The role of CCL3 and CCL4 encoding genes in control of HIV-1 infection
Date
2017
Authors
Mncube, Sizanani
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Two CC chemokine ligands that bind the CCR5 receptor (major coreceptor used by HIV-1), namely
CCL3 and CCL4, are encoded by the CCL3 and CCL4 genes found in 2 copies per diploid genome
and their non-allelic variants, CCL3L and CCL4L, respectively, which are found in variable copy
numbers. CCL3L and CCL4L gene copy numbers lower than the population specific median have
been associated with various HIV-1 related outcomes including increased susceptibility to HIV-1
infection, faster rate of progression to AIDS and high viral loads. These associations have however
been controversial and the controversy has in part been attributed to technical challenges of accurate
high copy number determination. In this study we investigated the association of the variable copy
number genes, CCL3La and CCL3Lb (the sum of which is CCL3L) and CCL4La and CCL4Lb (the
sum of which is CCL4L) and previously identified CCL3 haplotypes (Hap-A1, Hap-A3 and Hap-
2SNP) in natural control of HIV-1 infection in the South African black population.
The study was conducted on antiretroviral (ARV) naive HIV-1-infected controllers (HICs, N=52),
comprised of elite controllers (ECs, N=11), viraemic controllers (VCs, N=30) and high viral load
long term non progressors (HVL LTNPs, N=11) and HIV-1-infected progressors (N=74). HIV-1
controllers were also categorised and analysed on the basis of their viral loads as individuals with
less than 400 RNA copies/ml (<400 HICs, N=20) and greater than 400 RNA copies/ml (>400 HICs,
N=32). Droplet digital PCR (ddPCR), which is a much more accurate method for determination of
high copy numbers compared to real-time PCR, was used to determine CCL3L and CCL4L gene
copy numbers. We developed and validated ddPCR assays for CCL3L copy number determination
and used previously developed ddPCR assays for CCL4L copy number determination. CCL3L and
CCL4L gene copy numbers were compared between HICs, HIC subgroups and progressors,
individually or in combination as continuous variables or stratified around the population median.
CCL3 haplotypes were determined using previously designed CT shift real-time PCR assays and
these were compared between HICs, HIC subgroups and progressors.
We found CCL3La copy number to be equal to CCL4L copy number in all individuals (with the
exception of one individual who had 8 copies of CCL3La and 7 copies of CCL4L), thus any
associations involving these genes would mirror each other and were designated as
CCL3La/CCL4L. CCL3L and CCL4L copy numbers compared as continuous variables did not yield
any significant differences between HICs, HIC subgroups and progressors. However, a strong trend
of higher representation of individuals with high CCL3La/CCL4L copy numbers was seen in VCs
(p=0.07) compared to progressors. When individuals were grouped and compared according to
copy number > population median (high) and ≤ population median (low), significant
overrepresentations of individuals with CCL3La/CCL4L copy number > population median were
seen when comparing VCs (p=0.03, OR=0.39) and <400 HICs (p=0.01, OR=0.21) to progressors.
Comparing the total controller group (HICs) to progressors, however only showed a strong trend of
overrepresentation of individuals with CCL3La/CCL4L copy number > population median (p=0.05,
OR=0.4). Comparison of HICs, HIC subgroups and progressors having high CCL3L and/or high
CCL4L copy numbers revealed a significant difference in VCs (p=0.04, OR=0.28) compared to
progressors, with VCs having a higher representation of individuals with high CCL3L and/or
CCL4L. Combined associations were however more significant when comparing individuals with
high CCL3La and/or CCL4La with HICs (p=0.01, OR=0.38), VCs (p=0.02, OR=0.31) and <400
HICs (p=0.01, OR=0.21) having a higher representation of individuals with high CCL3La and/or
CCL4La compared to progressors.
Given that the CCL3L and CCL4L gene copy numbers occur in different combinations in different
individuals, we next explored the specific patterns of copy number possession by assigning pattern
identifiers (IDs) to each individual based on the number of copies of CCL3L, CCL3La and CCL3Lb
or CCL4L, CCL4La and CCL4Lb. The 651 CCL3L pattern (i.e 6 copies of CCL3L, 5 copies of
CCL3La and 1 copy of CCL3Lb) was overrepresented in ECs (p=0.02, OR=0.17) and <400 HICs
(p=0.04, OR=0.26) whereas a 633 CCL4L pattern was overrepresented in HICs (p=0.03, OR=0.07)
and >400 HICs (p=0.03, OR=0.06) with a combined 550541 CCL3L and CCL4L pattern being
overrepresented in VCs (p=0.02, OR=0.05). Moreover, we found a CCL3La and CCL3Lb copy
number difference of 2 to be underrepresented in HICs (p=0.03, OR=3.58) compared to
progressors, with a copy number difference of 5 being overrepresented in VCs (p=0.02, 0.29) when
compared to progressors.
No significant differences were seen when CCL3 Hap-A1, Hap-A3 and Hap-2SNP allelic and
genotypic frequencies were compared between HICs, HIC subgroups and progressors, however a
trend of higher representation of Hap-A1 heterozygosity was seen in >400 HICs when compared to
progressors (p=0.09, OR=0.14) or healthy HIV-1 uninfected controls (p=0.09, OR=0.20).
Additionally, combined analysis of CCL3 haplotypes and CCL3La copy numbers does not seem to
be more significantly associated with natural control of HIV-1 infection compared to CCL3La copy
numbers alone, with the exception of CCL3La and Hap-A3 which showed a slightly higher
significant association in VCs (p=0.02, OR=0.11) compared to progressors.
Overall, our findings suggest that the CCL3 and CCL4 chemokines play a role in natural control of
HIV-1 infection, in agreement with previous studies which found high CCL3L copy numbers to be
protective against disease progression. However, future studies with larger sample sizes would be
beneficial to further elucidate the associations of the CCL3L and CCL4L gene copy numbers with
natural control of HIV-1.
Description
A dissertation submitted to the Faculty of Health Sciences, University of the Witwatersrand,
Johannesburg, in fulfilment of the requirements for the degree of Master of Science (Med)
Virology. Johannesburg, 2017.