Regulatory variants of apobec3 genes potentially associate with covid-19 severity in populations with african ancestry

ABSTRACT Since November 2019, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has caused the worldwide pandemic of the coronavirus disease 2019 (COVID-19), the impact of

which is huge to the lives of world populations. Many studies suggested that such situation will continue due to the endless mutations in SARS-CoV-2 genome that result in complexity of the

efforts for the control of SARS-CoV-2, since the special enrichment of nucleotide substitution C>U in SARS-CoV-2 sequences were discovered mainly due to the editing by human host factors

_APOBEC3_ genes. The observation of SARS-CoV-2 variants Beta (B.1.351) and Omicron (B.1.1.529) firstly spreading in South Africa promoted us to hypothesize that genetic variants of _APOBEC3_

special in African populations may be attributed to the higher mutation rate of SARS-CoV-2 variants in Africa. Current study was conducted to search for functional variants of _APOBEC3_

genes associate with COVID-19 hospitalization in African population. By integrating data from the 1000 Genomes Project, Genotype-Tissue Expression (GTEx), and Host Genetics Initiative (HGI)

of COVID-19, we identified potential functional SNPs close to _APOBEC3_ genes that are associated with COVID-19 hospitalization in African but not with other populations. Our study provides

new insights on the potential contribution of _APOBEC3_ genes on the evolution of SARS-CoV-2 mutations in African population, but further replication is needed to confirm our results.

SIMILAR CONTENT BEING VIEWED BY OTHERS NOVEL GENETIC ASSOCIATION OF THE FURIN GENE POLYMORPHISM RS1981458 WITH COVID-19 SEVERITY AMONG INDIAN POPULATIONS Article Open access 03 April 2024

NEXT-GENERATION SEQUENCING OF HOST GENETICS RISK FACTORS ASSOCIATED WITH COVID-19 SEVERITY AND LONG-COVID IN COLOMBIAN POPULATION Article Open access 11 April 2024 GENETIC VARIABILITY IN

COVID-19-RELATED GENES IN THE BRAZILIAN POPULATION Article Open access 02 April 2021 INTRODUCTION Since November 2019, the emergence of severe acute respiratory syndrome coronavirus 2

(SARS-CoV-2) has resulted in coronavirus disease 2019 (COVID-19), affecting everything from daily routines to mental wellbeing on global populations. Three years into the pandemic,

SARS-CoV-2 continues to have a profound effect on human health, such as the emergence of a new type of disease: long COVID, which is characterized by a diverse range of symptoms, including

over 200 symptoms that have been reported so far1. Study suggests that the ongoing mutations in the genome of SARS-CoV-2 contribute to the complexity of controlling the pandemic by

false-negative result in viral RNA genome sequencing2, with the 12 types of nucleotide substitutions reported. It is noteworthy that the C>U mutation among the 12 types of nucleotide

substitutions occurs much more frequently than other types3. Regarding previously discovered viral mutations of human immunodeficiency virus (HIV) and monkeypox virus (MPOXV), the potential

role of non-random driver proteins, such as Apolipoprotein B mRNA-editing catalytic polypeptide-like 3 (APOBEC3) was assessed and analyzed by researchers4. These APOBEC3 proteins share a

highly conserved zinc-dependent deaminase domain that is activated by a zinc ion that coordinates a water molecule for nucleophilic attack of C4 of the pyrimidine ring of C, which in

conjunction with the glutamic acid residue leads to the deamination of the base to U2. Therefore, the hypermutation events of C>U take place in viral genome during subsequent cellular

infection5,6. Furthermore, the APOBEC3 family comprises cytidine deaminases that contribute to the innate response against retroviral and retrotransposon infection7. They use deaminase and

deaminase-independent mechanisms to suppress various endogenous and exogenous viruses8. Among seven members in the APOBEC3 family, at least five of these enzymes, namely APOBEC3B, APOBEC3D,

APOBEC3F, APOBEC3G, and stable haplotypes of APOBEC3H, exhibit anti-HIV activity9. As one of the potent mutators in the human genome, APOBEC3 stands out as the only one whose expression is

significantly upregulated after HIV infection10. In recent years, MPOXV, a double-stranded DNA virus, has also undergone a surprising surge in mutation and was declared a Public Health

Emergency of International Concern by WHO in 2022. A curious observation is that most cases of HIV and original strains of MPOXV11 were reported in Africa. Notably, the SARS-CoV-2 variants

Beta (B.1.351) and Omicron (B.1.1.529) initially emerged in South Africa and eventually spread to other countries12. Furthermore, the pivotal mutation such as D614G in the Spike protein that

is caused by the C>U mutation in SARS-CoV-2 genome, was identified in almost all Beta (B.1.351) and Omicron (B.1.1.529) isolates, but not in Alpha (B.1.1.7), Gamma (P.1), and Delta

(B.1.617.2) strains13. This suggests that the beta and omicron variants may have contributed a higher number of key mutations to the global COVID-19 pandemic. All of these observations

indicate that the African population may differ from other populations in their role in driving viral mutations, whether in the context of SARS-CoV-2 or other viruses. One potential

explanation for the prevalence of the SARS-CoV-2 variants in Africa is the low vaccination rate on the continent, which might increase the likelihood of virus transmission, subsequently

leading to a higher probability of mutagenesis with each replication cycle and the emergence of multiple variants14. However, despite the vaccine rollout, morbidity and mortality rates have

remained low in Africa. This cannot be adequately accounted for by the younger age of the African population alone, suggesting the potential involvement of genetic factors in COVID-19

susceptibility or severity15. Thus, the current study was conducted to investigate whether functional genetic variants of _APOBEC3_ genes associate with COVID-19 severity in African

populations. MATERIALS AND METHODS ASSOCIATION OF SINGLE NUCLEOTIDE POLYMORPHISMS (SNPS) OF _APOBEC3_ GENES WITH COVID-19 HOSPITALIZATION We firstly downloaded the summary statistics of

