Population characteristics
The present study sought to look at the variation within the L1 region of the HPV genome in 98 samples. All participants were female and a subpopulation was from Botswana (n = 72) and another from Kenya (n = 26). The distribution of HPV genotypes stratified by age and country are shown in Figure 2. Out of 98 generated sequences, 83.7% (82/98) participants had HR-HPV genotypes while 16.3% (16/98) had LR-HPV genotypes. However, in some patients, we identified other additional HPV genotypes which were previously missed when the same samples were genotyped using Abbott and LA-HPV. Among participants with HR-HPV genotypes, 51.2% (42/82) were coinfected with HIV (HIV/HPV), while 47.5% (39/82) were from Botswana and 3.7% (3/82) were from Kenya based on MY09/11 Sanger sequencing-based method. We could not assess the cancer stage of HPV sequences isolated from Kenyan participants because their demographics and cancer stage information at diagnosis were not available. The most predominant cancer stage among participants from Botswana was stage 2. Amongst individuals with LR-HPV genotypes, 12.5% were infected with HIV, cancer stage 3 was the most predominant cancer stage. The prevalence of HR-HPV genotypes was statistically higher in the Botswana population compared to Kenya (p-value < 0.001). We did not record any statistical significance in other variables among HR and LR- HPV genotypes among the Botswana population.
Table 1. Baseline demographics for participants
|
Total n=98
|
HR-HPV
n=82
|
LR-HPV
n=16
|
P-value
|
Median Age in years Median (IQR)
|
50 (42-54)
|
50 (42-61)
|
51.5 (47.5-57)
|
0.67a
|
HIV Status, n (%)
HIV negative
HIV positive
|
31 (31.6)
41 (41.8)
|
27 (32.9)
39 (47.6)
|
4 (25)
2 (12.5)
|
0.22b
|
Country, n (%)
Botswana
Kenya
|
72 (73.5)
26 (26.5)
|
66 (80.5)
16 (19.5)
|
6 (37.5)
10 (62.5)
|
<0.001b
|
Cancer Stage, n (%) Botswana only
Stage 1
Stage 2
Stage 3
Stage 4
|
6 (6.1)
21 (21.4)
15 (15.3)
1 (1.0)
|
6 (7.3)
18(22.0)
14 (17.0)
1 (1.2)
|
0 (0.0)
3 (18.8)
1 (6.3)
0 (0.0)
|
0.69b
|
a P-value was calculated using the Rank sum test
b P-values were obtained by chi-square test
HIV, human immunodeficiency virus; HPV, human papillomavirus; HR, high risk; LR, low risk; IQR, interquartile range
HPV genotypes
Human papillomavirus genotypes were identified using both phylogenetic tree (Figure 3) and online L1 Taxonomy Tool Analysis. After quality control (QC), 3 sequences were excluded for mutation and phylogenetic analyses because sequences were too short (>450bp). In total, 19 HPV genotypes were determined, -6, -11, -16, -18, -35, -39, -42, -45, -53, -54, -55, -56, -58, -59, -73, -81, -89, -122, and -159 (Figure 2). Out of 19 different genotypes found in Botswana and Kenya populations, five showed statistically significant differences in their frequency (p-value < 0.05) between countries (Figure 2). Thus, HPV genotypes -6, -54 and -73 were higher in Kenyan population while genotypes -16 and -18 were recorded higher in Botswana population. The WLWHIV had more HPV genotypes compared to WNLWHIV even though the difference was not statistically significant (p-value > 0.05).
Phylogenetic analysis
Phylogenetic analysis included all the 95 HPV sequences obtained in this study that were adequate for phylogeny (>400bp sequence length). All the studied HPV sequences clustered with reference and had posterior probabilities > 90%, and could be used to assign the genotypes with confidence. Phylogenetic inference with maximum likelihood and MCMC methods showed that there were no isolated clusters among HPV sequences from both countries. This was also true when trees for HPV-16 (n=29) and HPV-18 (n=12) sequences from this study and the respective GenBank references for the L1 gene region were constructed (Figure 4). However, there was a general increase in nucleotide genetic diversity among the HPV sequences isolated from WLWHIH as opposed to WNLWHIV (Figure 4) as shown by multiple branches within the phylogenetic tree. The overall mean pairwise distances of HPV-16 and HPV-18 sequences isolated from WLWHIV versus WNLWHIV were not statistically significant.
Mutations Analysis
Nucleotide Diversity
In total, 68 nucleotide base substitutions were detected using reference sequences for each genotype. Table 1 summarises the different polymorphisms detected per genotype and only genotypes with counts greater than 1 per strata were considered. Although there were more HPV types among WNLWHIV, most of the HPV genotypes among WLWHIV were HR-HPV genotypes.
Mutations at the Amino Acid level
The present study defined mutations as amino acid changes that differed from that of reference sequence and candidate escape mutations were those that have not yet been reported in 1000 most similar sequences. Six (6) new variants were identified based on the sequencing of the L1 region, HPV-16 (L441P, S343P), HPV-18 (S424P), HPV-45 (Q366H, Y365F), belonging to the HPV HR-HPV group and HPV-84 (F458L) belonging to the LR-HPV group.
Table 2. Novel genetic mutations found in the 5′ and 3′-ends of HPV-16, -18, -45, -84 -L1 regions.
HPV genotype
|
Mutations
|
HIV status
|
Cancer Stage
|
Count of Sequences in GRS
|
Deleterious
|
16
|
L441P
|
Positive
|
2
|
-
|
√
|
S343P
|
Positive
|
n/a
|
-
|
√
|
18
|
S424P
|
Positive
|
n/a
|
-
|
√
|
45
|
Q366H
|
Positive
|
3
|
51
|
√
|
84
|
Y365F
|
Negative
|
2
|
-
|
√
|
F458L
|
Negative
|
n/a
|
-
|
√
|
GRS, gene recruitment sequence, HIV, human immunodeficiency virus; HPV, human papillomavirus; n/a, not applicate