Abstract
Background The viral entry of SARS-CoV-2 requires host-expressed TMPRSS2 to facilitate the viral spike (S) protein priming.
Objectives To test the hypothesis that Mg treatment leads to DNA methylation changes in TMPRSS2.
Methods This study is nested within the Personalized Prevention of Colorectal Cancer Trial (PPCCT), a double-blind 2×2 factorial randomized controlled trial, which enrolled 250 participants from Vanderbilt University Medical Center. Target doses for both Mg and placebo arms were personalized.
Results We found that 12-week of personalized Mg treatment significantly increased 5-mC methylation at cg16371860 (TSS1500, promoter) by 7.2% compared to placebo arm (decreased by 0.1%) in those aged < 65 years old. The difference remained statistically significant after adjusting for age, sex and baseline methylation as well as FDR correction (FDR-adjusted P =0.014). Additionally, Mg treatment significantly reduced 5-hmC level at cg26337277 (close proximity to TSS200 and 5’UTR, promoter) by 2.3% compared to increases by 7.1% in the placebo arm after adjusting for covariates in those aged < 65 years old (P=0.003). The effect remained significant at FDR of 0.10 (adjusted P value=0.088).
Conclusion Among individuals aged younger than 65 years with the Ca:Mg intake ratios equal to or over 2.6, reducing Ca:Mg ratios to around 2.3 increased 5-mC modifications (i.e. cg16371860) and reduced 5-hmC modifications (i.e. cg26337277) in the TMPRSS2 gene. These findings, if confirmed, provide another mechanism for the role of Mg intervention for the prevention of COVID-19 and treatment of early and mild disease by modifying the phenotype of the TMPRSS2 genotype.
INTRODUCTION
As of December 4th 2020, in the US alone, coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), led to 14.2 million confirmed cases and 276,000 deaths. Up to 20% of symptomatic individuals will progress to severe or critical illness ranging from hospitalization to death while some with mild symptoms may also experience a variably prolonged period of recovery with long-term complications [1]. Current antiviral pharmaceutical therapeutics, such as remdesivir, targeting hospitalized patients with COVID-19 have not achieved statistically significant benefits on mortality in randomized trials [2–5]. While awaiting global vaccination for SARS-CoV-2 to end the COVID-19 pandemic, and confirmation that vaccination provides complete protection among adults and for multiple SARS-CoV-2 variants [6–8], the National Institute of Health (NIH) highlights an urgent need for interventions which can be administered early during the course of infection to prevent disease progression to severe COVID-19, speed recovery, and prevent long-term complications [1].
In a precision-based randomized trial [9], we reported that an improved magnesium (Mg) status led to an optimal level of vitamin D. A recent trial further confirmed this finding that Mg treatment improved vitamin D status [10]. We also previously found high circulating levels of vitamin D were prospectively related to reduced risk of cardiovascular (CVD) mortality only when Mg intake was adequate [11]. Based on our findings, a recent study conducted by Tan et al. found a combinatorial treatment of Mg and vitamin D3 within the Recommended Dietary Allowance (RDA) plus vitamin B12 reduced the risk of severe illness from COVID-19 by 80% [12]. Although cross-sectional studies investigating the correlation between vitamin D levels and COVID-19 severity and mortality [13] were inconclusive and a large-scale prospective cohort study in the UKBiobank found that serum 25(OH)D (25-hydroxyvitamin D) was not related to risk of severe COVID-19 after adjusting for confounding factors[14], a recent open-label trial using calcifediol (i.e. 25(OH)D) treatment starting at 64 times the RDA levels[15] substantially reduced the need for intensive care unit (ICU) care among hospitalized COVID-19 patients [16].
In addition to optimizing 25(OH)D levels, Mg also activates the conversion from 25(OH)D to 1,25(OH)2D, the active form of vitamin D [17]. Patients with “Mg-dependent vitamin-D-resistant rickets” [18] caused by Mg deficiency were resistant to vitamin D treatment alone with no change in blood measures of 1,25(OH)2D at doses up to 600,000 IU [17]. However, they dramatically responded to Mg treatment, particularly treatment of Mg plus vitamin D [17,19]. Collectively these varied findings support the hypothesis that the synergistic interaction between Mg and vitamin D may substantially reduce the doses required for vitamin D to reduce the severity of COVID-19. Since 79% of US adults do not meet the RDA of Mg [20], thus, using the combinatorial prophylaxis strategy of vitamin D plus Mg could be a crucial strategy for treating COVID-19.
