|[Frontiers in Bioscience 1, e42-54, August 1,1996]|
MYCOPLASMAS AND HIV INFECTION: FROM EPIDEMIOLOGY TO THEIR INTERACTION
WITH IMMUNE CELLS
Catherine Brenner, Olivier Neyrolles, Alain Blanchard
Institut Pasteur, Unité d'Oncologie Virale, Département SIDA et Rétrovirus, 28, rue du Dr. Roux, 75724 Paris Cedex 15, France
Received 07/05/96; Accepted 07/09/96; On-line 08/01/96
It was difficult to evaluate the prevalence of these mycoplasmas (M. fermentans and M. penetrans) in HIV-infected subjects from the initial studies based on isolation procedures, because control groups were not always included and the number of recruited patients was limited. Various epidemiological studies were undertaken concomitantly with the development of detection methods including direct procedures such as immuno-histochemistry and PCR-based assays, or indirect methods such as enzyme-linked immuno-sorbent assay (ELISA) and Western blotting (WB) for anti-mycoplasma antibodies in patients' sera.
The presence of M. fermentans in AIDS patients was first reported in immuno-histochemistry studies (29, 38, 39) (see Table 1). The work of Lo et al. (29) indicated the frequent presence of M. fermentans in numerous tissues of patients dying with AIDS. Another group reported a much lower prevalence with only one patient out of 42 being positive at autopsy for this mycoplasma (38). Mycoplasma antigen was detected in spleen, liver and bone marrow of this patient whereas the brain, heart, testis and lymph nodes were negative. From these autopsy investigations, it is very difficult to evaluate the significance of the presence of M. fermentans in tissues of patients who died with an extensive destruction of their immune system.
M. fermentans has also been detected by immuno-histochemistry and electron microscopy in renal tissue taken at autopsy from a series of patients with AIDS (39). Twenty patients out of 203 had evidence of AIDS-associated nephropathy (AAN) and M. fermentans was detected in all the 15 renal tissue samples available from these patients. Renal tissues from patients dying with AIDS without renal disease or without AIDS showed no M. fermentans-specific staining. A case report from another group confirmed the association between this mycoplasma and AAN (40).
A M. fermentans-specific PCR detection assay was developed by Wang et al. (41). Various studies using this assay report approximately the same M. fermentans prevalence, about 10%, in peripheral blood from HIV-infected patients (42-44) (table 1). Interestingly, M. fermentans was also detected in throat and genital specimens whereas M. genitalium and M. pneumoniae were not detected in these sites or in blood samples (44); (43). However, it seems that M. fermentans is not a common respiratory pathogen in AIDS patients because it was not detected in or cultured from bronchoalveolar lavages from HIV-infected subjects (45). The M. fermentans prevalence in blood from HIV seronegative individuals was evaluated in various control groups: M. fermentans was found in 9% of 73 seronegative patients that consulted for sexually-transmitted diseases (44), whereas Hawkins et al. (42) did not detect this mycoplasma in any of 26 blood donors.
These results are in apparent contradiction with those from another study which reports no evidence for mycoplasmas in peripheral blood mononuclear cells from HIV-infected subjects (46). However, these authors used a PCR assay with mycoplasma genus-specific primers for which the sensitivity of M. fermentans detection was not provided. Therefore, it is likely that these negative results were due to a lack of sensitivity rather than the absence of mycoplasma.
Serological assays were developed, on an empirical basis, for the detection of M. penetrans-specific antibodies (47). The antigen preparation used is a Triton X-114 extract of mycoplasma cells, containing 2 major polypeptides of apparent molecular mass of 35 (P35) and 38 kDa (P38) (47). P35 and P38 are lipoproteins and the amino acid sequence deduced from p35 gene sequence contains a signal sequence (V21ATVPVIVSSC31) with a specific acylation site (48). The paper of Grau et al. 49) brought some confusion to the nomenclature by designating P35 and P38 as P38 and P42, respectively. In addition, the criteria for defining a M. penetrans-positive serum differs between the 2 groups: whereas a positive ELISA result is sufficient for Wang et al. (47), Grau et al. (49) propose that this must be confirmed by reactivity for P35.
The initial study of Wang et al. (47), as shown on Figure 2, established that 40% of AIDS patients and 20% of asymptomatic HIV infected patients were M. penetrans seropositive, whereas only 0.9% of STD patients and 0.3% of blood donors were M. penetrans seropositive. An association between M. penetrans and HIV infections was therefore suggested. However, others propose rather an association between this mycoplasma and homosexual practices (50). An epidemiological study by the same group indicated that HIV-infected male homosexuals whether suffering from AIDS or asymptomatic were much more frequently M. penetrans seropositive than other risk exposure categories (1% for intravenous drug users, and 0.6% among haemophiliacs) (51). The M. penetrans seroprevalence among French patients is lower: 18% of HIV-infected patients and 1.3% of HIV-seronegative patients were found to be M. penetrans seropositive (Figure 2). Both studies were consistent with an association between HIV and M. penetrans and suggest an association between homosexual behaviour and M. penetrans, and demonstrate that M. penetrans seroprevalence increases with progression of the disease. However, the two studies differ about the association between M. penetrans and Kaposi's sarcoma (KS): Wang et al.(51) found an association, whereas Grau et al. (49) did not.
A novel herpes virus, HHV8, has been detected in all KS lesions tested and its detection in the blood is predictive of KS development. This has changed our view of the genesis of the disease (52, 53). Whether M. penetrans also contributes to or facilitates KS development is unclear.
There is little data available concerning the serological response to M. fermentans. Unlike for M. penetrans, no serological assay has been specifically developed for this species. Without indicating the method of detecting antibodies, it was reported that 15% of HIV-infected patients vs. 1.3% of HIV-seronegative individuals were M. fermentans-seropositive (51). Other serological studies based on growth or metabolic inhibition have given varied results, which are difficult to interpret (Table 1).