Vitamin C improves gut Bifidobacteria in humans

Abstract

Aims: Numerous beneficial effects of vitamin C (ascorbic acid) supplementation have been reported in the literature. However, data on its effects toward the gut microbiome are limited. We assessed the effect of vitamin C supplementation on the abundance of beneficial bacterial species in the gut microbiome. Materials and methods: Stool samples were analyzed for relative abundance of gut microbiome bacteria using next-generation sequencing-based profiling and metagenomic shotgun analysis. Results: Supplementation with vitamin C increased the abundance of bacteria of the genus Bifidobacterium (p = 0.0001) and affected various species. Conclusion: The beneficial effects of vitamin C supplementation may be attributed to modulation of the gut microbiome and the consequent health benefits thereof.

Plain language summary

Vitamin C, also known as ascorbic acid, is used as a supplement for fighting infectious disorders. Many disorders, including COVID-19 and cancer, harmfully disrupt the levels of bacteria that naturally reside in the gut, which may contribute to symptoms. The aim of the study was to understand whether high-dose vitamin C could improve the types of bacteria in the human gut. To do this we characterized the gut bacteria before and after 23 individuals took vitamin C, as prescribed by their respective physicians. We observed that vitamin C increased levels of a gut bacterium called Bifidobacterium which has positive health benefits, including fighting infection. This study suggests the possibility that vitamin C could be successful for improving infection outcomes, possibly even COVID-19, partially because it improves the gut bacteria present.

Tweetable abstract

Patients receiving ascorbic acid supplementation had increased abundance of Bifidobacterium in their gut microbiome, which may help to explain some of the apparent health benefits and antiviral properties of vitamin C.

Summary points

• Low vitamin C concentrations have been reported in cognitively impaired patients, such as those with Alzheimer’s disease and dementia, and in patients with advanced cancer and severe SARS-CoV-2 infection.

• We hypothesized that vitamin C administration could modulate the gut microbiome contributing to protection from severe outcomes associated with viral illness, including SARS-CoV-2 infection.

• Supplementation with vitamin C increased the relative abundance of bacteria of the genus Bifidobacterium. Families Lachnospiraceae and Bifidobacteriaceae also significantly increased, and various species changed.

• Our observational study shows that vitamin C has microbiome-modulating properties, presenting a new potential mechanism for its therapeutic value.

• Moreover, the data demonstrate that vitamin C has a potential for creating Refloralization™ (restoration of the human gut microbiome) after Bifidobacterium depletion.

Keywords: :

• ascorbic acid
• bacterial diversity
• Bifidobacterium
• gut microbiome
• Lachnospiraceae
• vitamin C

Eminent chemist and two-time Nobel laureate Linus Pauling’s controversial scientific conjecture about the health benefits of vitamin C (ascorbic acid) has been the subject of much debate [1,2]. Pauling’s book [1] provoked resentment among several professionals because it was written for the lay public and it presented information that was not widely accepted by the medical establishment [3]. Since then, at least 100 studies have sought to determine the potential role of vitamin C in reducing the incidence, severity or duration of the common cold [3]. Two meta-analyses, published more than 40 years after Pauling’s book, reported that vitamin C supplementation only reduced the duration of colds in the general population by an average of 8% [2] and that extra doses of vitamin C given at the onset of cold symptoms could reduce duration and relieve symptoms [4]. The meta-analysis used a cutoff of <0.2 g of vitamin C per day, a dosage much lower than the 2–18 g per day recommended by Pauling [2]. Although a few studies have shown statistically significant reductions in incidence [^5-7], a consistent decrease in duration or symptom reduction can be observed in many more studies [5,6,^8-21], verifying some of Pauling’s claims. Many studies have also failed to demonstrate beneficial effects, and it may be possible that the public’s perception of beneficial effects of vitamin C may be responsible for some of the favorable findings [22].

Low vitamin C concentrations have been reported in cognitively impaired patients, such as those with Alzheimer’s disease and dementia [23], and in advanced cancer [24] and severe SARS-CoV-2 infection [25]. Between 0.8 and26% of adults in high-income countries appear to be vitamin C deficient, as defined by levels <11 μmol/l [26]. A US survey found that about 13% of the population was deficient, with the overall occurrence of age-adjusted vitamin C deficiency being closer to 7% and higher among lower socioeconomic classes [27]. However, it has also been suggested that because depletion of tissue stores can happen rapidly, short-term or intermittent vitamin C deficiency prevalence in the population could be much higher [28,29].