The viral entry of SARS-CoV-2 depends on binding of the viral spike (S) proteins to cellular receptors (i.e. angiotensin-converting enzyme (ACE2)) which requires host-expressed transmembrane serine protease 2 (TMPRSS2) to facilitate S protein priming [21]. In vitro studies showed that TMPRSS2 inhibitor partially blocked SARS-CoV-2 from entering into lung epithelial cells [21]. Animal studies found that TMPRSS2 knockout mice infected with H1N1 influenza had minimal infection and largely attenuated disease severity[22].
Supporting the findings in mice, a recent cohort found that TMPRSS2 expression levels and genetic variants played an essential role in modulating COVID-19 severity [23]. The effect is also likely through regulating coagulation cascade and arterial thrombosis [24–26], one key factor for COVID-19 severity and mortality [27,28]. TMPRSS2 expression is influenced by both genetic variants and epigenetic changes such as DNA methylation. 5-methylcytosine (5-mC), namely the methylation of the fifth carbon of cytosines, was shown to be associated with transcription repression [29,30] whereas 5-hydroxymethylcytosine (5-hmC) is specifically enriched in tissue-specific enhancers [31] and critical in maintaining active gene expression [32,33]. DNA methylation changes are inducible by environmental exposures including nutrients[34,35]. In addition to optimizing vitamin D synthesis and metabolism, Mg also affects the metabolism of α-ketoglutarate [36,37], one key factor for the ten-eleven translocation (TET) enzymes [38], which catalyze the oxidation of 5-mC to 5-hmC in an active demethylation pathway [39]. We, therefore, aim to test the hypothesis that Mg treatment leads to DNA methylation changes (i.e. 5-mC and 5-hmC) in TMPRSS2 in the Personalized Prevention of Colorectal Cancer Trial (PPCCT), a precision-based magnesium supplementation trial.
MATERIALS and METHODS
Participants, randomization and blinding
This is an ancillary study nested in the parent study, the “Personalized Prevention of Colorectal Cancer Trial” (PPCCT; NCT01105169 at ClinicalTrials.gov). The PPCCT is a double-blind 2×2 factorial randomized controlled trial conducted at Vanderbilt University Medical Center, Nashville, TN. The detailed study design has been reported previously [9,40]. In brief, participants aged 40 to 85 years old were recruited including: 1) 236 individuals with adenomas or hyperplastic polyps diagnosed from 1998 to 2014 or 2) 14 polyp-free individuals with a high risk of colorectal cancer. Dietary and total intakes of Ca, Mg, and the Ca: Mg ratios were derived from six 24-hr dietary recalls over the course of the trial including intakes of Ca and Mg supplements. Eligible participants were those who had a Ca intake ≥ 700 mg/day and < 2000 mg/day and in whom the Ca:Mg intake ratio was ≥ 2.6 based on the average of the first two baseline 24-hr dietary recalls. Eligible participants were randomized to Mg glycinate or placebo (microcrystalline cellulose) capsules. The Mg treatment used a personalized dose of Mg supplementation that would reduce the Ca:Mg intake ratio to around 2.3, suggested by previous studies [9,41–43]. Participants, study investigators and staff were blinded to the assigned interventions. Blood samples were collected and processed at each clinic visit. Anthropometric measurements (weight, height, waist and hip circumference) were measured at least twice at each clinic visit.
265 participants were randomized and allocated to the Mg treatment arm or the placebo arm. 15 participants withdrew consent before taking Mg treatment or placebo. Of these, 250 randomized participants started treatment, and 239 completed the trial with 11 participants finishing part of the study before withdrawing [9]. Six of the withdrawals were due to self-reported adverse events (four withdrawals in the treatment arm and two in the placebo arm). One of them had donated blood at baseline and at the end of the trial. Therefore, in the current study, 240 participants were included who had blood collected at baseline and at the end of the trial.