COVID-19 hospitalization genome-wide association studies (GWASs) conducted separately among samples with European or African ancestries from the COVID-19 Host Genetics Initiative (HGI,

release 7)16,17. The links to the two GWAS summary statistics are freely available at HGI: (1) HGI-B2-EUR GWAS of hospitalized COVID-19 vs. general population controls among European samples

(https://storage.googleapis.com/covid19-hg-public/freeze_7/results/20220403/pop_spec/sumstats/COVID19_HGI_B2_ALL_eur_leave23andme_20220403.tsv.gz; reference in hg38 build) and (2) HGI-2-AFR

GWAS of hospitalized COVID-19 vs. general population controls with African ancestry

(https://storage.googleapis.com/covid19-hg-public/freeze_7/results/20220403/main/sumstats/COVID19_HGI_B2_ALL_leave_23andme_20220403.tsv.gz; reference in hg38 build). The sample sizes for

both GWASs are as follows: HGI-B2-EUR (cases = 32,519 and controls = 2,062,805) and HGI-B2-AFR (cases = 2,589 and controls = 123,225). Their summary statistics were undertaken standard

quality controls by HGI, and we thus used the data directly in our analyses for _APOBEC3_ genes. The association signals around the _APOBEC3_ genes, including _APOBEC3A_, _APOBEC3B_,

_APOBEC3C_, _APOBEC3D_, _APOBEC3F_, _APOBEC3G_, and _APOBEC3H_, are located in a gene cluster on chromosome 22 (chr22:38, 939, 327–39, 106, 168; hg38). Therefore, the COVID-19 association

signals for the _APOBEC3_ genes in the above two GWASs were extracted, with only top SNPs passed the association _P_ < 0.01 were selected for downstream functional evaluation. The use of

relaxed association threshold was arbitrarily set since we thought that our study was a typical candidate gene study. MINOR ALLELE FREQUENCIES FOR PRIORITIZED _APOBEC3_ GENES ACROSS

POPULATIONS FROM THE 1000 GENOME PROJECT For 6 candidate SNPs passed the nominal association threshold _P_ < 0.01, we determined their minor allele frequencies across multiple major

populations, including African (AFR), Ad Mixed American (AMR), East Asian (EAS), European (EUR), and South Asian (SAS), as well as the corresponding subpopulations for these major

populations from Ensembl database18. Ensembl curated the population frequency data from the Phase 3 of the 1000 Genome Projects19. The module “Population genetics” of Ensembl was used to

extract the allele frequencies of 6 target _APOBEC3_ SNPs. CIS-EXPRESSION QUANTITATIVE TRAIT LOCUS (CIS-EQTL) ANALYSIS OF _APOBEC3_ SNPS IN GENOTYPE-TISSUE EXPRESSION (GTEX) DATABASE For

these top SNPs of _APOBEC3_ genes that passed the COVID-19 association threshold _P_ < 0.01, we annotated them individually with Haploreg420 and the GTEx V821. As both Haploreg4 and GTEx

provide expression quantitative trait locus (eQTL) information for these SNPs or its high linkage disequilibrium (LD) SNPs (R2 > 0.8), we curated all eQTL information for these top SNPs

of _APOBEC3_ genes. When a SNP was not included by the two databases, we obtained one of its high LD SNPs via Haploreg4 in European or African population dependent on whether the SNP was

emerged from HGI-B2-EUR or HGI-B2-AFR GWAS, and then searched for the selected high LD SNPs in the two databases. _APOBEC3_ GENE EXPRESSION ANALYSIS BETWEEN EUROPEAN AMERICAN (EA) AND

AFRICAN AMERICAN (AA) IN GTEX DATABASE The GTEx database encompasses the gene expression data in 49 tissues for individuals with multiple ancestries, such as European ancestry, African

ancestry, as well as Asian ancestry. Recently, a study published by Gay N.R. et al.22 estimated the ancestries of GTEx samples, the results of which are openly accessible. According to Gay

N.R. et al.22, among 838 GTEx individuals, there are 103 individuals with African ancestry, and most of others (n = 715) are EA. Based on these EA and AA sample identification labels

provided by Gay N.R. et al., we mapped RNA-seq gene expression data downloaded from GTEx of each sample to its corresponding ancestry by looking up with the sample names from GTEx and Gay

N.R. et al.22. The GTEx TPM (transcripts per million) matrix was downloaded from this link:

https://storage.googleapis.com/gtex_analysis_v8/rna_seq_data/GTEx_Analysis_2017-06-05_v8_RNASeQCv1.1.9_gene_tpm.gct.gz. We thus conducted differential gene expression analysis by stratifying

_APOBEC3_ genes expression by ancestry among 49 GTEx tissues. The multiple testing adjusted statistical significance was set at _P_ < 1.5 × 10–4 [i.e., 0.05/(7 × 49)] for differential

gene expression analysis between EA and AA samples using SAS statement “lsmeans”, which is included in the SAS procedure “proc GLM” via the freely available SAS OnDemand for Academics.

