Dietary nitrate and population health: a narrative review of the translational potential of existing laboratory studies

Background Dietary inorganic nitrate (NO3−) is a polyatomic ion, which is present in large quantities in green leafy vegetables and beetroot, and has attracted considerable attention in recent years as a potential health-promoting dietary compound. Numerous small, well-controlled laboratory studies have reported beneficial health effects of inorganic NO3− consumption on blood pressure, endothelial function, cerebrovascular blood flow, cognitive function, and exercise performance. Translating the findings from small laboratory studies into ‘real-world’ applications requires careful consideration. Main body This article provides a brief overview of the existing empirical evidence basis for the purported health-promoting effects of dietary NO3− consumption. Key areas for future research are then proposed to evaluate whether promising findings observed in small animal and human laboratory studies can effectively translate into clinically relevant improvements in population health. These proposals include: 1) conducting large-scale, longer duration trials with hard clinical endpoints (e.g. cardiovascular disease incidence); 2) exploring the feasibility and acceptability of different strategies to facilitate a prolonged increase in dietary NO3− intake; 3) exploitation of existing cohort studies to explore associations between NO3− intake and health outcomes, a research approach allowing larger samples sizes and longer duration follow up than is feasible in randomised controlled trials; 4) identifying factors which might account for individual differences in the response to inorganic NO3− (e.g. sex, genetics, habitual diet) and could assist with targeted/personalised nutritional interventions; 5) exploring the influence of oral health and medication on the therapeutic potential of NO3− supplementation; and 6) examining potential risk of adverse events with long term high- NO3− diets. Conclusion The salutary effects of dietary NO3− are well established in small, well-controlled laboratory studies. Much less is known about the feasibility and efficacy of long-term dietary NO3− enrichment for promoting health, and the factors which might explain the variable responsiveness to dietary NO3− supplementation between individuals. Future research focussing on the translation of laboratory data will provide valuable insight into the potential applications of dietary NO3− supplementation to improve population health.


Background
Dietary inorganic nitrate (NO 3 − ) is a polyatomic ion present in large quantities in green leafy vegetables and certain root vegetables such as beetroot [1]. In recent years, inorganic NO 3 − has attracted substantial attention as a potential health promoting and exercise performance-enhancing dietary compound. These effects have largely been attributed to its ability to serve as a substrate for the ubiquitous gasotransmitter, nitric oxide (NO; Fig. 1) [2]. Following consumption, inorganic NO 3 − is absorbed in the upper gastrointestinal tract, increasing plasma NO 3 − concentration [3]. In the blood, exogenous NO 3 − mixes with endogenous NO 3 − produced via oxidation of NO. Most (~60%) of the ingested NO 3 − is excreted in the urine [4]. However,~25% is actively taken up by the salivary glands via the transporter protein sialin [5], and secreted into the oral cavity, where it is reduced to nitrite (NO 2 − ) by facultative anaerobic bacteria residing primarily on the dorsal surface of the tongue [6,7]. Salivary (in the saliva) NO 2 − is then swallowed and a portion is converted into NO and other nitrogen oxides in the acidic environment of the stomach [2,8,9]. A further portion of the swallowed NO 2 − reaches the systemic circulation, where it can be transported to various tissues and reduced to NO by a range of enzymatic and non-enzymatic catalysis [2,3]. By increasing the bioavailability of NO and other nitrogen oxides, which play a role in the regulation of multifarious physiological processes, inorganic NO 3 − has the capacity to elicit far-reaching effects in the human body.
Current research has provided valuable insight into optimisation of NO 3 − supplementation strategies (e.g. pharmacokinetics, dose-response and supplementation duration) [13,64,65] and mechanisms of action [51,53,57,66]. Nevertheless, more research is needed to understand whether findings from typically small, wellcontrolled laboratory studies are likely to translate into clinically relevant improvements in population health. This article highlights key areas for further research that could help in this regard. Such research is warranted to help guide practitioners, influence policy, and form guidelines for the effective and safe consumption of inorganic NO 3 − .