Measurement of 5-mC and 5-hmC at single resolution for the TMPRSS2 gene
All 240 participants who were enrolled in the PPCCT and had blood DNA samples available at the baseline and the end of the trial were included in the current study to examine the effect of personalized Mg treatment on methylation modifications in the TMPRSS2 gene. In order to minimize between-batch variations, samples were randomly organized in treatment-placebo (i.e. one treatment arm with one placebo arm) sets (4 samples in each set: 2 from pre-, and 2 from post-treatment). Lab staff were blinded to sample status (i.e. treatment vs. placebo and pre-treatment vs. post-treatment status).
Genomic DNA was extracted from buffy coat fractions collected using a QIAamp DNA mini-kit (Qiagen Inc, Germantown, MD) according to the manufacturer’s protocol [44]. DNA quality was examined using standard molecular biology protocols. We used the TET-assisted bisulfite (TAB)-Array, which combines TET-assisted bisulfite conversion with the Infinium Methylation EPIC array (EPIC array) that interrogates ∼850,000 CpG or non-CpG methylation sites to differentiate 5-hmC and 5-mC signals at base resolution [45]. Our detailed approach was reported previously [45–47] Using this technique, we were able to differentiate 5-mC from 5-hmC.
The β-values for 5-mC and 5-hmC were estimated using the Maximum Likelihood Estimate from the paired bisulfite conversion and TAB-treated samples 39. The following quality control steps were taken: 1) We excluded low-quality probes where the number of beads < 3 or the detection P-value > 0.05 [48]; 2) Exclusion of CpGs with a detection rate <95% and samples with the percentage of low-quality methylation measurements >5% or extremely low signal of BS probes [48]; 3) Exclusion of extreme outliers, as defined by the Tukey’s method [49], based on average total signal value across CpG probes; 4) Remaining samples were preprocessed using the R package ENmix to improve accuracy and reproducibility [48]; 5) Dye bias was corrected using regression on a logarithm of internal control probes [50]; 6) Quantile-normalization of signal for Infinium I or II probes; and 7) Lastly, extreme β value (i.e., the proportion of methylated signal in total signals from 0 to 1) outliers across samples, defined by Tukey’s method, were set as missing. In total, the TAB-Array data for 224 participants out of 240 passed the seven quality control steps. In the current study, we kept all 32 CpG sites related to the TMPRSS2 gene selected by the EPIC array for 224 participants whose methylation data passed the quality control steps.
Statistical analyses
Continuous demographic variables (mean ± standard deviation) and categorical demographic variables (percent) were compared between treatment and placebo arms. The Wilcoxon rank sum test was conducted to evaluate continuous variables, while Pearson chi-squared tests were conducted to compare categorical variables. Linear regression models were fitted to examine the effect of Mg treatment on overall changes of cytosine modifications (5-mC and 5-hmC) in the TMPRSS2 gene in three models, respectively. Model 1: crude value; model 2: sex and age; model 3: sex, age and baseline methylation. We also conducted stratified analyses by age because elderly adults are at increased risk for severe illness and mortality from COVID-19 [51]. All P values are two sided and statistical significance was determined using an alpha level of 0.05. The data analyses used R software (version 3.5.1).
RESULTS
The baseline demographic characteristics of 240 participants are presented in Table 1. The treatment arm was not significantly different from the placebo arm for baseline demographic variables, including age, sex, smoking status, alcohol drinking status, physical activity status, educational achievement, race, daily intake of total energy, total Ca, intake ratio of Ca to Mg and factors related to cardiovascular events, including body mass index (BMI), blood pressure and eGFR (Table 1).
The mean daily dose of personalized Mg supplementation was 205.59 mg with a range from 77.25 mg to 389.55 mg. Compliance with the pill regimen was very high for both the placebo and treatment arms (mean (standard deviation) based on pill counts were 96.1% (8.3) and 95.9% (10.2), respectively, and P=0.37 for difference between the arms). The mean Ca:Mg ratios (standard deviations) for the treatment and placebo arms after administering Mg and placebo supplementation were 2.27 (0.13) and 3.87 (1.46) respectively (P for difference, <0.001), based on the two 24-hour dietary recalls performed at baseline and remained stable at 2.13 (0.68) and 3.51 (1.32), respectively (P for difference, <0.001) based on the four 24-hour dietary recalls conducted over the 12-week period of the trial.