An increasing body of evidence has shown that the gut microbiome is a key regulator of immunity and host defense mechanisms. Disturbance of homeostasis involving interactions between the gut microbiome and the immune system can adversely influence resistance to viral infections, increase disease risk and alter neurocognitive function [^30-32]. Although previous studies have recognized that vitamin supplementation can alter the gut microbiome, no vitamins are presently classified as prebiotics (agents that promote the growth of beneficial microorganisms in the gut) by the International Scientific Association for Probiotics and Prebiotics [33,34]. Vitamin supplementation in patients with Crohn’s disease resulted in an altered gut microbiome composition when patients were administered riboflavin [35] or vitamin D [36]. Several additional studies on vitamin D and the gut microbiome have been performed [37] linking the mucosal immune system and the microbiome in inflammatory bowel disease [38], identifying host–microbe interactions and mutations in the vitamin D receptor as risk factors for inflammatory bowel disease [39] and suggesting that the vitamin D receptor can affect the gut microbiome [40]. Therefore substantial interest in the clinical significance of this finding should prompt further studies on vitamin-mediated modulation of the microbiome for the treatment and prevention of dysbiotic microbiota-related diseases.

Although the most common SARS-CoV-2 infection symptoms are respiratory, SARS-CoV-2 infections also target the gastrointestinal tract; investigations have shown that SARS-CoV-2 infection causes changes to the gut microbiota, including an overall decline in microbial diversity, enrichment of opportunistic pathogens and depletion of beneficial commensal microorganisms [^41-45]. It has been found that there is an association between SARS-CoV-2 infection severity and the abundance of certain bacteria, with symptomatic SARS-CoV-2-infected patients having significantly less bacterial diversity and lower relative abundances of Bifidobacterium and Faecalibacterium, while having increased Bacteroides [41,42,45]. This particular dysbiosis pattern may be amenable to pre- or post-infection intervention through probiotic supplementation or fecal microbiota transplantation [41,43].

It has been hypothesized that the use of vitamin C could reduce SARS-CoV-2 infection via its beneficial immunomodulating properties, which include neutralization of the inflammatory response, reduction of oxidative stress and stimulation of antiviral cytokines [^46-48]. During viral infections vitamin C has been shown to increase production of α/β interferons and downregulate the production of proinflammatory cytokines such as TNF and IL1-α/β [49,50]. However, more evidence-based clinical data are needed to support these findings.

A recent study by Pham and colleagues compared the effects of colon-targeted vitamins C, B2 and D on the human gut microbiome and reported that vitamin C produced the most distinct effect on the microbiome, increasing microbial alpha-diversity and short-chain fatty acids [51]. However, one meta-analysis found that vitamin C therapy did not reduce major health-related poor outcomes in SARS-CoV-2-infected patients [52]. However this meta-analysis did not consider the gut microbiome, specifically baseline Bifidobacterium levels. Also, it has been noted that larger prospective randomized trials are needed to evaluate the effect of isolated vitamin C administration [^46-48,52]. It is possible that vitamin C might help in a certain population of SARS-CoV-2-infected patients via modulation of the SARS-CoV-2-induced dysbiotic gut microbiome.

A large variety of factors act to shape and potentially disrupt individuals’ microbiomes, such as genetics, aging, diet, infections and medications [53]. We hypothesized that vitamin C administration could modulate the gut microbiome, which is a known regulator of immunity [^53-55]. It is possible that such microbiome changes could contribute to protection from viral illnesses associated with microbiome changes, including SARS-CoV-2 infection [41,56,57], and this is worthy of exploration.

Materials & methods

Study design & participants

Twenty-three participants (11 males and 12 females) with varying pre-existing medical conditions who were taking oral vitamin C supplementation daily, prescribed by their primary care or naturopath doctor, were enrolled in our study to understand how this vitamin consumption was affecting their microbiomes. No subjects were declined for participation, and subjects were recruited from 1 November 2019 until 31 January 2022. The dose of vitamin C ranged from 3 to 25 g/day and was administered either orally (daily) or intravenously (weekly). The duration of vitamin C supplementation varied among the subjects, ranging from 5 days to 1.5 years. There was no other change in the medication regimens of subjects during the period of vitamin C administration.

This study was approved by the ‘Ethical and Independent’ (E&I) review board (https://eandireview.com/).

Stool sample collection & processing

The procedure for collection and processing of fecal samples was published previously [41]. Stool samples from subjects were collected prior to baseline and 24 h after the last dose of vitamin C was given. Specifically, for baseline, the samples were collected 1 week to 7 months prior to vitamin C administration, which is valid in light of our analysis and data processing methods; specifically, our microbiome data were processed to remove the component of bacteria due to daily dietary and other fluctuations. We have internally validated these methods and demonstrated consistency in subjects’ microbiome readings over extended time periods, in the absence of major interventions or disease.

