MtDNA variations at C5178a and A249d decrease the risk of severe COVID-19 in a Han Chinese population from Central China


 Background: Mitochondria have been shown to play vital roles during SARS-CoV-2 infection and COVID-19 development. Currently, whether mitochondrial DNA (mtDNA) variations, which define mtDNA haplogroups and determine OXPHOS performance and ROS production, are associated with COVID-19 risk is unclear. Methods: A population-based case-control study was conducted to compare the distribution of mtDNA variations defining mtDNA haplogroups between healthy controls (n = 615) and COVID-19 patients (n = 536). COVID-19 patients were diagnosed based on molecular diagnostics of the viral genome by qPCR and chest X-ray or computed tomography (CT) scanning. The exclusion criteria for the healthy controls were any history of diseases in the one-month preceding study assessment. MtDNA variations defining mtDNA haplogroups were identified by PCR-RFLPs and HVS-I sequencing and determined based upon mtDNA phylogenetic analysis using Mitomap Phylogeny. Student’s t-test was used for continuous variables, and Pearson’s chi-squared test or Fisher’s exact test was used for categorical variables. To assess the independent effect of each mtDNA variation defining mtDNA haplogroups, multivariate logistic regression analyses were performed to calculate the adjusted odds ratios (ORs) and 95% confidence intervals (CIs) with adjustments for the possible confounding factors of age, sex, smoking and diseases (including cardiopulmonary diseases, diabetes, obesity and hypertension) determined through clinical and radiographic examinations. Results: Multivariate logistic regression analyses revealed that mtDNA variations at C5178a and A249d were associated with a reduced risk of severe COVID-19 (OR = 0.590, 95% CI = 0.428-0.814, p = 0.001; and OR = 0.654, 95% CI = 0.457-0.936, p = 0.020, respectively), while A4833G, A4715G, T3394C and G5417A/C16257a/C16261T were related to an increased risk of severe COVID-19 (OR = 2.336, 95% CI = 1.179-4.608, p = 0.015; OR = 2.033, 95% CI = 1.242-3.322, p = 0.005; OR = 3.040, 95% CI = 1.522-6.061, p = 0.002; and OR = 2.890, 95% CI = 1.199-6.993, p = 0.018, respectively). Conclusion: mtDNA variations C5178a and A249d might contribute to an individual’s resistance to developing severe COVID-19, whereas A4833G, A4715G, T3394C and G5417A/C16257a/C16261T might increase an individual’s risk of developing severe COVID-19. Trial registration: no.


Introduction
The COVID-19 pandemic caused by SARS-CoV-2 has resulted in a worldwide crisis of formidable morbidity and mortality. The epidemiology, diagnosis, risk factors and treatments of COVID-19 have been explored intensively since the outbreak in Wuhan (Hubei Province, China) in December 2019. Many studies show that the clinical features of COVID-19 range from an asymptomatic state to acute respiratory distress syndrome (ARDS) and multiorgan dysfunction. Most COVID-19 patients develop a respiratory tract infection with common symptoms of cough, fever and shortness of breath. Other reported symptoms are weakness, malaise, respiratory distress, muscle pain, sore throat and loss of taste and/or smell. A number of patients develop severe fatal consequences resulting from a surge of in ammatory events (also known as cytokine storms) [1,2]. These clinical characteristics hint that although SARS-CoV-2 infection is the causative factor of COVID-19, not all of those exposed to SARS-CoV-2 will develop COVID-19, especially severe COVID-19, strongly suggesting that gene-environment interactions exist in COVID-19 progression, and an individual's hereditary susceptibility and innate capacities of antioxidant and immune responses to SARS-CoV-2 might contribute to this process. Currently, advanced age; male sex; blood group A; comorbidities of cardiopulmonary diseases, diabetes, obesity and hypertension [3]; a genomic segment of ~50 kb inherited from Neanderthals and carried by ~50% of people in South Asia and ~16% of people in Europe today [4]; and a 3p21. 31 gene cluster have been reported to be related to an individual's susceptibility to severe COVID-19 [5]. Whether there are other speci c molecular markers to predict the risk of COVID-19 remains unclear.
of innate and adaptive immune responses [8] and activation, development, maintenance and survival of the speci c phenotypes of immune cells [9]. These ndings provide clues that mitochondria might be associated with an individual's susceptibility to  As hubs of cellular oxidative homeostasis, mitochondria generate approximately 85% of intracellular reactive oxygen species (ROS) when they produce usable energy through oxidative phosphorylation (OXPHOS). In contrast to other cellular organelles, mitochondria have their own DNA (mtDNA). Interestingly, the common and "nonpathological" mtDNA variations de ning mtDNA haplogroups determine OXPHOS performance and ROS production in humans and mice [10]. Additionally, these mtDNA variations exert considerable in uence on longevity [11], help human beings adapt to different environments [12,13] and are associated with susceptibility to many human diseases in conditions where ROS generated by mitochondria are supposed to play a part [14][15][16][17]. Thus, we hypothesized that certain mtDNA variations de ning mtDNA haplogroups might be related to an individual's susceptibility to COVID-19. To test this hypothesis, a population-based case-control study was performed to compare the distribution of mtDNA variations de ning mtDNA haplogroups between COVID-19 patients and healthy controls in a Han Chinese population from Central China.