Shown in Supplemental Table S1 and Table S2 are the effects of personalized Mg treatment on cytosine modifications 5-mC and 5-hmC, respectively, in the TMPRSS2 for 32 CpG sites. We found, compared to the placebo arm, Mg treatment significantly increased 5-mC level at cg16371860 CpG site (P=0.011), even after adjusting for age and sex (P=0.010), and further adjusting for baseline 5-mC levels (P=0.019) (Table 2). However, the significance disappeared after false discovery rate (FDR) correction at 0.10 based on Benjamini–Hochberg approach. In stratified analyses by age (Table 2), we found that personalized Mg treatment significantly increased 5-mC methylation at cg16371860 (TSS1500, promoter) by 7.2% compared to placebo arm (decreased by 0.1%) in those aged < 65 years old. The difference remained statistically significant after adjusting for age, sex and baseline methylation as well as FDR correction (FDR-adjusted P =0.014). The effect of Mg treatment was not significant among those aged ≥ 65 years. On the other hand, we found that Mg treatment affected 5-hmC level at cg16371860 compared to the placebo arm after adjusting for age, sex, but the effect was not significant after further adjusting for baseline levels. In the stratified analysis by age, among those aged < 65 years, the effect of Mg treatment remained after adjusting for sex, age and baseline levels, but disappeared at FDR of 0.10.
Additionally, we found that Mg treatment reduced 5-hmC at cg26337277 (close proximity to TSS200 and 5’UTR, promoter) compared to placebo arm after adjusting for age, sex and baseline levels (P=0.012). However, the significant effect disappeared at FDR of 0.10 (FDR-adjusted P=0.386). In the stratified analysis by age, we found that Mg treatment significantly reduced 5-hmC level by 2.3% compared to increases by 7.1% in the placebo arm after adjusting for covariates in those aged < 65 years old (P=0.003). The effect remained significant at FDR of 0.10 (adjusted P value=0.088). No effect was found in those aged ≥ 65 years old.
DISCUSSION and CONCLUSION
In this personalized precision-based randomized trial, we found that 12-week of Mg treatment significantly increased 5-mC DNA methylation at cg16371860 CpG site and decreased 5-hmC cytosine modification at cg26337277 CpG site in the TMPRSS2 compared to placebo among participants aged less than 65 years old. To our best knowledge, no study has evaluated how to modify TMPRSS2 cytosine modification, nor examined the effect of Mg treatment on the cytosine modification in the TMPRSS2.
Cg16371860 is a CpG site located at a CpG-rich island 200–1500 bp upstream of the transcriptional start site (TSS1500) while cg26337277 CpG site resides with close proximity to 0–200 bp upstream of the TSS (TSS200) and the 5’-untranslated region (5’UTR). These loci are within the promoter region of the gene, therefore potentially playing a role in transcription initiation and regulation of gene expression. Although 70-80% CpG sites of human genome are methylated to maintain a stable molecular phenotype, regions of CpG island promoters of actively transcribed genes are frequently lowly methylated [52]. Hypermethylation of the promoter regions could impede transcription activity and repress related gene expression [53]. Similarly, reduced 5-hmC level is also associated with decreased gene expression because conversion from 5-mC to 5-hmC is an active demethylation process [54]. Our observations that Mg treatment induced increases in 5-mC methylation at cg16371860 and decreases in 5-hmC methylation at cg26337277 CpG in the promoter indicate a hindered process of transcription initiation and, subsequently, lower levels of TMPRSS2 expression.
Given that TMPRSS2 plays an essential role in facilitating SARS-CoV-2 entry, higher expression of TMPRSS2 may relate to higher viral loads in the host and subsequently, worse clinical outcomes. This has been supported by a recent finding that African Americans, individuals that were disproportionately affected by severe COVID-19 [55], had significantly higher average nasal epithelial gene expression of TMPRSS2 compared with other races/ethnicities [56]. High viral loads can further induce cytokine storm which triggers a violent inflammatory immune response that contributes to acute respiratory distress syndrome (ARDS), multiple organ failure, and finally death in severe cases of SARS-CoV-2 infection [57,58]. The association between high viral loads and fatal clinical outcomes has been observed in both human influenza A (H5N1) [59] and SARS-CoV-2 [60].