Main text
Research focus 1: large-scale, longer duration trials Although NO 3 − consumption has been linked with a range of positive health outcomes, the majority of trials exploring the salutary effects of inorganic NO 3 − have involved short-term supplementation regimens, typically a few days in duration. Only a handful of trials have explored the medium-to longer-term effects of NO 3 − consumption (4 weeks to 6 months), usually focusing on BP or endothelial function as an outcome. Whilst not a universal finding [67,68], beneficial effects of medium-to longer-term NO 3 − supplementation protocols have been reported in some trials [69][70][71]. For example, Siervo et al. [69] found that 2 months supplementation with NO 3 − -rich beetroot juice (~6.5 mmol/d NO 3 − ) decreased 24-h systolic and diastolic BP by − 10.8 and − 5.4 mmHg, respectively, in a Sub-Saharan African setting. Similar effects were also observed when NO 3 − rich beetroot juice was co-ingested alongside folate. In another study, Mills and colleagues [70] showed that 6 months consumption of NO 3 − -rich beetroot juice (~11 mmol/d NO 3 − ) decreased central systolic pressure by − 2.6 mmHg. Likewise, Kapil et al. [71] reported reductions in 24-h systolic and diastolic BP (7.7 and 5.2 mmHg, respectively) and improved endothelial function and arterial stiffness with 4 weeks NO 3 − -rich beetroot juice supplementation (6.4 mmol/d NO 3 − ) with no change after placebo. Although focusing on different outcomes to the above trials, a study by Thompson et al. [72] also showed greater adaptations to sprint interval training in individuals consuming NO 3 − rich beetroot juice (13 mmol/d) over a 4-week period, providing further evidence of a benefit of this supplement when given over prolonged periods.
By contrast, studies by Blekkenhorst [67] and Sundqvist [68] observed no effects of 4-and 5-week NO 3 − interventions on BP. The lack of effect in these studies could be related to the relatively low NO 3 − doses administered (2.4 and 4.8 mmol/d NO 3 − , respectively). Conversely, the source of NO 3 − (vegetables or NO 3 − pills rather than NO 3 − -rich beetroot juice) could be relevant in explaining the lack of effect in these studies, given different foods providing equivalent doses of NO 3 − appear to have divergent effects on plasma NO 2 − concentration and BP [16], which could be linked to the (poly) phenol and ascorbate content of these foods [73]. Indeed, in most studies to date NO 3 − has been administered as beetroot juice, which is also rich in a constellation of different bioactive compounds, particularly (poly) phenols and the betalains [74]. Independent of NO 3 − , betalains have been shown to possess antioxidant [75], antiinflammatory [76], and vasodilatory [77] properties, although studies in humans are still scarce. To isolate the effects NO 3 − from other compounds in beetroot juice, researchers often compare the effects of a NO 3 − rich beetroot juice to a taste-, smell-and appearancematched NO 3 − depleted juice. One limitation of this strategy is that it cannot account for any synergistic interactions between NO 3 − and the other bioactive compounds that may augment the physiological effects of beetroot juice; in other words, we cannot be certain if the positive effects in these studies are simply due to NO 3 − or its interactions with the other bioactive compounds present. Thus, studies chiefly aimed at untangling the mechanistic effects of NO 3 − may prefer to administer NO 3 − in the form of NO 3 − salts instead of food-based supplements that contain other compounds.
Overall, additional comparisons of NO 3 − rich beetroot juice and sodium NO 2 − or NO 3 − supplements are required. When interpreting the findings of the studies discussed in this review, it is important the reader is aware that studies with NO 3 − salts and NO 3 − -rich beetroot juice do not contain the same compounds and therefore different effects are possible. Notwithstanding, as discussed in Section 4 of this review, cross-talk between NO 3 − and other dietary components or participant-level differences in the response to NO 3 − could also account for the lack of effect of NO 3 − in the studies of Blekkenhorst [67] and Sundqvist [68].