_APOBEC3_ GENE EXPRESSION AMONG BLOOD SAMPLES DERIVED FROM HEALTHY CONTROLS AND COVID-19 PATIENTS These 7 _APOBEC3_ genes were subjected to differential gene expression analysis in blood

samples of COVID-19 patients and healthy controls published by Thair et al.23 using the COVID19db, a gene expression database related to SARS-CoV-2 infection24. Statistical significance was

determined by one-way ANOVA, with the multiple testing significance threshold set at _P_ < 0.01 = 0.05/7_._ ANALYSIS OF _APOBEC3_ INDUCED C>U CODING MUTATION AMONG ~ 8.4 MILLION

SARS-COV-2 GENOMES FROM AFRICAN, EUROPE, AND NORTH AMERICA Mutational data from SARS-CoV-2 genomes were obtained from the open database shared by nextstrain

(https://nextstrain.org/ncov/open/global/6m). SARS-CoV-2 coding mutations were identified by nextstrain using Wuhan-Hu-1/2019 (GenBank accession number MN908947) as the reference genome. To

quantify the total number of C>U mutations in coding regions and the overall number of coding mutations in each SARS-CoV-2 genome within the downloaded mutational metadata

(https://data.nextstrain.org/files/ncov/open/metadata.tsv.zst), we developed a customized Perl script. And our analysis retained only SARS-CoV-2 genomes that met two criteria: they had at

least 10 coding mutations, and they were sampled from three different geographic regions, including Africa, Europe, and North America. The percentage of C>U coding mutations relative to

the total number of coding mutations (referred to as the "C>U coding mutation percentage") in each SARS-CoV-2 genome was determined using SAS OnDemand for Academics. This

software was also used to generate histograms illustrating the distribution of C>U coding mutation percentages across the three geographic regions. Furthermore, given that the average

C>U coding mutation percentage is approximately 40% across all genomes from these three geographic regions (comprising around 8.4 million viral genomes), we further calculated the

percentage of SARS-CoV-2 genomes exhibiting a higher C>U coding mutation percentage. We stratified SARS-CoV-2 samples within each geographic region based on a C>U coding mutation

percentage threshold of 40%. Finally, we employed a Chi-square test to assess differences in the percentage of SARS-CoV-2 genomes with higher C>U coding mutation percentages between the

African samples and the samples from the other two regions. ISSUE OF MEDICAL ETHICS As the two GWAS summary statistics are freely available at HGI and no patients’ identification information

were revealed by HGI, our study was thus a typical secondary analysis of previously published data. In addition, we didn’t generate any new data in our study. All data published by other

researchers were further processed to investigate the potential genetic variants of _APOBEC3_ genes associated with COVID-19 hospitalization. Based on this, we believed that our study did

not involve any issues of medical ethics and no committee permission is applicable to our study. Finally, we confirm that all methods were carried out in accordance with relevant guidelines

and regulations in our universities or institutes, and no experiments were performed with human subjects in our study, thus we emphasize that no approval and consent forms are required for

our investigation. RESULTS To investigate potential functional SNPs in _APOBEC3_ genes involved in COVID-19 severity, we evaluated the COVID-19 association signals around 7 _APOBEB3_ genes,

comprising _APOBEC3A_, _APOBEC3B_, _APOBEC3C_, _APOBEC3D_, _APOBEC3F_, _APOBEC3G_, and _APOBEC3H_, in the two COVID-19 hospitalization GWASs with European and African ancestries (HGI-B2-EUR

and HGI-B2-AFR, respectively). Around these 7 _APOBEC3_ genes, with an arbitrary association threshold of _P_ < 0.01, we obtained 2 SNPs from HGI-B2-AFR and 4 SNPs from HGI-B2-EUR. Of

these 6 COVID-19 risk SNPs, rs12168809 (_P_ = 0.002; OR = 1.12; 95% CI [1.04–1.2]) and rs76929059 (_P_ = 0.004; OR = 0.82 [0.71–0.94]) are unique to AA; they are located in the intergenic

and promoter region of _APOBEC3A_, respectively. For other 4 SNPs, including rs2076109 (_P_ = 0.008; OR = 1.04 [1.01–1.08]), rs1807558 (_P_ = 0.01; OR = 1.04 [1.01–1.08]), rs2244104 (_P_ = 

0.008; OR = 0.96 [0.93–0.99]), and rs13057307 (_P_ = 0.009; OR = 1.03 [1.01–1.05]), they are unique to EA (Fig. 1A). It is important to point out that these SNPs are only nominally

significant in AA or EA. In conclusion, 2 and 4 prioritized SNPs close to the _APOBEC3_ gene cluster were revealed nominally associated with COVID-19 hospitalization in EUR and AFR samples,

respectively. To study whether these SNPs are specific polymorphisms to populations with unique ancestries, we checked the minor allele frequencies of these 6 SNPs in five major populations

(including African [AFR], Ad Mixed American [AMR], East Asian [EAS], European [EUR], and South Asian [SAS]) as well as in its corresponding subpopulations (Fig. 1B; Table S1). We revealed

that the protective SNP rs76929059 derived from HGI-B2-AFR is only polymorphic in AFR populations [minor allele frequency (MAF) = 0.07]. Another SNP, rs12168809 from HGI-B2-AFR, is a risk

