Study population and collection of specimens.
Our study population was recruited in the rural provinces of Abel Iturralde (the lowlands of the Amazonas at altitudes of 150 to 600 meters above sea level) and Caranavi (the adjacent lower highlands at 1000 meters above sea level) from the Department of La Paz in Bolivia during five field trips; three in the Abel Iturralde province during the dry seasons (July to October) between 2015 and 2017 and two in Caranavi in March 2018 and May 2019. Our research group, a unit working at the Universidad Mayor de San Andrés (UMSA), visited the three villages Tumupasa, San Silvestre and Santa Rosa de Maravilla (mostly indigenous population below 1200 inhabitants) (13), the small town San Buenaventura (population approximately 8000) and the large town Caranavi (population of 50 000) and spread information through the local authorities explaining our objectives and the benefits of our study.
Participants were recruited either by direct invitations to small mother groups, unions, local authorities and indigenous leaders or via general invitations by radio and television. The examination and collection of samples was done in the nearest local health center or hospital by laboratory technicians from UMSA. Women who were menstruating or were more than 12 weeks pregnant were excluded from the study. Eligible participants (N=394) received extended oral and written information regarding STIs and signed an informed consent form. In the case of women under the age of 18 informed consent was signed by their guardians. Demographic data was collected through a questionnaire and a personal interview. Blood samples (N=389) and cervicovaginal cell samples (N=376) were collected from female volunteers.
Because of the lack of sanitary services (including electricity) in some of these areas, sampling was performed using filter papers that preserve the biological material and allowed the transportation and assessment of the samples. Whole blood samples were therefore collected as Dried Blood Spots (DBS) in Whatman 903 filter papers (Protein Saver TM 903 Card) for the evaluation of antibodies against HSV-2, HBV and HIV as well as for HBV antigen. Exfoliated cervicovaginal cells were collected by swabbing the lateral walls using sterile cotton swabs that were then transferred in 1 mL of essential D-MEM medium (Sigma–Aldrich, MO, USA) and kept at 4ºC. At the laboratory in La Paz city the samples were centrifuged at 1500 rpm for 10 minutes at 4ºC and suspended cells in 150 µL of medium were transferred and applied on 2 × 2.5 cm Nucleic-card color matrix spots (NUCLEIC-CARD Thermo Fisher Scientific, USA) as dried cervicovaginal cell spots (DCCS). DBS and DCCS were kept at -20ºC until their further transport to and analysis in Sweden. Data was managed anonymously according to the ethical permission CEI-UMSA 0215 obtained from the Ethical committee at UMSA, La Paz, Bolivia.
Elution of DBS
Blood proteins were extracted from DBS by punching one bloodstained circle of 5 mm diameter from the filter paper as previously described (14, 15), The spots were soaked for 24 hours at 4ºC in 300 µL of diluent buffer. The papers were removed and the elutions were centrifuged at 4500 rpm for 10 minutes. A total volume of 250 µL of elution was obtained from each spot and either assessed immediately (see below) or kept at -20ºC until analysis.
Extraction of DNA from the vaginal spots
DNA was extracted and purified from DCCS using QIAGEN mini kit as described (16), One spot of 2 × 2.5 cm was placed in 600 µL of sterile PBS buffer in a 1.5 mL tube for 1 hour at 4ºC. After that, 400 µL of the supernatant was collected in a different tube in order to do the lysis and the extraction of genomic DNA according to the manufacturer’s instructions. About 150 µL of eluted purified DNA was obtained and quantified by using a Nanodrop N-1000 instrument. DNA samples were stored at -20ºC until assessment.
Elutions of DBS were used to assess antibodies for HSV-2, HBV, and HIV using commercial ELISA kits; HSV-2 (Herpeselect2, FOCUS Diagnostics), anti-HBV anti-core antibodies (Murex anti-HBc total, Diasonin S.p.A. UK Branch), and HIV (Murex HIV-1.2.0, Diasonin S.p.A. UK Branch). In addition, we measured the presence of HBV surface antigens in eluted DBS using ELISA (Murex HBsAg Version3, Diasonin S.p.A. UK Branch). As the dried blood samples contain lysed red blood cells, which can interfere with ELISA readout and give false-positive results, we increased the cut-off values provided by the manufacturers of the ELISA kits by multiplying with a factor of 1.5 (17, 18).
Equivocal and low positive samples for HSV-2 by FOCUS ELISA were confirmed with the detection of antibodies to the mature portion of glycoprotein G-2 using Western blot as described (19). Briefly, HSV-2 antigens were subjected to polyacrylamide gel electrophoresis under reducing conditions by using NuPAGE 7% Tris-acetate gels (Novex). Antigens were electrotransferred to a membrane (Milipore Corp.) and the strips were incubated overnight with eluted samples at a 1:3 dilution. Peroxidase-labeled rabbit anti-human IgG (DAKO) was used as conjugate and 4-chloro-1-naphthol was used as substrate to visualize binding. A positive profile was defined as reactivity to mgG-2 (~120kDa).
Purified DNA from DCCS were assessed by a Taqman real-time PCR assay which identifies 12 high-risk (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58 and 59) and two low-risk (6 and 11) types, by targeting segments of the E6/E7 region (20). The amplification was performed in 8 duplex real-time PCR reactions, including the 14 HPV types and the human gene (the β-globin gene) which served as a control for sample quality and overall PCR efficiency. The PCR was run for 45 cycles (15 s at 95°C, 60 s at 58°C) on an ABI 7500 instrument (Applied Biosystems). Serial dilutions of pUC57 plasmids containing target segments of each HPV type, sized 82-134 base pairs (synthesized by GenScript Corp.) were used to verify a high PCR efficiency. Only samples yielding at a cycle threshold (CT) for β-globin below of 37 were included in the analysis.
Descriptive data are presented as percentages, median, prevalence estimates and 95% confidence intervals (CIs). Statistical analyses were calculated using χ2 test comparing among groups using PRISM 7.0® from GraphPad Software Inc., San Diego, CA. A p value below 0.05 was considered statistically significant.
The associations between the dependent variable (viral STI) and the independent variables; age, occupation, number of children, current family planning methods, residence and coinfections were tested by binary logistic regression. This analysis was used to estimate crude and adjusted odds ratios and 95% CIs for each category compared to a reference category (Ref). Variables or categories that showed association with the outcome at p<0.1 were included in the adjusted models as in other studies (21). Data analysis were performed using SPSS 24.0 IBM, Chicago, USA.