DNA/RNA Shield™ fecal collection tubes (Zymo Research, Cat # R1101, CA, USA) were used to collect 1 ml of fresh stool sample. DNA was then extracted from samples using the QIAmp^® PowerFecal^® Pro DNA extraction kit (Qiagen, Cat#51804, MD, USA). The isolated DNA was then quantified and normalized for downstream library fabrication using shotgun methodology. The prepared libraries were then pooled and sequenced using NextSeq 500/550 High Output v2.5 300 cycle kit (Illumina, Cat# 20024905, CA, USA) and run on the Illumina NextSeq 550 system as we previously reported [58]. Briefly, run setup parameters on the NextSeq Control Software (Illumina Local Run Manager) included paired-end sequencing set to 150 cycles with both Index 1 and 2 at 10 bp. Sequencing acceptance criteria were a Q-score (AQ30) ≥75%, cluster density between 120 and 240 K/mm², and clusters passing filter (PF%) ≥80%. Following successful next-generation sequencing quality control, sequences were mapped utilizing the minimap2 sequencing alignment tool in One Codex’s (CA, USA) bioinformatics analysis pipeline (open source, available at http://github.com/onecodex). A detailed description of the bioinformatics methods is available at http://docs.onecodex.com.

Data analysis

The DNA sequences of microbial strains were analyzed using metagenomic sequencing analysis and then compared for bacterial species present before and after vitamin C supplementation at all taxonomic levels. The data were uploaded to One Codex and analyzed against the One Codex database, which contains more than 115,000 complete microbial reference genomes. During processing, reads were first screened against the human genome, then mapped to the microbial reference database using a k-mer-based classification. Individual sequences (next-generation sequencing read or contig) were compared against the One Codex database by exact alignment using k-mers, where k = 31. Based on the relative frequency, unique k-mers were filtered to control for sequencing or reference genome artifacts. The relative abundance for the microbial taxa was then assessed, based on the depth and coverage for the available reference genomes in the database, as we previously reported [41]. Bacterial diversity was assessed using the Shannon (richness of bacterial composition) and Simpson alpha-diversity (evenness of bacterial composition) indices, calculated at the genus level [59].

The bioinformatic pipeline employed for bacterial identification (One Codex) matches all overlapping k-mers in a given read to the most specific organism and taxonomic level possible. Because not all k-mers are unique to an individual species or strain, each k-mer is classified to the lowest common ancestor within a taxonomic/phylogenetic tree. Finally, aggregated individual k-mers are matched across a given read and the most specific, consistent taxonomic ID is assigned to the read (i.e., the highest weighted root-to-leaf path of k-mer matches across the taxonomic tree). This bioinformatic approach utilizing the lowest common ancestor is our employed operational taxonomic unit, with more stringent criteria to increase taxonomic accuracy.

One Codex calculates relative abundance for all bacteria in each sample, many of which are at practically zero abundance. We collected data on 730 families, 2734 genera and 16,527 species. To focus our analysis, statistical analysis was performed on the families, genera and species studied by Otten et al. [60] in their similar, albeit preliminary, study that pioneered the question of vitamin C’s effect on the microbiome.

GraphPad Prism v. 9.4 (GraphPad Software, CA, USA) was used for statistical analysis and graphical image generation; p-values were calculated using Wilcoxon matched-pairs signed rank test, and p < 0.05 was considered statistically significant. All analyses and comparisons were performed for each subject individually, both prior to and after vitamin C supplementation. Mean and standard error of mean (SEM) are used when describing patient characteristics, and median and interquartile range are used when describing changes in relative abundance or diversity of bacteria. When calculating p-values, multiple comparisons were incorporated, using false discovery rate. Fold changes were calculated for each subject, and then the median was found for all subjects for a given bacterium. Individual points with zero initial relative abundance of bacteria were excluded. Throughout the study, relative abundances are presented on a scale of 0–1.

Results

This observational study enrolled 23 participants aged 5–80 years with varying medical conditions who received outpatient care at sites in California or via telehealth. Demographic clinical characteristics of subjects (n = 23) are presented in Table 1. The mean (SEM) age of participants was 49 ± 3.7 years and the mean ± SEM BMI was 25.2 ± 1.1 kg/m² (Table 2).

Table 1. Demographics, clinical characteristics and dosage of vitamin C supplementation in subjects.