Study population
The study was approved by the Ethics Committee of Hubei University of Medicine and Wuhan University (Hubei Province, China). COVID-19 patients (n = 536) were recruited from Taihe Hospital (the First A liated Hospital of Hubei University of Medicine, Shiyan, Hubei Province, China) and People's Hospital of Hubei Province (the A liated Hospital of Wuhan University, Wuhan, Hubei Province, China) from February 2020 to March 2020. COVID-19 patients were diagnosed based on molecular diagnostics of the viral genome by qPCR and chest X-ray or computed tomography (CT) scanning. Age-and sex-matched volunteers (n = 615) were recruited from healthy individuals who underwent physical examinations at the two hospitals. The exclusion criteria for the healthy controls were any history of diseases in the one-month preceding study assessment. All the subjects were unrelated for at least three generations. After explaining the purpose and procedures of the study, all the participants signed a written informed consent form and completed a detailed questionnaire on smoking habits. Three milliliters of peripheral blood from each subject were drawn into Na-EDTA tubes. After incubation at 55°C for 30 min to inactivate the potential SARS-CoV-2, the blood samples were stored at -80°C prior to genomic DNA extraction.
Detection of mtDNA variations de ning mtDNA haplogroups Genomic DNA was extracted from peripheral blood using an Ezup Column Blood Genomic DNA Puri cation Kit (Lot#: B518253-0100, Sangon Biotech Co., Ltd, Shanghai, China). MtDNA variations de ning mtDNA haplogroups were identi ed as described [13,14,16,17]. Brie y, after the entire mtDNA was ampli ed into 22 overlapping PCR fragments, the PCR fragments were digested with different restriction endonucleases (PCR-restriction fragment length polymorphism, PCR-RFLP) and replenished by sequencing hypervariable segment I (HVS-I). Two × Taq Plus Master Mix II was used for PCR-RFLP and HVS-I sequencing (Lot#: P213-01, Vazyme Co., Ltd, Nanjing, China). The restriction endonucleases Alu I, Ava II, Bam HI, Bst NI, Dde I, Hae II, Hae III, Hha I, Hinc II, and Hinf I were used in the study (Takara Co., Ltd, Dalian, China). The primers for PCR-RFLPs and HVS-I sequencing were synthesized by Sangon Biotech Co., Ltd. (Shanghai, China), and the primer sequences are presented in Supplementary Table 1. MtDNA variations de ning mtDNA haplogroups were determined based upon mtDNA phylogenetic analysis using Mitomap Phylogeny [12]. The Human Genome Variation Society (HGVS) validation of mtDNA variations is presented in Supplementary Table 2.