SNP to COVID-19 hospitalization and shows higher MAF in AFR as 0.43, compared to all other 4 major populations (mean MAF = 0.19 ± 0.057). In terms of these 4 SNPs emerged from HGI-B2-EUR,

the risk SNP rs13057307 is less frequent in AFR (MAF = 0.24) than in other 4 major populations (mean MAF = 0.45 ± 0.09); other 3 SNPs, including rs1807558, rs2076109, and rs2244104, display

similar MAFs between AFR (MAFs = 0.28, 0.32, and 0.23, respectively) and other 4 major populations (mean MAFs for the 3 SNPs: 0.30 ± 0.11, 0.38 ± 0.08, and 0.29 ± 0.05, respectively). The

latter 3 SNPs were only emerged in HGI-B2-EUR as candidate SNPs and further evaluation revealed that they were not nominally significant in HGI-B2-AFR; given the relative high MAFs for these

SNPs across all populations (MAFs ≥ 0.28), if they were associated with COVID-19 hospitalization, they might be expected to show association signals close to the association _P_ threshold 

as _P_ < 0.01 in HGI-B2-AFR GWAS, however, we didn’t observe this phenomenon. Therefore, based on this observation, we decided to prioritize only 3 out of 6 SNPs as top candidates, which

are rs76929059, rs12168809, and rs13057307, with the first 2 SNPs are more frequent in AFR populations. Since these 6 SNPs only showing nominal association significance with COVID-19

hospitalization, we wanted to further evaluate their potential involvement in COVID-19 hospitalization based on their potential regulatory roles on _APOBEC3_ gene expression. We manually

collected cis-eQTL results for these 6 SNPs from GTEx and Haploreg4. Our investigation uncovered all these 6 SNPs, except rs1807558, are nominal significant eQTLs (Fig. 1C). Among the 3

prioritized candidate SNPs, including rs76929059, rs12168809, and rs13057307, the last SNP is highly associated with _APOBEC3C_/_D_/_G_ gene expressions across multiple GTEx tissues, with

subcutaneous fat and nerve fibers are two tissues where higher correlations between rs13057307 and _APOBEC3_ expression were observed (Fig. 1C). While rs12168809 (represented by its high LD

SNP rs5757372) is the only eQTL of _APOBEC3A_/C in blood tissue. Another top prioritized SNP rs76929059 (represented by its high LD SNP rs113819742) is an eQTL for multiple _APOBEC3_ genes,

including _APOBEC3A_ (brain caudate basal), _APOBEC3B_ (breast mammary tissue, pancreas, brain caudate basal, esophagus muscularis), _APOBEC3C_ (colon sigmoid), and _APOBEC3D_ (ovary, heart

atrial appenda, artery coronary, skin sun exposed low part, esophagus mucosa). In terms of another 3 SNPs that show close MAFs across different major populations, rs1807558 is not an eQTL

for all _APOBEC3_ genes, rs2076109 is an eQTL for both _APOBEC3B_ (cell-cultured fibroblasts and thyroid) and _APOBEC3F_ (thyroid); rs2244104 is an eQTL for _APOBEC3C_ (muscle skeletal,

cells EBV-transformed lymphocytes, lung, adipose subcutaneous, esophagus muscularis, thyroid, nerve tibial, and artery tibial), _APOBEC3D_ (adipose subcutaneous), and _APOBEC3F_/_G_

(esophagus mucosa). Taken together, our prioritized 3 SNPs, including rs76929059, rs12168809, and rs13057307, are all eQTLs of _APOBEC3_ genes. Additionally, we evaluated _APOBEC3_ genes

expression in 49 normal tissues by ancestry from GTEx and conducted differential expression analysis for these genes between blood tissues derived from COVID-19 patients and healthy

controls. _APOBEC3C_/_G_ were highly expressed in > 20 GTEx tissues (median TPM > 2), while other _APOBEC3_ genes were only moderately expressed in whole blood, spleen, lung, and cells

culture fibroblasts (Fig. 2A). We further performed differential gene expression analysis for these 7 _APOBEC3_ genes among 49 GTEx tissues between European American (EA) and African

American (AA), and the significance threshold after multiple testing adjustment was set at _P_ < 0.00015 = 0.05/(7 × 49). Figure 2B shows the expression profiles of seven _APOBEC3_ genes

among five major tissues, including liver, lung, pancreas, spleen, and whole blood. Only 3 _APOBEC3_ genes, including _APOBEC3F_ (in liver and pancreas), _APOBEC3G_ (in pancreas), and

_APOBEC3H_ (in spleen) display significant differential expression. In whole blood, although all 7 _APOBEC3_ genes demonstrate nominally significantly (_P_ < 0.05) differential gene

expression between EA and AA, no ones were survived after multiple testing. Furthermore, we re-analyzed previously published data to determine the expression of _APOBEC3_ genes upon

SARS-CoV-2 infection in whole blood derived from COVID-19 patients and healthy controls. Cluster analysis and one-way ANOVA analysis of _APOBEC3_ gene expression post SARS-CoV-2 infection

showed _APOBEC3A_, _APOBEC3B_, _APOBEC3G_, and _APOBEC3H_ were significantly upregulated in blood samples from patients with COVID-19 disease compared to healthy controls (Fig. 2C,D). Taken

together, _APOBEC3_ genes show differential expression profiling across multiple tissues, with _APOBEC3C/G_ ubiquitously expressed in more than 20 tissues_,_ and 3 tissues, including lung,

whole blood, and spleen demonstrate similar expression pattern for these _APOBEC3_ genes; _APOBEC3F_/G/H display differential expression levels between EA and AA in specific tissues and