Age (years)	Gender	Race	Past medical history	Ascorbic acid dose*	BMI (kg/m^2)
6	M	Lebanese	ASD; allergic to penicillin; tonsils and adenoids removed	12,000 mg/day p.o. for 10 days	18.2
80	F	Caucasian	Bipolar disorder; allergy to sulfa; elevated cholesterol; arthritis; osteopenia; cyst near pancreas	10,000 mg/day p.o. for 5 days	24.1
38	M	Caucasian	Lyme, mast cell and cardiovascular disease; CIDP; appendectomy; sepsis; seizures; kyphoplasty; GERD; cyclic vomiting syndrome; hypertension	10,000 mg/ day p.o. for 5 days	24.4
55	F	Moroccan	Hashimoto’s thyroiditis	10,000 mg/day p.o. for 5 days	NA
5	M	Lebanese	Healthy	12,000 mg/day p.o. for 10 days	21.7
34	F	Caucasian	Overweight	10,000 mg/day p.o. for 10 days	27.5
43	F	Caucasian	Healthy	5000 mg/week iv. + 3000 mg/day p.o. for 3 weeks	26.6
53	F	NA	Hypothyroid	10,000 mg/day p.o. for 10 days	NA
27	F	Caucasian/Hispanic	Healthy	5000 mg/day p.o. for 5 days	NA
53	F	Caucasian	Hypothyroid; IBD; PCOS	3000 mg/day p.o. for 1 year	32.0
54	F	Caucasian	Mold illness; cesarean section surgery	15,000 mg/week iv. + 3000 mg/day p.o. for 4 weeks	20.5
57	M	Caucasian	Anxiety; hernia repair; knee, shoulder and forearm surgeries	15,000 mg/week iv. + 3000 mg/day p.o. for 4 weeks	23.2
44	F	Caucasian/Asian	Past SARS-CoV-2 infection; cesarean section surgeries	20,000 mg/week iv. + 3000 mg/day p.o. for 3 weeks	22.3
54	M	Caucasian	Uses alcohol	3000 mg/week p.o. for 1.5 years	34.4
50	M	Caucasian	Anxiety; uses nicotine	10,000 mg/week iv. + 3000 mg/day p.o. for 3 weeks	27.1
54	M	Caucasian	Healthy	5000 mg/week p.o. for 2 weeks	29.0
46	M	Caucasian	Uses nicotine	20,000 mg/week iv. + 3000 mg/day p.o. for 3 weeks	21.8
55	M	Caucasian	Parkinson’s disease	20,000 mg/week iv. + 3000 mg/day p.o. for 3 weeks	20.2
58	F	Caucasian	Parkinson’s disease	20,000 mg/week iv. + 3000 mg/day p.o. for 3 weeks	20.0
63	M	Caucasian	Guillain–Barré syndrome; bacteremia	20,000 mg/week iv. + 3000 mg/day p.o. for 3 weeks	25.6
61	F	African–American	Low RBC, low WBC; hypothyroid; fibrocystic breasts	20,000 mg/week iv. + 3000 mg/day p.o. for 3 weeks	26.2
73	F	Caucasian	Diverticulitis and irregular heartbeats; hypothyroid	10,000 mg/week iv. + 3000 mg/day p.o. for 3 weeks	24.0
64	M	Caucasian	Colon polyps; hypertension; hypothyroid; high cholesterol; depression; long COVID	25,000 mg/week iv. + 3000 mg/day p.o. for 1 week	35.6

ASD: Autism spectrum disorder; CIDP: Chronic inflammatory demyelinating polyneuropathy; F: Female; GERD: Gastroesophageal reflux disease; IBD: Inflammatory bowel disease; iv.: Intravenous; M: Male; NA: Not available; PCOS: Polycystic ovary syndrome; p.o.: Per orem; RBC: Red blood cells; WBC: White blood cells.

Table 2. Summary of patient demographics and clinical characteristics.

Parameter	Value
Age (years): mean ± SEM; range	49 ± 3.7; 5–80
Male	11 (47.8%)
Female	12 (52.2%)
Relatively healthy^†	7 (30.4%)
BMI (kg/m^2): mean +/- SEM; range	25.2 ± 1.1; 18.2–35.6
Race: Caucasian Other Not specified	16 (69.5%) 6 (26.1%) 1 (4.3%)

† Includes healthy individuals, or those with history of only one or more of the following: hypothyroidism, past surgeries and/or nicotine/alcohol use. Otherwise healthy overweight/obese individuals are not counted as relatively healthy.

SEM: Standard error of the mean.