In addition to viral entry, TMPRSS2 expression levels may affect disease severity by regulating coagulation cascade and arterial thrombosis through the protease activated receptor (PAR)-signaling pathway [24–26]. This is further supported by a recent study reporting that polymorphisms near the TMPRSS2 were associated with thrombocytes count [25]. Moreover, nafamostat [61], one TMPRSS2 inhibitor has already been used in clinical practice as an effective anti-coagulant, and camostat mesylate [21], a closely related compound that are undergoing clinical trials to test their utility for COVID-19 treatment, further indicating the important role of TMPRSS2 in regulating thrombosis and, in turn, disease severity. Given that nearly half of patients with COVID-19 pneumonia developed thrombotic complications [28] and deceased patients were distinctively characterized with widespread vascular thrombosis [27], TMPRSS2 could be a key target for interventional strategies in reducing COVID-19 severity and mortality. Since TMPRSS2 plays a similar role during influenza infection [22], the findings from the current study suggest Mg status may also affect influenza’s infection and severity.
Strikingly, TMPRSS2 was first identified as a key regulator in prostate cancer. Strongly upregulated TMPRSS2 expression was observed in prostate cancer cell lines [62]. Consistent with racial disparities in nasal epithelial gene expression of TMPRSS2, African Americans also have higher incidence and mortality of prostate cancer than European Americans [63], further indicating that the high prostate cancer mortality among African Americans may be attributed to upregulated TMPRSS2 expression. However, factors that may impact modified TMPRSS2 expression remain unknown. Previous evidence found that African Americans had significantly lower levels of serum Mg or lower intake of Mg compared to European Americans [64] while low blood Mg levels and a high Ca/Mg ratio were significantly associated with high-grade prostate cancer [65]. Our findings from the current study in which Mg status was improved by modulating Ca: Mg intake ratio to around 2.3 suggest a protective effect from severe COVID-19 by Mg among individuals aged younger than 65 years through regulating DNA methylation/demethylation modifications and subsequently suppressing TMPRSS2 expression.
Although no significant effect on TMPRSS2 methylation was observed among those aged 65 years or older, it is likely Mg treatment may exert benefits through other mechanisms. First, our and other studies found Mg optimized body vitamin D status and substantially reduced the doses requirement for vitamin D [9,18,66]. Vitamin D has been proposed to generate beneficial effects in ARDS through activation of the vitamin D receptor (VDR) signaling pathway by reducing the cytokine storm, regulating the renin-angiotensin system, maintaining the integrity of the pulmonary epithelial barrier, and tapering down the increased coagulability [16]. Secondly, Mg deficiency is a prevailing, yet under-recognized driver for increased risks of cardiometabolic diseases including diabetes [67], hypertension, coronary heart disease, heart failure and thrombosis [68]. Previous studies found that low Mg plays an essential role in promoting endothelial cell dysfunction and generating a proinflammatory, pro-thrombotic and pro-atherogenic environment that could contribute to the pathogenesis of cardiovascular diseases and severe COVID-19 [69]. Thirdly, Mg deficiency caused by a gene mutation led to reduced cytotoxic activity of T cells and increased viral load, but Mg treatment reduced B cells positive for EBV, indicating Mg is critical in antivirus immunity [70]. Lastly, a study found that COVID-19 patients who carry the apolipoprotein E (ApoE) genotype were at a 2.3-fold increased risk of severe COVID-19 [71]. In addition to being a major risk factor for dementia [72], the ApoE genotype is strongly linked to lower human longevity [73], a genetic mark of early aging and is linked to lower plasma ApoE [74]. We reported from a randomized trial that in those age ≥65 years, Mg treatment improved cognitive function, particularly among elderly via demethylation in the ApoE gene which is expected to result in increased ApoE levels [47].
This study has several strengths, including the double-blinded randomized trial design. Furthermore, a precision-based design was used. Intakes of Mg and Ca from both diet and supplements were measured twice before and four times during the treatment and a personalized dosing strategy of Mg supplementation was administered to each participant. The Ca:Mg ratios remained stable over the 12-week study period. In addition, we had a high compliance with the study medication and the dropout rate was very low. The study has some weaknesses though. The primary concern is that TMPRSS2 expression was not measured in the PPCCT, and thus the association between DNA methylation changes and level of TMPRSS2 expression and phenotype patterns is not confirmed. However, the observed DNA methylation changes were internally consistent, with increased 5-mC and reduced 5-hmC indicating a reduced level of TMPRSS2 expression.