Based around the current evidence it is likely that, under the right circumstances (which remain to be fully elucidated), consumption of inorganic NO 3 − could elicit longer-term health benefits. In order to fully appreciate the potential applications of NO 3 − on population health, large-scale (e.g. n= > 1000), longer duration (e.g. 2-5 years) trials which focus not only on risk factors (e.g. BP, endothelial function, cognitive function), but also incidence of key non-communicable diseases (e.g. CVD, dementia) are warranted. Specific considerations for the design of such studies are provided in Table 1. Whilst likely to be logistically complex and require substantial financial backing from funders, this research could be justified by the promising evidence from short-term trials and the potential application of findings to ease the unsustainable societal and financial burden of conditions such as CVD (annual global costs~$863 billion [78]) and dementia (annual global cost~$1 trillion [79]). Prior to undertaking this research, it is essential to obtain more data on the feasibility and acceptability of different strategies to increase habitual NO 3 − intake by a sufficient quantity and for a sufficient period to obtain long lasting health benefits. This information is critical for the design of feasible longer-term trials and translation to the general population, and will be explored in more detail in the next section.
Research focus 2: feasibility and acceptability of different strategies to facilitate prolonged, increased consumption of nitrate To date, a limited number of studies have reported data on the feasibility and acceptability of beetroot juice as a vehicle for increasing habitual NO 3 − intake. Mixed findings have been reported. For example, Ormesher and colleagues [80] gave 40 pregnant women 70 mL/d concentrated beetroot juice (~400 mg of NO 3 − ) and, after 8 days of ingestion, 97% of participants indicated they would consume the supplement again, if they were experiencing benefits. However, only 62% of participants reported finding it easy to consume the beetroot juice and just over half of the participants rated the drink as palatable (54%). These findings suggest that longer-term consumption of beetroot juice may be difficult in this cohort, which could impede longer-term adherence. More recently, Kandhari et al. [81] evaluated the feasibility of a 60-day concentrated beetroot juice and folate intervention to treat hypertension in Sub-Saharan Africa. No serious adverse events were reported, and compliance was > 90%, suggesting beetroot juice was well accepted in this population. In addition, all participants rated the taste as "good" or "very good" and most participants (~87%) indicated a preference for beetroot juice over BP medication. The studies by Ormesher et al. [80] and Kandhari et al. [81] both administered the same brand of concentrated beetroot juice, such that the different findings cannot be attributed to a different type of supplement administered. Alternatively, it is possible that the different findings of Ormesher et al. [80], which was conducted in the UK, and Kandhari et al. [81], which was conducted in Tanzania, reflect cultural/ regional differences in food preference. However, it is noteworthy that participants in the Ormesher et al. [80] study were also pregnant, which may have further

Other lifestyle factors
Avoidance of smoking Outcomes Inclusion of hard clinical endpoints (e.g., CVD or dementia incidence) to build upon promising findings on risk factors for these conditions contributed towards the difference in palatability given pregnancy is known to influence taste [82]. In another study, Babateen et al. [83] examined the feasibility of different doses of concentrated beetroot juice in overweight and obese older adults over a 13week period. Compliance was high, no adverse events were reported, and the attrition rate was 19%, which is similar or lower than the dropout rates reported in other human intervention trials [84,85]. Collectively, these studies suggest that beetroot juice may represent an acceptable strategy to facilitate increased consumption of NO 3 − , at least in certain cohorts. However, future studies need to evaluate the feasibility and acceptability of beetroot juice consumption over longer periods (e.g., > 6 months) and in other populations.