Changes in relative abundances of bacteria, grouped by rank (family, genus, species) before and a vitamin C administration are listed in Table 3. At the family level, we found that vitamin C supplementation significantly increases the Lachnospiraceae and Bifidobacteriaceae families (Figure 1A). Vitamin C supplementation also appears to alter the degree of abundance of the genus Bifidobacterium, as well as the species Collinsella aerofaciens and Barnesiella intestinihominis (Figure 1B & C). Specific to the Bifidobacterium genus, vitamin C supplementation increased the species Bifidobacterium adolescentis (Figure 1D).

Table 3. Changes in relative abundances of bacteria, grouped by rank (family, genus, species), before and after vitamin C administration.

Rank		Before vitamin C		After vitamin C		p-value
		Median	IQR	Median	IQR
Family	Bacteroidaceae	0.1508	(0.0669, 0.2165)	0.1001	(0.0754, 0.1306)	0.1982
	Bifidobacteriaceae	0.0044	(0.0010, 0.0342)	0.0468	(0.0328, 0.0704)	0.0001
	Lachnospiraceae	0.1557	(0.1088, 0.2091)	0.1983	(0.1456, 0.2336)	0.0301
	Oscillospiraceae	0.0051	(0.0023, 0.0101)	0.0029	(0.0014, 0.0049)	0.2722
	Ruminococcaceae	0.1262	(0.0492, 0.1946)	0.1578	(0.1019, 0.2239)	0.7314
Genus	Bacteroides	0.1508	(0.0669, 0.2165)	0.1001	(0.0754, 0.1307)	0.0501
	Bifidobacterium	0.0044	(0.0010, 0.0342)	0.0468	(0.0328, 0.0704)	0.0001
	Blautia	0.0247	(0.0113, 0.0332)	0.0301	(0.0179, 0.0542)	0.2726
	Dialister	0.0000	(0.0000, 0.0000)	0.0000	(0.0000, 0.0023)	0.2500
	Enterococcus	0.0000	(0.0000, 0.0001)	0.0000	(0.0000, 0.0000)	0.7695
	Roseburia	0.0251	(0.0113, 0.0406)	0.0138	(0.0083, 0.0365)	0.6650
	Veillonella	0.0000	(0.0000, 0.0002)	0.0000	(0.0000, 0.0002)	0.9219
Species	Akkermansia muciniphila	0.0000	(0.0000, 0.0000)	0.0000	(0.0000, 0.0014)	0.8125
	Bacteroides fragilis	0.0003	(0.0000, 0.0024)	0.0003	(0.0000, 0.0012)	0.4954
	Bacteroides vulgatus	0.0350	(0.0023, 0.0679)	0.0370	(0.0094, 0.0677)	0.8288
	Barnesiella intestinihominis	0.0000	(0.0000, 0.0000)	0.0000	(0.0000, 0.0007)	0.0313
	Bifidobacterium adolescentis	0.0000	(0.0000, 0.0016)	0.0000	(0.0000, 0.0265)	0.0029
	Bifidobacterium animalis	0.0000	(0.0000, 0.0000)	0.0000	(0.0000, 0.0154)	0.0580
	Bifidobacterium bifidum	0.0000	(0.0000, 0.0000)	0.0000	(0.0000, 0.0116)	0.0781
	Bifidobacterium breve	0.0000	(0.0000, 0.0000)	0.0000	(0.0000, 0.0000)	0.3125
	Bifidobacterium dentium	0.0000	(0.0000, 0.0000)	0.0000	(0.0000, 0.0000)	0.7500
	Bifidobacterium longum	0.0022	(0.0000, 0.0160)	0.0068	(0.0000, 0.0235)	0.0507
	Bifidobacterium pseudocatenulatum	0.0000	(0.0000, 0.0004)	0.0000	(0.0000, 0.0007)	0.3828
	Christensenella minuta	0.0000	(0.0000, 0.0000)	0.0000	(0.0000, 0.0000)	0.0625
	Collinsella aerofaciens	0.0000	(0.0000, 0.0063)	0.0000	(0.0000, 0.0266)	0.0322
	Faecalibacterium prausnitzii	0.0630	(0.0149, 0.0717)	0.0573	(0.0164, 0.0789)	0.1601
	Gemmiger formicilis	0.0012	(0.0000, 0.0099)	0.0020	(0.0000, 0.0111)	0.4657
	Holdemanella biformis	0.0000	(0.0000, 0.0000)	0.0000	(0.0000, 0.0000)	0.8750
	Intestinibacter bartlettii	0.0000	(0.0000, 0.0005)	0.0001	(0.0000, 0.0020)	0.0785
	Prevotella copri	0.0000	(0.0000, 0.0000)	0.0000	(0.0000, 0.0000)	0.2500
	Streptococcus thermophilus	0.0000	(0.0000, 0.0006)	0.0003	(0.0000, 0.0022)	0.3778

Median and IQR of relative abundance (depicted on a scale 0–1) are shown, with bold p-values and bacterial names indicative of significant change.