In summary, among individuals aged younger than 65 years with the Ca:Mg intake ratios equal to or over 2.6, reducing Ca:Mg ratios to around 2.3 increased 5-mC modifications (i.e. cg16371860) and reduced 5-hmC modifications (i.e. cg26337277) in the TMPRSS2 gene. The NIH has recently highlighted the crucial need for early intervention to reduce the likelihood of developing severe outcomes and reduce demand on healthcare system [1]. These findings, if confirmed, provide another mechanism for the role of Mg intervention for the prevention of COVID-19 and treatment of early and mild disease by modifying the phenotype of the TMPRSS2 genotype in addition to affecting ApoE methylation in the elderly [47] and optimizing the levels of vitamin D [9]. In addition, these findings also indicate a possible mechanism of Mg involved in prostate cancer and influenza.
Data Availability
The data have been deposited to the NIH database of Genotypes and Phenotypes (dbGaP).
Author contribution
Lei Fan: Conceptualization, Formal analysis, Writing- Original draft preparation; Xiangzhu Zhu: Data curation, Methodology; Yinan Zheng: Investigation, Writing- Reviewing and Editing; Wei Zhang: Investigation, Writing- Reviewing and Editing; Douglas L Seidner: Supervision, Writing- Reviewing and Editing; Reid Ness: Supervision, Writing- Reviewing and Editing; Harvey J Murff: Supervision, Writing- Reviewing and Editing; Chang Yu: Methodology, Funding acquisition, Writing- Reviewing and Editing; Xiang Huang: Validation, Formal analysis; Martha J Shrubsole: Supervision, Writing- Reviewing and Editing; Lifang Hou: Supervision, Funding acquisition, Writing- Reviewing and Editing; Qi Dai: Conceptualization, Project administration, Funding acquisition, Writing- Reviewing and Editing.
Funding Statement
This study was supported by R01 CA149633 (to Qi Dai & Chang Yu) and R01 CA202936 (to Qi Dai & Lifang Hou) from the National Cancer Institute, Department of Health and Human Services as well as the Ingram Cancer Center Endowment Fund. Data collection, sample storage and processing for this study were partially conducted by the Survey and Biospecimen Shared Resource, which is supported in part by P30CA68485. Clinical visits to the Vanderbilt Clinical Research Center were supported in part by the Vanderbilt CTSA grant UL1 RR024975 from NCRR/NIH. The parent study data were stored in Research Electronic Data Capture (REDCap) and data analyses (VR12960) were supported in part by the Vanderbilt Institute for Clinical and Translational Research (UL1TR000445).
ACKNOWLEDGEMENT
Footnotes
Author disclosures: All authors have no conflicts of interest.
The Personalized Prevention of Colorectal Cancer Trial (PPCCT) was registered at clinicaltrials.gov as NCT01105169.
CONFLICT of INTEREST SATEMENT All authors have no conflicts of interest.
References
- [1].↵
- [2].↵
- [3].
- [4].
- [5].↵
- [6].↵
- [7].
- [8].↵
- [9].↵
- [10].↵
- [11].↵
- [12].↵
- [13].↵
- [14].↵
- [15].↵
- [16].↵
- [17].↵
- [18].↵
- [19].↵
- [20].↵
- [21].↵
- [22].↵
- [23].↵
- [24].↵
- [25].↵
- [26].↵
- [27].↵
- [28].↵
- [29].↵
- [30].↵
- [31].↵
- [32].↵
- [33].↵
- [34].↵
- [35].↵
- [36].↵
- [37].↵
- [38].↵
- [39].↵
- [40].↵
- [41].↵
- [42].
- [43].↵
- [44].↵
- [45].↵
- [46].
- [47].↵
- [48].↵
- [49].↵
- [50].↵
- [51].↵
- [52].↵
- [53].↵
- [54].↵
- [55].↵
- [56].↵
- [57].↵
- [58].↵
- [59].↵
- [60].↵
- [61].↵
- [62].↵
- [63].↵
- [64].↵
- [65].↵
- [66].↵
- [67].↵
- [68].↵
- [69].↵
- [70].↵
- [71].↵
- [72].↵
- [73].↵
- [74].↵