Concentrated, commercially available beetroot juice shots have the advantage of being readily available (they are now sold in many major supermarket chains) and contain a standardised dose of NO 3 − sufficient to influence myriad health outcomes. This form of beetroot juice has also been shown to be more effective at reducing BP (and presumably eliciting other physiological changes) than non-concentrated beetroot juice when the same dose is administered [86]. In addition, as mentioned in the previous section, beetroot juice also contains other bioactive compounds that may contribute to overall health. Nevertheless, as participants do not always enjoy the taste of beetroot juice and the relatively high cost of commercially available beetroot 'shots' (£ 1-2 or $2-3 each) may be prohibitive to some users, it is essential for researchers to explore the feasibility and acceptability of other strategies to increase NO 3 − consumption. This could include other NO 3 − -rich foods (e.g. lettuce, rhubarb, spinach, radish), gels, powders, crystals, capsules and non-beetroot drinks. To this end, both Blekkenhorst et al. [67] (> 98% compliance) and Sundqvist et al. [68] (> 97% compliance) demonstrated excellent compliance to 4 and 5 week interventions, respectively, with NO 3 − -rich vegetables, which were well tolerated with minimal side effects. Importantly, Sundqvist et al. [68] reported similar compliance between NO 3 − -rich vegetables and NO 3 − -containing pills (> 97% vs. > 98%). Nevertheless, neither Blekkenhorst et al. [67] nor Sundqvist et al. [68] reported beneficial physiological effects of their interventions, which could be related to the relatively modest NO 3 − doses provided (~2.4 and 4.8 mmol/d respectively) or other methodological factors which were discussed in Section 1 of this review. A comprehensive investigation of patient preferences and the real and perceived barriers of adopting a high-NO 3 − diet or consuming NO 3 − -rich supplements warrants further investigation. In addition, studies need to determine the amount of NO 3 − -rich vegetables required to elicit beneficial physiological effects, whether this is achievable for different populations, and whether effects are superior to non-vegetable NO 3 − sources. Finally, it is worth exploring whether there are regional and population preferences, as this knowledge could be used to develop more targeted NO 3 − products.
Research focus 3: nitrate intake and health outcomes in epidemiological studies The role of dietary NO 3 − for human health has gradually shifted over the last five decades. Indeed, this compound was initially considered as a risk factor for cancer, endocrine disorders and infant methaemoglobinaemia. However, the stigma attached to dietary NO 3 − has gradually dwindled, and NO 3 − is now viewed by many a potential health-promoting compound (see Section 6 for further details). The initial results suggesting a harmful role of dietary NO 3 − intake (from food) were mostly derived from animal models and weakly designed epidemiological studies which have had a prominent, almost demonizing, influence on defining the role of dietary NO 3 − for human health [87]. These initial studies informed the still contentious WHO nutritional recommendations for dietary NO 3 − intake in humans which was set at 3.7 mg/ kg body weight [88]. The perception of dietary NO 3 − as a risk factor started to change with the discovery of the role of NO 3 − as key substrate for the NO 3 − -NO 2 − -NO pathway and the evidence of a beneficial effect of NO 3 − on health parameters such as BP.
After the study by Larsen et al. [10] in 2006, which first demonstrated a BP lowering effect of sodium NO 3 − , there was a rapid surge in research testing the effects of dietary NO 3 − on health outcomes [89]. However, the research strategy in the last decade has almost taken an inverse approach to that typically adopted in nutritional science as the conduction of clinical trials have surpassed epidemiological investigations, which are generally considered as a first step to validate research hypotheses [90][91][92]. One of the primary reasons for the inverse trend is the lack of reliable and representative food databases of NO 3 − content to support an accurate dietary assessment [93]. An additional limitation is the severe lack of validation studies testing the accuracy of dietary assessment methods against valid biomarkers of NO 3 − intake (e.g. 24-h urinary NO 3 − concentrations) [94]. This is compounded by the fact that the NO 3 − content of vegetables will vary by farming method (whether NO 3 − fertiliser is used or not), growing conditions, time of year the crop is harvested, and storage conditions [1], such that there is likely to be a degree of error in estimated NO 3 − intake values [95]. Several research groups have developed independent databases by collecting data on NO 3 − food content from published sources in an attempt to obtain valid estimates of NO 3 − intake and evaluate associations with health outcomes [96,97].
Although this is a step in the right direction, it remains difficult to accurately estimate long-term habitual dietary NO 3 − intake for the reasons mentioned above. In addition, NO 3 − concentrations measured in biological fluids have been used in some analysis as indirect markers of NO 3 − intake [98]. Whether these objective markers of NO 3 − intake show stronger links with health outcomes compared with subjective, self-reported NO 3 − intake values, is the subject of ongoing research. A summary of the key non-cancer related epidemiological studies testing the association of inorganic NO 3 − with health outcomes is provided in Table 2.