_APOBEC3A_/_B/G/H_ are upregulated upon SARS-CoV-2 infection in whole blood. Given the cumulative studies supporting the involvement of _APOBEC3_ genes25,26, particularly _APOBEC3A_27,28, in

the prevalent C>U coding mutations observed in SARS-CoV-2 genomes, we posited that SARS-CoV-2 genomes sampled from African regions would exhibit a higher percentage of C>U coding

mutations compared to those sampled from Europe or North America, where the majority of the population shares European ancestry. To validate this hypothesis, we analyzed mutational data from

SARS-CoV-2 genomes provided by nextstrain database, comparing the C>U coding mutation percentage per SARS-CoV-2 genome across three geographic regions: Africa, Europe, and North America.

The nextstrain database shared open data encompassed 21,404, 4,968,953, and 3,377,244 samples from these respective regions (Fig. 3A–C). In our analysis, which is consistent with prior

research3, we found that the C>U coding mutation comprised roughly 40% of all coding mutations per SARS-CoV-2 genome. Notably, samples from Africa displayed an increasing percentage of

SARS-CoV-2 genomes containing higher C>U coding mutation rates (≥ 40%; Fig. 3D). Furthermore, upon comparing the percentage of SARS-CoV-2 genomes harboring elevated C>U coding mutation

rates across the three geographic regions, we observed that Africa exhibited a significantly higher proportion of samples with advanced C>U coding mutation rates (40.38%). In contrast,

Europe and North America had lower percentages of SARS-CoV-2 genomes passed the coding mutation rate threshold (C>U mutation percentage ≥ 40%) (Fig. 3D). In summary, through the analysis

of approximately 8.4 million SARS-CoV-2 genomes sampled from Africa, Europe, and North America, we have confirmed a significantly higher percentage of SARS-CoV-2 genomes sampled from Africa

displaying prominent C>U coding mutation rates. DISCUSSION Previous research has strongly suggested the involvement of _APOBEC3_ genes in viral infection through RNA editing, with

APOBEC3A was reported to play a critical role in inducing C>U mutation following SARS-CoV-2 infection in vitro28. Our exploratory analyses suggested rs12168809 and rs76929059 are located

within the intergenic and promoter region of _APOBEC3A_, respectively, and are polymorphic only in AFR populations. These SNPs have also been identified as eQTLs for multiple _APOBEC3_ genes

across variable tissues, which suggests that they may regulate gene expression levels. Therefore, it is hypothesized that the rs12168809 and rs76929059 SNPs may be responsible for the more

frequent emergence of new SARS-CoV-2 variants in AFR populations. The mechanism underlying this association is currently unknown but may involve the regulation of _APOBEC3A_ expression

levels by rs12168809 and rs76929059 SNPs, which would influence the RNA editing capacity of APOBEC3A. In addition, analyses of over 8.4 million SARS-CoV-2 genomes from Africa, Europe, and

North America have enabled the coding mutation rate of C>U to be compared among these continents. Notably, the coding mutation rate of C>U in Africa was found to be significantly

higher (40.38%) compared to Europe (36.88%) and North America (37.91%) with _P_ values less than 1 × 10–4. These observations indicate that the unique genetic makeup in the loci _APOBEC3_

genes and the significantly prominent coding mutation rate of C>U of SARS-CoV-2 genomes in AFR population might be the underlying reason why the new SARS-CoV-2 variants are more

frequently emerging in AFR, since these _APOBEC3_ genes are likely involved in the generation of SARS-CoV-2 mutations. There were cumulative reports supporting the involvement of _APOBEC3_

genes in viral lifecycle. One of the important roles played by APOBEC3 proteins is to direct restrict the virus infection/replication29, as all seven APOBEC3 proteins could bind RNA and

single strand DNA30 to combat retroviruses as well as other pathogenic viruses. APOBEC3A was shown to decrease E2A SUMOylation and interfered with replication of a DNA virus—human adenovirus

by deamination31. Also, another double strand DNA virus—human papillomaviruses was reported to be edited and inhibited by over expression of APOBEC3A in vitro32. Notably, in cells infected

by SARS-CoV-2, the introduction of APOBEC3A through exogenous expression led to UC-to-UU mutations in viral RNA, while the expression of other APOBEC proteins did not show the similar

effect. Moreover, the mutated C bases were frequently observed at the ends of bulge or loop regions in the secondary structure of the viral RNA27. In lines with our finding that whole

blood-derived _APOBEC3A_/_B_/_G_/_H_ were significantly upregulated after SARS-CoV-2 infection, a recent study suggested SARS-CoV-2 adapts and evolves through APOBEC3A/G and APOBEC1-mediated

UC-to-UU mutations in vitro28. In terms of APOBEC3B, it was predominately expressed in nuclear that limited its anti-viral spectrum. By examining the samples from EA and AA, one study found

heterozygous deletions of _APOBEC3B_ had no effect, but homozygous deletions had effect on a direct association with HIV-1 acquisition, progression to AIDS, and viral set points33.