IQR: Interquartile range.

Figure 1. Relative abundances (expressed as a fraction) of bacteria, grouped by rank.

(A) Family. (B) Genus. (C & D) Species. Bars indicate median + interquartile range. Symbols indicate individual points: x = prior to vitamin C, square = after vitamin C.

*p < 0.05; **p<0.01; ***p < 0.001.

Changes in relative abundance for all bacteria analyzed, grouped by rank and calculated as median of fold changes for individual subjects, are shown in Figure 2A. Relative abundance of genera of bacteria for each subject before and after and vitamin C administration is shown in Figure 2B. Figure 2B provides an overview, focused on the most common bacterial genera. Figure 2A & Table 3 focus on bacteria of interest, which are similar to the bacteria analyzed by Otten et al. [60].

Figure 2. Changes in relative abundance of bacteria in response to Vitamin C adminstration.

(A) Fold change of relative abundance for all bacteria analyzed, calculated as median of fold changes for individual subjects. (B) Relative abundance of genera of bacteria for each subject (left) before and (right) after vitamin C administration.

Of note, a few subjects (Figure 2B) appeared to have a sizably different response to vitamin C administration. For instance, subject 19’s Bifidobacterium relative abundance drastically decreased, while other subjects’ abundance all increased. Regardless, the change in Bifidobacterium abundance remained highly significant. Likewise, the Bacteroides abundance typically decreased after vitamin C administration; however, subjects 4 and, again, 19 showed opposite results.

Three bacterial types were chosen for closer analysis: genera Bifidobacterium and Bacteroides, and family Lachnospiraceae. A significant increase in abundance of the genus Bifidobacterium (p = 0.0001) and family Lachnospiraceae (p = 0.0301) and a strong trend of decrease in the bacterial genus Bacteroides (p = 0.0501) were seen after supplementation of vitamin C (Figure 3A–C). Vitamin C supplementation appears to increase the abundance of Bifidobacterium approximately threefold higher compared with baseline (Figure 3A).

Figure 3. Relative abundance of bacteria for organisms potentially affected by vitamin C.

Relative abundance before and after vitamin C administration shown for individual subjects for (A) Bifidobacterium (p = 0.0001), (B) Bacteroides (p = 0.0501) and (C) Lachnospiraceae (p = 0.0301). Graphs on right panel plot median + interquartile range.

*p < 0.05; ***p < 0.001.

There was no difference in the Shannon diversity index (p = 0.7069) or Simpson index (p = 0.5839) pre and post vitamin C supplementation (Figure 4A & B).

Figure 4. Alpha diversity at the genus level.

As measured by (A) Shannon diversity index (p = 0.7069) and (B) Simpson diversity index (p = 0.5839) for each subject before and after vitamin C administration.

Discussion

With the advent of high-throughput sequencing technology and bioinformatics, it is now feasible to explore the function of the gut microbiome at a more detailed level. This observational study explored the effect of the micronutrient vitamin C on the composition and diversity of the gut microbiome. Our results indicate that vitamin C increases the abundance of gut bacteria of the genera Bifidobacterium. A study by Otten et al. investigated vitamin C supplementation at a dose of 1 g per day for 2 weeks. They observed a more than fourfold increase in the mean relative abundance of Bifidobacterium, supporting our findings [60]. Differences between our study cohort and the cohort in that study included the mean age of participants, and participants being moderately active and healthy in Otten et al. versus participants with pre-existing medical conditions in this study [60].

Our observational study shows that vitamin C has microbiome-modulating properties, presenting a new potential mechanism for its therapeutic value. The use of probiotics has been shown to result in a statistically significant reduction in the incidence of upper respiratory tract infections (p = 0.002) [61]. In another study investigating probiotic treatment, CD8⁺ cytotoxic T cells and T suppressor cells were enhanced with probiotic treatment during the first 14 days of supplementation [62]. This microbiome mechanism may explain its potential role in improving the common cold and other respiratory viral illnesses.

Members of the genus Bifidobacterium are considered beneficial bacteria and are an indicator of a healthy gut [54]. Bifidobacterium are among the first microbes to colonize the human gastrointestinal tract [54,63] and are used as probiotics due to their health-promoting properties [64]. Bifidobacterium plays a role in several beneficial functions such as increased ATP production, modulation of the immune system, mucosal barrier integrity and production of short-chain fatty acids [54,55,^64-66].