Data analysis
Student's t-test was used for continuous variables, and Pearson's chi-squared test or Fisher's exact test was used for categorical variables. For multiple comparisons of mtDNA variations de ning mtDNA haplogroups, Bonferroni correction was applied (the required signi cance level = 0.05/number of comparisons). To assess the independent effect of each mtDNA variation de ning mtDNA haplogroups, multivariate logistic regression analyses were performed to calculate the adjusted odds ratios (ORs) and 95% con dence intervals (CIs) with adjustments for the possible confounding factors of age, sex, smoking and diseases (including cardiopulmonary diseases, diabetes, obesity and hypertension) determined through clinical and radiographic examinations. All statistical analyses were performed using SPSS Statistics 25 for Mac (SPSS Inc., Chicago, IL, USA).

Distribution of mtDNA variations de ning mtDNA haplogroups in controls and severe cases
Pearson's chi-squared test or Fisher's exact test demonstrated that mtDNA variations A4833G, A4715G, T3394C and G5417A/C16257a/C16261T were signi cantly higher (p = 0.003, 0.001, 0.010, and 0.023, respectively), while mtDNA variations C5178a and A249d were signi cantly lower in severe COVID-19 patients than in controls (p = 0.000 and 0.008, respectively). When Bonferroni correction was applied, mtDNA variations A4833G, A4715G and C5178a reached the required p value of < 0.0033. Multivariate logistic regression analyses with adjustments for age, sex, smoking and diseases (including cardiopulmonary diseases, diabetes, obesity and hypertension) showed that the C5178a and A249d mtDNA variations were associated with a reduced risk of severe COVID-19 (OR = 0.590, 95% CI = 0.428-0.814, p = 0.001; and OR = 0.654, 95% CI = 0.457-0.936, p = 0.020), while the A4833G, A4715G, T3394C and G5417A/C16257a/C16261T mtDNA variations were related to an increased risk of severe COVID-19 (OR = 2.336, 95% CI = 1.179-4.608, p = 0.015; OR = 2.033, 95% CI = 1.242-3.322, p = 0.005; OR = 3.040, 95% CI = 1.522-6.061, p = 0.002; and OR = 2.890, 95% CI = 1.199-6.993, p = 0.018, respectively) ( Table 4 and Figure 1). As hubs of cellular oxidative homeostasis, mitochondria are not only interrelated and play a role in the oxidative stress and in ammation in SARS-CoV-2 infection and COVID-19 development [6,7,[18][19][20][21] but are also indispensable regulators of the innate and adaptive immune responses in the process of SARS-CoV-2 infection and COVID-19 development [8,9,[21][22][23]. Moreover, mitochondrial residency of SARS-CoV-2 with a stronger signal compared to its coronavirus relatives implied further that mitochondria are the major cellular organelle affected by oxidative stress and in ammation resulting from SARS-CoV-2 infection [24]. In comparison with nuclear DNA (nDNA), mtDNA is particularly susceptible to oxidative damage due to its direct exposure to ROS, limited DNA repair capacity and absence of protection by histones. The decline in mitochondrial function with aging might explain the phenomena of high mortality in elderly COVID-19 patients to a certain extent [1,11,15]. Thus, when SARS-CoV-2 infects cells, the common and "nonpathological" mtDNA variants, which de ne mtDNA haplogroups and determine OXPHOS performance and ROS production, contribute to an individual's capacity for antioxidant and immune responses to protect cells from SARS-CoV-2 infection and COVID-19 development or aggravate the process. Consistent with this, mtDNA variation C5178a (L237M, Leu→Met substitution) in ND2, de ning mtDNA haplogroup D and proposed to be an e cient oxidant scavenger [25], was signi cantly lower both in the pooled COVID-19 patients and severe COVID-19 patients compared to controls in the study. The protective effect of the C5178a mtDNA variation has been reported to be propitious to human longevity [11], bene cial for diabetic patients against atherosclerotic and myocardial infarction [26], and to decrease an individual's risk of developing acute mountain sickness (AMS), lung cancer, chronic obstructive pulmonary disease (COPD) and other diseases [14][15][16]26]. Therefore, the protective effect of the C5178a mtDNA variation against oxidative damage as an e cient oxidant scavenger might protect cells from oxidative destruction caused by SARS-CoV-2 infection and decrease an individual's risk of developing COVID-19, especially severe COVID-19.