Likewise, APOBEC3B was reported to deaminate both the negative-sense and positive-sense strand of the para-retrovirus Hepatitis B Virus in vitro and in vivo, resulting in a low proportion of

G to A hypermutated viral genome34. Furthermore, the upregulated expression of _APOBEC3B_ induced by folate deficiency was associated with the inhibition of replication of vesicular

stomatitis virus in vitro and in vivo35. Recently, APOBEC3B was shown to combine with Poly (A) binding protein cytoplasmic 1 to stimulate protein kinase R (PKR) and overturned the impaired

activity of PKR that caused by Sendai virus infection, since stimulation of PKR would shutoff cellular translation thus cutoff viral gene expression36. This research hinted APOBEC3B could

affect viral infection via not only editing viral genome but also regulating host innate immunity response. Meanwhile, APOBEC3D is expressed in the cytoplasm, and it can hypermutate the

HIV-1 genome, thereby playing a role in HIV-1 diversification37. APOBEC3G has drawn significant attention for its exceptional intrinsic anti-HIV activity, and it is currently the most

extensively studied protein in the human APOBEC3 family. Most _APOBEC3G_ variants show high population-specificity38. In contrast, APOBEC3F has lower mutagenicity than APOBEC3G and can

induce HIV-1 evolution and drug resistance37,39. In our study, we demonstrated that a SNP of _APOBEC3A_ located in its promoter is only polymorphic in AFR population and also displays

suggestive association with COVID-19 hospitalization. Since APOBEC3A is suggested to the key player to contribute the prevalent C>U mutations in SARS-CoV-2 genomes, our mutational

analysis of ~ 8.4 million SARS-CoV-2 genomes from Africa, Europe, and North America, supports the potential involvement of APOBEC3A in the more variable mutational profiles of SARS-CoV-2

genes in Africa. Further investigation with experiments conducted at population level is warranted to confirm the role of APOBEC3A in the more prevalent mutation rate of SARS-CoV-2 in AFR

population. In our study, we found 3 prioritized SNPs are eQTLs of multiple _APOBEC3_ genes, two of the SNPs are located into regulating area of _APOBEC3A_ and are uniquely polymorphic in

AFR. These _APOBEC3_ genes all show suggestive differential gene expression in blood samples with African ancestry compared to blood samples with European ancestry. We also noted that

_APOBEC3A_ expression tended to be higher in blood samples with African ancestry. Furthermore, _APOBEC3A_/_B_/_G_/_H_ were upregulated upon SARS-CoV-2 infection in blood samples of COVID-19

patients. Finally, we observed that the Africa region-derived SARS-CoV-2 genomes yielded higher C>U coding mutation percentage than that from Europe and North America. Recently, two

publications reported that the C>U mutation in SARS-CoV-2 genome is contributed by APOBEC3A27,28. Thus, it is warrant for further replication of the association of 3 prioritized _APOBEC3_

eQTLs in association with COVID-19 hospitalization and determine it is APOBEC3A but not other APOBEC3 proteins involved in the generation of high transmissible SARS-CoV-2 in AFR

populations. The primary limitation of this study lies in our inability to employ direct experimental methods for assessing whether the C>U coding mutation in SARS-CoV-2 genomes,

attributed to APOBEC3A, occurs at a faster rate in Africans compared to Europeans. To address this question, extensive experimentation involving a substantial number of cell lines, such as

lymphoblastoid cell lines derived from both European and African populations, would be required. While this avenue holds significant promise for future research, the associated costs and

time required for such experiments fall outside the scope of the current study. DATA AVAILABILITY All data generated or analyzed during this study are provided with downloadable links in

this article, and the analysis codes and intermediate data will be available from the corresponding author (Dr. Zhong-Shan Cheng) upon reasonable request. REFERENCES * Davis, H. E. _et al._

Long COVID: major findings, mechanisms and recommendations. _Nat. Rev. Microbiol._ 21(3), 133–146 (2023). Article  PubMed  PubMed Central  CAS  Google Scholar  * Saldivar-Espinoza, B. _et

al._ The mutational landscape of SARS-CoV-2. _Int. J. Mol. Sci._ 24(10), 1 (2023). Article  Google Scholar  * Liu, X. _et al._ Rampant C-to-U deamination accounts for the intrinsically high

mutation rate in SARS-CoV-2 spike gene. _RNA_ 28(7), 917–926 (2022). Article  PubMed  PubMed Central  CAS  Google Scholar  * Chen, Y., Li, M. & Fan, H. The monkeypox outbreak in 2022:

Adaptive evolution associated with APOBEC3 may account for. _Signal Transduct. Target Ther._ 7(1), 323 (2022). Article  PubMed  PubMed Central  Google Scholar  * Mangeat, B. _et al._ Broad

antiretroviral defence by human APOBEC3G through lethal editing of nascent reverse transcripts. _Nature_ 424(6944), 99–103 (2003). Article  ADS  PubMed  CAS  Google Scholar  * Smith, H. C.