The administration of probiotics containing Lactobacillus gasseri PA 16/8, Bifidobacterium longum SP 07/3, Bifidobacterium bifidum MF 20/5 (5 × 107 CFU/tablet) for at least 3 months has been shown to reduce the severity of cold symptoms and shorten the duration by almost 2 days [62]. Several studies in mice have revealed that Bifidobacterium can protect mice from influenza infection by increasing anti-influenza IgG [67], balancing Th1/Th2 responses against infection and decreasing IL-6 production in the lungs, leading to increased survival rates [68]. Bifidobacterium also mediates anti-influenza effects through the production of the metabolites valine and coenzyme A [69]. A human trial demonstrated that oral administration of B. longum BB536 resulted in lower influenza infection rates in the elderly [70]. Neutrophil, phagocytic cell and NK cell activity remained higher through the end of the study in the BB536 group compared with the control group. Recently, studies have reported a decrease in abundance of Bifidobacterium associated with SARS-CoV-2 infection and severity [41,56,57]. The promising results of recent preclinical and clinical trials investigating dietary supplementation of probiotics, including Bifidobacteria, for treating SARS-CoV-2 infection have provided hope that these bacteria can be an important means to fight SARS-CoV-2 infection; however, further work is needed [43,^71-73]. Moreover, this study would suggest vitamin C could be used for the process of restoring microbe levels, such as Bifidobacterium, a process we call Refloralization™.

Work by Shu et al. [74] found that vitamin C concentration can increase the growth of B. bifidum BB01 and BB02 in vitro; however, it is also affected by many factors, including oxygen levels and pH. However, obligate anaerobes are especially susceptible to oxidative stress, and the presence of oxygen can severely compromise the growth of Bifidobacterium [^75-79]. The redox potential of the gut is linked to the ratio of aerobic or facultative anaerobic and anaerobic species [77]. As an antioxidant, vitamin C would be expected to exert a direct effect on the redox balance in the gut and may modulate the microbiome via this mechanism. It is important to note that electrons from ascorbate also have the capacity to reduce metals, some of which may be present in the stool (e.g., iron and copper), which may lead to the production of superoxide and hydrogen peroxide and the subsequent generation of reactive oxygen species when high concentrations are achieved [80,81]. More work is needed to determine the concentrations of ascorbate in the gut that are achieved by supplementation, the role of the route of administration (oral vs intravenous) and how this influences the redox potential, ascorbate’s antioxidant versus pro-oxidant behavior and, ultimately, the gut microbiome and metabolomics.

Also consistent with the study by Otten et al., we observed an increase in the relative abundance of the family Lachnospiraceae. Lachnospiraceae are the predominant bacteria which likely produce short-chain fatty acids in the gut of healthy subjects [82]. Increased populations of Lachnospiraceae may be associated with markers of good health, such as improved epigenetic states, increased fatty acid metabolism and decreased inflammatory markers [82]. Nonetheless, when one examines Lachnospiraceae in various human diseases and disease models, the results prove more complex: either high or low levels are associated with the presence of various disorders [82]. Whether the increase in Lachnospiraceae due to vitamin C is beneficial or harmful would require further studies, and may depend on the individual’s diet, microbiome, health history and other parameters.

Gut microbiome diversity and mammalian microbial co-metabolism are closely interconnected with overall human health and wellbeing. Reduced levels of microbial diversity are linked to several acute and chronic diseases, including SARS-CoV-2 infection [41,42,83]. The study by Pham et al. that investigated colon-targeted vitamins found that vitamin C significantly increased microbial alpha-diversity and composition [51]. Although we observed no changes in the overall diversity of the gut flora post vitamin C supplementation, specific supplementation parameters may make it possible to achieve this effect.

Given that this was an observational study and vitamin C was being administered for clinical as well as research purposes, there was a sizable variation in not only the length, route of administration and dose of vitamin C, but also the ages, weights and medical histories of the patients. As such, these parameters are worth discussing. One systematic review and meta-analysis of six randomized controlled clinical trials involving a total of 572 SARS-CoV-2-infected patients reported dosing of vitamin C ranging from 50 mg/kg/day to 24 g/day, with routes of administration including both intravenous (four studies) and oral (two studies) [52]. This meta-analysis showed that administration of vitamin C did not have any effect on major health outcomes (positive or negative) in SARS-CoV-2-infected patients, compared with either placebo or standard therapy, irrespective of its dosage, route of administration and disease severity. A recent phase I clinical trial found that intravenous vitamin C, administered in a dose of 1.5 g/kg three times weekly, appears to be safe and essentially free of adverse side effects [84].