The first studies to evaluate the association between dietary NO 3 − intake and health outcomes were conducted in 2016 in Iran (two studies) [99,101] and in the United States (one study) [100]. The former evaluated the association of vegetable NO 3 − intake with risk of chronic kidney disease in the Tehran Lipid and Glucose Study and found a higher prevalence of chronic kidney disease (CKD) at baseline (cross-sectional analysis) in the high-NO 3 − intake group whereas no significant association with CKD risk was observed after a 3-year follow up [101]. Using the same dataset, Bahadoran et al. [99] found that dietary NO 3 − intake, overall and from animal sources, was not associated with prospective risk of diabetes. The US study was conducted in a very large sample (> 100,000 participants) and assessed dietary NO 3 − intake in the Nurses' Health Study and the Health Professionals Follow-up Study [100]. The results showed a significantly lower risk of primary open-angle glaucoma in participants with higher NO 3 − intake [100]. However, a subsequent analysis conducted in the Nurses' Health Study found a non-significant association between dietary NO 3 − and prospective risk of coronary heart disease [111]. More recently, several crosssectional and longitudinal studies have observed significant associations between high NO 3 − intake or urinary NO 3 − concentrations (as a proxy for NO 3 − intake) with cardiovascular outcomes including lower BP [107], risk of hypertension [109], common carotid intimal medial thickness [103], congestive heart failure [114] and CVD mortality [109]. Conversely, higher plasma NO 3 − concentrations in the Framingham Offspring Study [106] were associated with an increased risk of all-cause mortality, which may be explained by the rise in plasma NO 3 − concentrations in participants with impaired kidney function included in the analysis and highlights the potential risk of reverse causality in these investigations. The improvements in physical performance and cognition observed in some of the NO 3 − supplementation trials were also explored in two cross sectional studies [98,112]. Improved hand-grip strength and timed up and go tests (a test of functional mobility) were observed in middle-aged and older Australian participants with a higher NO 3 − intake [112] whereas NO 3 − concentrations measured in spot urine samples were not associated with improved cognition in 1015 older Americans participants enrolled in the National Health and Nutrition Examination Survey [98]. The NIH workshop on dietary NO 3 − held in 2016 [115] advocated for more epidemiological research to be conducted to better define the predictive role of dietary NO 3 − consumption for the prevention as well as treatment of chronic diseases. The consensus statement also encouraged the development of detailed and country-specific NO 3 − food composition tables for a more accurate assessment of the exposure to dietary NO 3 − [115]. The current epidemiological evidence points towards a protective role of dietary NO 3 − intake for cardiovascular events and mortality whereas the predictive role for cancer risk is still undefined as latest meta-analyses on the topic indicate a lack of association between dietary NO 3 − consumption and cancer risk [116,117]. There is still scarce or no data from prospective studies on the association of dietary NO 3 − intake with other chronic conditions with established links with NO 3 − /NO 2 − and NO pathways such as diabetes, hypertension, physical disability or dementia. Further epidemiological studies in this area are therefore warranted. Such research will complement the findings from RCTs, by providing information on the effectiveness of a NO 3 − for disease reduction in real-world circumstances with greater sample sizes and longer follow up than is logistically feasible in most RCTs [90,91].