As expected, the combined subhaplogroups F1 and F3 (haplogroup F) were shown to be associated with a decreased risk of COVID-19 and severe COVID-19. In Asian populations, haplogroup F functions as a positive factor for a long-life span [28], confers bene cial effects to resist metabolic syndrome (MetS) [29], and improves the physical performance of athletes [30].
Because only subhaplogroups F1 and F3 were detected and mtDNA variations at T6392C and G10310A were synonymous, we deduced that variation at A249d in the control region (HVS-II) of mtDNA might account for its protective effect against COVID-19.
MtDNA variations A4833G, A4715G, T3394C and G5417A/C16257a/C16261T were found to increase an individual's risk of developing severe COVID-19 in the study. Of note, A4715G, T3394C and G5417A/C16257a/C16261T are reported to be associated with an increased risk of type II diabetes mellitus (T2DM) in the Chinese population [31]. A4715G is a risk factor for moderate and severe nonalcoholic fatty liver disease (NAFLD) [32]. The mtDNA variation T3394C is bene cial for native Tibetans to adapt to hypoxic environments because it has higher complex I activity [13] [13], which might help native Tibetans adapt to hypoxic environments. Similarly, A4833G constitutes a risk factor for lung cancer [16], COPD [17] and recurrent oral ulceration (ROU) in plain areas [37]. Therefore, T3394C and A4833G might be risk factors for severe COVID-19 through the same mechanism as in the other human diseases occurring in the plain areas.
MtDNA variation G5417A (speci c for mtDNA haplogroup N9) confers a higher risk of MetS development in HIV-infected patients [38]. In the Chinese population, the mtDNA variation G5417A/C16257a/C16261T is a risk factor for diabetic nephropathy because of having more ROS and fragmented mitochondria [31], which might account for it being a risk factor for both moderate and severe COVID-19 in this Han Chinese population.
Of note, a mtDNA deletion of approximately 800 bp was detected during PCR-RFLP analysis in the study. In our previous studies, an 822-bp mtDNA deletion was identi ed and demonstrated to be positively associated with cigarette smoking and mtDNA haplogroups [16,17]. Because the blood samples of the study were incubated at 55°C for 30 min to inactivate the potential SARS-CoV-2, the mtDNA deletion was not further analyzed to avoid the possible disturbance of mtDNA breakage caused by incubation at higher temperature.
This study has some limitations. First, during the extreme clinical circumstances of the pandemic, especially at the beginning of the epidemic, we were unable to collect detailed clinical data (for example, levels of in ammatory cytokines and immune factors and disease outcome) in a very short period of time, which will be important to investigate in follow-up studies. Second, because we only analyzed mtDNA variations in a Han Chinese population from Central China, large-scale studies are expected in other populations.

Conclusions
Our ndings revealed that mtDNA variations C5178a and A249d contribute to an individual's resistance to COVID-19 development, whereas variations A4833G, A4715G, T3394C, and G5417A/C16257a/C16261T might be risk factors for this process, providing evidence that gene-environment interactions do exist in COVID-19 progression.

Declarations Ethics approval and consent to participate
The study was approved by the Ethics Committee of Hubei University of Medicine and Wuhan University (Hubei Province, China). All the participants signed a written informed consent form to participate the project.

Consent for publication
All the authors have reviewed the nal version of manuscript and signed a written consent form for publication.

Availability of data and materials
The data that support the ndings of this study are available from the corresponding author upon reasonable request.

Competing interests
The authors declare no con ict of interest.  c , the letter "d" indicates nucleotide deletion, the letter "a" indicates nucleotide transversion. c , the letter "d" indicates nucleotide deletion, the letter "a" indicates nucleotide transversion.