Our study observed both intravenous and orally administered dosages, which likely have different pharmacokinetic profiles. Recent pharmacological modeling revealed that orally administered vitamin C, even at very large and frequent dosing, will increase plasma concentrations only modestly, from 0.07 to a maximum of 0.22 mM, while intravenously infused doses were predicted to result in peak plasma vitamin C levels over 60-times higher and urine concentrations 140-fold higher than oral doses, via pharmacokinetic modeling [85]. The actual bioavailability of the molecule is controlled by numerous factors, including absorption by the intestine and other tissues, kidney absorption and excretion, and other patient-specific factors [^86-90]. Studies of supplemental iron showed that oral iron affects the diversity of gut bacteria and the production of gut metabolites differently compared with intravenous iron [91,92]. In addition to the differences in concentration delivered via intravenous versus oral administration, oral administration delivers vitamin C directly to the gut microbiome, as opposed to via the bloodstream; thus its effect on the gut microbes could relate to kinetics and concentrations completely different from those affecting plasma levels. More work is needed to determine routes of administration and dosage schedules that achieve optimum, persistent vitamin C concentrations and a consistent physiological effect in the gut. Despite the potential for variation in results due to methods of administration, we observed clear statistical differences in the relative abundances of the genus Bifidobacteria.

Although the sample size available was small, we were able to observe statistically significant alterations in two bacteria present in the stool samples. Despite having a range of participants who varied in terms of age, gender, ethnicity, route of administration and pre-existing medical conditions, these trends were clearly identifiable. A larger-scale study would allow us to explore these correlations in more detail. Our use of metagenomic sequencing methods (as opposed to 16s ribosome sequencing) is a less commonly used approach that may allow for a more detailed characterization of the microbiome, including possible physiologically important components [93]. Future studies should seek to identify differences based on BMI, gender, diet and particular medical conditions. Specific factors that may affect the microbiome results are vitamin C supplementation dose and duration, blood and baseline vitamin C levels and the contribution of diet to overall vitamin C intake. Future studies should also look at disease-specific effects, such as the effects on prevention or reduction of respiratory viral illness.

Conclusion

Vitamin C as a therapeutic agent should be explored specifically for its potential to reverse or ameliorate disorders linked to microbiome dysbiosis, especially Bifidobacterium deficiencies. It may be able to restore the gut microbiome (i.e. carry out Refloralization) after Bifidobacterium depletion due to various conditions or acute illness, including respiratory viral illnesses such as SARS-CoV-2 infection.

In conclusion, we postulate that the microbiome, specifically Bifidobacterium, is a mediator of the numerous reported beneficial effects of supplementation and mega-dose administration of vitamin C and that vitamin C supplementation can be used to restore Bifidobacterium. Longer-term and larger studies are still needed to understand the effects of vitamin C on the microbiome. This study also points out the need for the measurement of baseline Bifidobacterium levels in placebo-controlled trials, to compare similar populations.

Author contributions

S Hazan: conceptualization, data curation, investigation, methodology, project administration, resources, software, formal analysis, writing (original draft), writing (review and editing). S Dave: formal analysis, investigation, software, writing (original draft), writing (review and editing). L Martin: investigation, writing (review and editing). M Howell: writing (review and editing). N Deshpande: writing (original draft), writing (review and editing). A Papoutsis: writing (review and editing), formal analysis

Ethical conduct of research

The authors state that they have obtained appropriate institutional review board approval (Ethical and Independent IRB: study number 20110) and have followed the principles outlined in the Declaration of Helsinki for all human or animal experimental investigations. In addition, for investigations involving human subjects, informed consent has been obtained from the participants involved.

Acknowledgments

The authors would like to thank K Hendricks for writing and editorial support and helpful discussion.

Financial & competing interests disclosure

S Hazan declares that she has pecuniary interest in Topelia Pty Ltd in Australia and Topelia Pty Ltd in USA. She is the founder and owner of Microbiome Research Foundation, ProgenaBiome and Ventura Clinical Trials. S Dave declares she has corporate affiliation to McKesson Specialty Health/Ontada and North End Advisory, LLC. A Papoutsis declares he has corporate affiliation to ProgenaBiome, LLC. N Deshpande and M Howell declare they have corporate affiliation to North End Advisory, LLC. L Martin declares corporate affiliation with LEI NanoTech, LLC and MNT SmartSolutions, LLC. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Medical writing support was provided by Dr. Sonya Dave of North End Advisory, LLC and was funded by the corresponding author and her institutions, ProgenaBiome, LLC and Microbiome Research Foundation. Additional medical writing support was provided by Dr. Nirupama Deshpande and Dr. Mark Howell, also of North End Advisory, LLC and were funded by ProgenaBiome, LLC.