Research focus 4: inter-individual differences in the response to nitrate
At the individual participant level, several groups have suggested the existence of possible 'responders' and 'non-responders' to NO 3 − , irrespective of the vehicle used to provide this inorganic anion [64,118,119]. It is important to note that random within-subject variation could explain much of the variability in response to NO 3 − supplementation between individuals [120,121]. Similarly, issues may also arise when attempting to establish whether an individual is a dependable 'responder' or 'non-responder' on different occasions [122,123]. Nevertheless, several factors have been identified which could explain genuine differences in the response to NO 3 − between individuals. These include individual characteristics such as age [124,125], health [126] and exercise training status [40,49], sex [14], genetic factors [127], and differences in the oral microbiome (explored further in Section 5 of this review). In addition, betweenparticipant differences in potentially plastic lifestyle factors such as smoking status [128], use of mouthwash [129], and habitual diet [49,130] might also impact an individual's response to NO 3 − . We briefly review the impact of these variables on the effects of NO 3 − below.  [40,131]. This theory stemmed from several studies reporting that while NO 3 − supplementation from any source enhanced exercise performance in recreational level athletes (V O 2peak 40-60 ml/kg/min), such effects were less pronounced or nonexistent in well-trained and elite endurance athletes (typically manifesting a V O 2max > 60 ml/kg/min) [132][133][134][135] − supplementation less effective, and these are discussed in detail elsewhere [40,131,136]. One prominent explanation is that elite endurance athletes might produce more NO via the canonical NOS pathways and are therefore less reliant on NO 3 − as a substrate for NO generation [132]. Furthermore, recent evidence indicates that NO 3 − might elicit preferential effects on type II compared with type I muscle fibres [54, 57, 58]. Well-trained endurance athletes might therefore benefit less from NO 3 − supplementation given a lower proportion of type II, and a higher proportion of type I, muscle fibres compared with recreationally active individuals [137,138] [144] both demonstrated BPlowering effects of NO 3 − (potassium NO 3 − and beetroot juice, respectively) in men with higher baseline BP and lower plasma NO 2 − concentrations but not in women. Likewise, in a meta-analysis by Jackson et al. [17], BP reductions with NO 3 − were greater in studies with more male participants. Those authors speculated that this could be related to a greater vascular production of NO in pre-menopausal women due to oestrogen-related release and activity of NO [145], diminishing the response to supplemental NO 3 − in women compared with men. Although studies remain scarce, there is some evidence that the heterogeneous responses to NO 3 − supplementation are partly explained by polymorphisms in the eNOS gene. This was first explored by Hobbs et al., [127], who examined the effects of NO 3 − supplementation on BP in patients with and without a specific polymorphism in the eNOS gene (G894T), which has been suggested to inhibit NO production from eNOS [127]. Although findings are equivocal [146], the G894T polymorphism, alongside being a T allele carrier, has been associated with cardiovascular disease [147][148][149], of which a key risk factor is diminished NO bioavailability [150,151]. Intriguingly, despite the small sample size (n = 14), Hobbs et al., [127] found that NO 3 − supplementation (beetroot bread) only reduced BP in patients who were both T allele carriers and had the G894T polymorphism in the eNOS gene. A more recent study examined the influence of the G894T polymorphism and NO 3 − therapy on mortality in chronic heart failure patients [146]. Somewhat at odds with the findings of Hobbs et al., [127], Azzam et al. [146] found that NO 3 − therapy (source not specified) increased the risk of mortality in patients with the G894T polymorphism, and to a greater extent in G allele carriers, suggesting that NO 3 − therapy might increase mortality in advanced heart failure. However, as this study was observational, cause-effect relationships cannot be established. Moreover, the findings are at contrast to the beneficial effects of NO 3 − shown in most [152][153][154][155], but not all [156,157], short term intervention trials which show that NO 3 − improves cardiac function and/or exercise capacity in heart failure patients. Clearly, more studies with larger cohorts are required to determine the extent to which genetic variation influences the responsiveness to NO 3 − supplementation, but the findings from these two studies raise the possibility that that genetic factors could contribute towards the inter-individual variability reported by many studies.
Smoking has been shown to increase plasma and salivary concentrations of thiocyanate [158], a compound which competitively inhibits uptake of NO 3 − into the salivary glands [159], potentially reducing the amount of 'substrate' available to the oral bacteria for reduction into NO 2 − . Consequently, it is possible that smokers will experience compromised NO 3 − metabolism and thus a diminished physiological response to NO 3 − supplementation versus non-smokers. Indeed, Bailey et al. [128] demonstrated a smaller increase in salivary NO 3 − , plasma NO 3 − and NO 2 − concentration, and an attenuated BP response, following a NO 3 − bolus (beetroot juice) in smokers compared to non-smoking controls.
It is possible that supplemental NO 3 − is ineffective at eliciting meaningful physiological changes in individuals habitually consuming a high NO 3 − diet. Nevertheless, as population intake of NO 3 − is typically low -Babateen et al. [93] reported a median intake of 108 mg/d in healthy individuals. With very few individuals regularly consuming NO 3 − levels to match those provided through supplementation [160], high habitual NO 3 − intake is unlikely to explain a lack of response to NO 3 − supplementation in most 'non-responders'. