Background:
Blood group antigens are expressed on red blood cells however; these antigens can also
be expressed on some other cells particularly on the surface of epithelial cells and may be
found in mucosal secretions. The gene known to determine the secretion of these blood
group antigens is the Secretor Fucosyltransferase 2 (FUT2) gene. In many human
populations 80% secrete ABO antigens (termed secretors) while 20% do not (termed nonsecretors).
Furthermore, there are disease conditions that are associated with secretor
status. It is against this background that this study was proposed.
Hypothesis:
There are correlations between Secretor FUT2 gene polymorphisms and blood group
antigen secretor status; non-secretors are less susceptible to HIV infections.
Broad Objective:
To investigate associations between mucosal blood group antigen expression profiles
(secretor status), Secretor FUT2 gene polymorphisms and susceptibility to HIV infection
among female sex workers in Nairobi, Kenya.
Study Design:
This was a cross-sectional study.
2
Study Area:
This study was conducted at the Nairobi Regional Blood Transfusion Centre, and at sex
worker outreach program clinics, from the Pumwani Majengo female sex worker cohort,
Nairobi, Kenya.
Study Population:
This study enlisted 142 adults of both gender (male and female, aged 18-65 years) from
the regional blood transfusion centre in Nairobi, Kenya. In addition, this study recruited
280 female sex workers from the well-established Pumwani Majengo cohort aged 18 to
65 years of age (n=422).
Materials and Methods:
Blood, saliva and female genital tract (vaginal and cervical) specimens were collected
from each study participant once informed written consent was obtained. The laboratory
analyses were carried out at the KAVI Institute of Clinical Research (KAVI-ICR) and the
University of Nairobi Institute of Tropical and Infectious Diseases (UNITID)
laboratories, Nairobi. Blood typing was determined using standard serological techniques
using monoclonal antibodies to the ABH, Rhesus (D) and Duffy (Fya, Fyb) blood group
antigens. Secretor phenotyping was determined using lectins specific to blood group H
antigen in both salivary and female genital tract samples. This was correlated to the HIV
sero-status. The correlation of secretor phenotypes to CD4+ T cell counts, was based on
retrospective data analyses, following immunophenotyping using a CD3/CD4/CD45
panel on an BD LSR II flow cytometer. For secretor genotyping, DNA was extracted
3
from frozen whole blood samples. Quantitative real-time amplification was performed
based on the polymerase chain reaction (PCR) on a Rotor-gene Q (Qiagen) and PCR
products sequenced on the next-generation pyrosequencer, Pyromark Q24 platform
(Qiagen).
Results:
Objective 1: ABO blood group phenotype frequency distribution was O>A>B>AB
comprising 199 (47.2%), 120 (28.4%), 84 (19.9%) and 19 (4.5%) respectively with 408
(96.7%) Rhesus (D) positive cases. Duffy positive phenotypes were reported in 2 (0.47%)
of the study participants.
Objective 2: Saliva testing showed that among the blood donors, 121 (85%) were
secretors and 21 (15%) were non-secretors, while among the female sex workers, 212
(76%) were secretors and 68 (24%) non-secretors
Objective 3: Based on HIV screening, 92 (32.9%) of the female sex workers were HIV-1
infected and 188 (67.1%) HIV-1 uninfected. There was a correlation between HIV
infections and secretor phenotypes. The proportion of secretors was significantly higher
among women with HIV infection (77/92 = 83.7%) in comparison to HIV un-infected
women (135/188 = 71.8%) (p=0.029). Furthermore, the incidence of HIV infection was
significantly higher among blood group A secretors (p=0.028) in comparison to O
secretors, but not B and AB.
Objective 4: The correlation to CD4+ T cell counts demonstrated, although the secretors
were more susceptible to HIV, following infection, the ABH secretors maintained
elevated levels of CD4+ T cell counts in comparison to non-secretors and this difference
4
was statistically significant (p=0.005).
Objective 5: Secretor FUT2 genotyping demonstrated the secretor status of the
population was not dependent on the allele at position 428 as has been previously
reported in African populations, suggesting the secretor and non-secretor phenotype
profiles may be due to novel polymorphisms in the Kenyan population.
Conclusion:
The prevalence of ABO, Rhesus and secretor/non-secretor phenotypes are similar to the
global profiles as previously reported across African populations. However, the
underlying genotypic variation resulting in the non-secretor phenotype in a subset of the
Kenyan population as screened among our study cases does not fall under a the „se
428
umbrella‟ of a restricted geographical distribution as has been previously reported. These
findings further suggest the non-secretor phenotype may confer a certain degree of
protection against HIV infection, as there were higher HIV infection incidence rates
among ABH secretors; but following infection, the same population of secretors
maintained elevated CD4+ T cell counts.
Recommendations:
1. It is important to know the frequencies of various blood group antigen phenotypes in
a population. This information is important to confirm the prevalence of both major
and minor blood group phenotypes in a population. There are a number of blood
group phenotypes, which have been associated to varying disease/infection
susceptibilities as seen in this study, blood group A individuals were at an increased
5
risk for HIV infections.
2. Secretor and non-secretor phenotype profiles in populations should be determined.
This data serves as a baseline for investigative studies on the correlation between
secretor and non-secretor phenotypes and associations to disease conditions. In the
Kenyan population, secretors were found to have a significantly higher incidence of
HIV infection rates.
3. HIV awareness programmes, should include information on secretor status testing as
individuals who are ABH secretors are at an increased risk of HIV infections.
4. HIV treatment and care packages, should include secretor testing as non-secretor
individuals have been shown to progress faster to infection, based on a rapid decline
in CD4+ T cell counts in comparison to the secretor counterparts.
5. Secretor testing should be comprehensive, based on both phenotyping and
genotyping. There may be silent and/or novel mutations specific to a population,
which confer varying phenotypic traits, hence dual screening will ensure an
individual is ascertained of their secretor status.