RNA binding to APOBEC deaminases; Not simply a substrate for C to U editing. _RNA Biol._ 14(9), 1153–1165 (2017). Article  PubMed  Google Scholar  * Milewska, A. _et al._ APOBEC3-mediated

restriction of RNA virus replication. _Sci. Rep._ 8(1), 5960 (2018). Article  ADS  PubMed  PubMed Central  Google Scholar  * Sadeghpour, S. _et al._ Human APOBEC3 variations and viral

infection. _Viruses_ 13(7), 1 (2021). Article  Google Scholar  * Hultquist, J. F. _et al._ Human and rhesus APOBEC3D, APOBEC3F, APOBEC3G, and APOBEC3H demonstrate a conserved capacity to

restrict Vif-deficient HIV-1. _J. Virol._ 85(21), 11220–11234 (2011). Article  PubMed  PubMed Central  CAS  Google Scholar  * Harris, R. S. & Liddament, M. T. Retroviral restriction by

APOBEC proteins. _Nat. Rev. Immunol._ 4(11), 868–877 (2004). Article  PubMed  CAS  Google Scholar  * Alakunle, E. _et al._ Monkeypox virus in Nigeria: Infection biology, epidemiology, and

evolution. _Viruses_ 12(11), 1 (2020). Article  Google Scholar  * Wolf, J. M. _et al._ Molecular evolution of SARS-CoV-2 from December 2019 to August 2022. _J. Med. Virol._ 95(1), e28366

(2023). Article  PubMed  CAS  Google Scholar  * Chakraborty, C. _et al._ The D614G mutation helps to increase the transmissibility and reduce the virulence of SARS-CoV-2 variants through

natural selection. _Int. J. Surg._ 109(2), 171–174 (2023). Article  PubMed  PubMed Central  Google Scholar  * Petersen, E. _et al._ Emergence of new SARS-CoV-2 Variant of Concern Omicron

(B.1.1.529) - highlights Africa’s research capabilities, but exposes major knowledge gaps, inequities of vaccine distribution, inadequacies in global COVID-19 response and control efforts.

_Int. J. Infect. Dis._ 114, 268–272 (2022). Article  PubMed  CAS  Google Scholar  * Aroldi, A. _et al._ Characterization of SARS-CoV-2 mutational signatures from 1.5+ million raw sequencing

samples. _Viruses_ 15(1), 1 (2022). Article  Google Scholar  * Covid_Host_Genetics_Initiative, Mapping the human genetic architecture of COVID-19. _Nature_ 600(7889), 472–477 (2021). *

Initiative, C.-H.G. A first update on mapping the human genetic architecture of COVID-19. _Nature_ 608(7921), E1–E10 (2022). Article  Google Scholar  * Cunningham, F. _et al._ Ensembl 2022.

_Nucleic Acids Res._ 50(D1), D988–D995 (2022). Article  PubMed  CAS  Google Scholar  * Genomes Project _et al._ A global reference for human genetic variation. _Nature_ 526(7571), 68–74

(2015). Article  ADS  Google Scholar  * Ward, L. D., & Kellis, M. HaploReg: a resource for exploring chromatin states, conservation, and regulatory motif alterations within sets of

genetically linked variants. _Nucleic Acids Res_. 40(Database issue), D930–D934 (2012). * GTEx-Consortium, The GTEx Consortium atlas of genetic regulatory effects across human tissues.

_Science_ 369(6509), 1318–1330 (2020). * Gay, N. R. _et al._ Impact of admixture and ancestry on eQTL analysis and GWAS colocalization in GTEx. _Genome Biol._ 21(1), 233 (2020). Article 

PubMed  PubMed Central  CAS  Google Scholar  * Thair, S. A. _et al._ Transcriptomic similarities and differences in host response between SARS-CoV-2 and other viral infections. _iScience_

24(1), 101947 (2021). Article  ADS  PubMed  CAS  Google Scholar  * Zhang, W. _et al._ COVID19db: a comprehensive database platform to discover potential drugs and targets of COVID-19 at

whole transcriptomic scale. _Nucleic Acids Res._ 50(D1), D747–D757 (2022). Article  MathSciNet  PubMed  CAS  Google Scholar  * Cervantes-Gracia, K. _et al._ APOBECs orchestrate genomic and

epigenomic editing across health and disease. _Trends Genet._ 37(11), 1028–1043 (2021). Article  PubMed  CAS  Google Scholar  * Li, Y. _et al._ C-to-U RNA deamination is the driving force

accelerating SARS-CoV-2 evolution. _Life Sci. Alliance_ 6(1), 1 (2023). Article  CAS  Google Scholar  * Nakata, Y. _et al._ Cellular APOBEC3A deaminase drives mutations in the SARS-CoV-2

genome. _Nucleic Acids Res._ 51(2), 783–795 (2023). Article  PubMed  PubMed Central  CAS  Google Scholar  * Kim, K. _et al._ The roles of APOBEC-mediated RNA editing in SARS-CoV-2 mutations,

replication and fitness. _Sci. Rep._ 12(1), 14972 (2022). Article  ADS  MathSciNet  PubMed  PubMed Central  CAS  Google Scholar  * Harris, R. S. & Dudley, J. P. APOBECs and virus

restriction. _Virology_ 479–480, 131–145 (2015). Article  PubMed  Google Scholar  * Salter, J. D., Bennett, R. P. & Smith, H. C. The APOBEC protein family: United by structure, divergent

in function. _Trends Biochem. Sci._ 41(7), 578–594 (2016). Article  PubMed  PubMed Central  CAS  Google Scholar  * Göttig, L., _et al._ Apobec3A deamination functions are involved in

antagonizing efficient human adenovirus replication and gene expression. _mBio_ 14(3), e0347822 (2023). * Warren, C. J. _et al._ APOBEC3A functions as a restriction factor of human

papillomavirus. _J. Virol._ 89(1), 688–702 (2015). Article  PubMed  Google Scholar  * An, P. _et al._ APOBEC3B deletion and risk of HIV-1 acquisition. _J. Infect. Dis._ 200(7), 1054–1058

(2009). Article  PubMed  CAS  Google Scholar  * Suspène, R. _et al._ Extensive editing of both hepatitis B virus DNA strands by APOBEC3 cytidine deaminases in vitro and in vivo. _Proc. Natl.

