescence and for their sensitivity to the antiviral and antiproliferative effects of interferon (IFN). The cells rapidly developed resistance to the antiviral effect of IFN as judged by the inability of IFN to inhibit the replication of exogenous Sindbis virus. This decrease was accompanied by a reduction in the induction of the 2′–5′ oligo A synthetase. Both cell lines were resistant to the antiproliferative effect of IFN. A decrease or absence of IFN receptors on the surface of the cells was not found to be the cause of their resistance to IFN.

Absence of HIV infection in prostitutes and women attending sexually-transmitted disease clinics in South Africa.

To assess the extent of HIV infection in the Black South African population HIV status and STD serologies were tested in a group of 56 prostitutes and 240 clients of a STD clinic. Sera were tested by ELISA (Elavia Pasteur Institute) and indirect immunofluorescence. There was only 1 serum positive for HIV by both assays a Malawian migrant women in the STD group. STD serologies were highly positive in both groups: 95% and 92% positive for Chlamydia 2% and 4% positive for HBsAg 41% and 37% positive for total HBsAG or anti-HBs and 25% positive for syphilis (STD group only) in the prostitute and STD groups respectively.

One year surveillance of HIV-1 infection in Johannesburg, South Africa

A sero-epidemiological surveillance study to monitor the prevalence of HIV-1 infection in Johannesburg, South Africa, was commenced in February 1988. The population selected for study were attenders at clinics for sexually transmitted diseases (STD) and at family planning (FP) clinics. In the 12 months of the study 6631 sera were tested. Of the STD attenders, 15 of 1224 black females (1.2%) and 21 of 2482 black males (0.8%) were positive. Of the 449 white males tested 49 were homosexual, amongst whom 10 (20.4%) were positive; in the heterosexual white male group 4 of 400 (1.0%) were positive. Of the FP clinic attenders, 4 of 1459 black females (0.3%) were positive. 68 of the 6631 sera tested were indeterminate for infection. No attenders were positive for HIV-2 infection. These data confirmed the entry of HIV infection into the black population in South Africa.

An analysis of indeterminate Western blot patterns of black African subjects

The significance of indeterminate Western blotting results was assessed by determining their distribution in attenders of sexually transmitted diseases (STD) and family planning (FP) clinics. As expected, it was found that the former had a highly significantly greater prevalence of HIV infection. Indeterminate results were not found more frequently in STD attenders refuting any association with sexual promiscuity. They were, however, found significantly more frequently amongst black compared to white subjects, but no difference was found between males and females. Indeterminate results appear to be a feature of black African sera and what the significance is, still remains to be elucidated.

An indirect radioimmunoassay for interferon

An indirect radioimmunoassay for testing the antiviral activity of interferon (IFN) is described. Vero cells are seeded in microtitre plates, treated with appropriate dilutions of interferon and challenged with Sindbis virus. Viral yield is measured using specific antibody and radiolabelled protein A. The assay is able to detect IFN levels of 5 international units (I.U.)/ml, has a high degree of reproducibility, and could be easily adapted to various cell and virus combinations. This microsystem is technically simple, allows testing of small volumes of test material, and eliminates subjectivity in reading of end-points.