Acad. Sci. USA_ 102(23), 8321–8326 (2005). Article  ADS  PubMed  PubMed Central  Google Scholar  * Wu, J. _et al._ FOLR1-induced folate deficiency reduces viral replication via modulating

APOBEC3 family expression. _Virol. Sin._ 38(3), 409–418 (2023). Article  PubMed  PubMed Central  CAS  Google Scholar  * Manjunath, L. _et al._ APOBEC3B drives PKR-mediated translation

shutdown and protects stress granules in response to viral infection. _Nat. Commun._ 14(1), 820 (2023). Article  ADS  PubMed  PubMed Central  CAS  Google Scholar  * Sato, K. _et al._

APOBEC3D and APOBEC3F potently promote HIV-1 diversification and evolution in humanized mouse model. _PLoS Pathog._ 10(10), e1004453 (2014). Article  PubMed  PubMed Central  Google Scholar 

* Singh, K. K. _et al._ Genetic variants in the host restriction factor APOBEC3G are associated with HIV-1-related disease progression and central nervous system impairment in children. _J.

Acquir. Immune Defic. Syndr._ 62(2), 197–203 (2013). Article  PubMed  PubMed Central  CAS  Google Scholar  * Armitage, A. E. _et al._ Possible footprints of APOBEC3F and/or other APOBEC3

deaminases, but not APOBEC3G, on HIV-1 from patients with acute/early and chronic infections. _J. Virol._ 88(21), 12882–12894 (2014). Article  PubMed  PubMed Central  Google Scholar 

Download references ACKNOWLEDGEMENTS We need to thank all researchers who involved in the HGI and nextstrain project. Since without their contributions in sharing the two COVID-19

hospitalization GWASs and SARS-CoV-2 genomes to public, we might be unable to have current discovery. FUNDING This project was funded by Science and Technology Supporting Plan ([2020]4Y163)

and Scientific and Technological Innovation Talent Team (CXTD[2022]004) of Scientific and Technological Department of Guizhou Province, China. AUTHOR INFORMATION Author notes * These authors

contributed equally: Ke Zhang and Fang Chen. AUTHORS AND AFFILIATIONS * The Key and Characteristic Laboratory of Modern Pathogenicity Biology, School of Basic Medical Sciences, Guizhou

Medical University, Guiyang, 561113, China Ke Zhang, Fang Chen, Hu-Yan Shen & Ping-Ping Zhang * The Department of Emergency ICU, The Affiliated Hospital of Guizhou Medical University,

Guiyang, 550004, China Han Gao, Hong Peng & Yu-Si Luo * The Department of Emergency, Liupanshui Hospital of The Affiliated Hospital of Guizhou Medical University, Liupanshui, 553000,

China Yu-Si Luo * Center for Applied Bioinformatics, St. Jude Children’s Research Hospital, 262 Danny Thomas Hospital, MS1122, Memphis, TN, 38105, USA Zhong-Shan Cheng Authors * Ke Zhang

View author publications You can also search for this author inPubMed Google Scholar * Fang Chen View author publications You can also search for this author inPubMed Google Scholar * Hu-Yan

Shen View author publications You can also search for this author inPubMed Google Scholar * Ping-Ping Zhang View author publications You can also search for this author inPubMed Google

Scholar * Han Gao View author publications You can also search for this author inPubMed Google Scholar * Hong Peng View author publications You can also search for this author inPubMed 

Google Scholar * Yu-Si Luo View author publications You can also search for this author inPubMed Google Scholar * Zhong-Shan Cheng View author publications You can also search for this

author inPubMed Google Scholar CONTRIBUTIONS Z-S.C., Y-S.L., and K.Z. conceived the study. Y-S.L. provided financial support to the study. Z-S.C., F.C., H-Y.S., P-P.Z., H.G., and H.P.

collected the raw data from databases. Z-S.C., K.Z., Y-S.L., and F.C. analyzed the data, generated the figures, drafted and revised the manuscript. All authors reviewed and approved the

manuscript. CORRESPONDING AUTHORS Correspondence to Yu-Si Luo or Zhong-Shan Cheng. ETHICS DECLARATIONS COMPETING INTERESTS The authors declare no competing interests. ADDITIONAL INFORMATION

PUBLISHER'S NOTE Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. SUPPLEMENTARY INFORMATION SUPPLEMENTARY TABLE 1.

RIGHTS AND PERMISSIONS OPEN ACCESS This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and

reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes

were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material.

If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to

obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Reprints and permissions ABOUT THIS ARTICLE CITE THIS

ARTICLE Zhang, K., Chen, F., Shen, HY. _et al._ Regulatory variants of _APOBEC3_ genes potentially associate with COVID-19 severity in populations with African ancestry. _Sci Rep_ 13, 22435

(2023). https://doi.org/10.1038/s41598-023-49791-x Download citation * Received: 14 July 2023 * Accepted: 12 December 2023 * Published: 17 December 2023 * DOI:

https://doi.org/10.1038/s41598-023-49791-x SHARE THIS ARTICLE Anyone you share the following link with will be able to read this content: Get shareable link Sorry, a shareable link is not

currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative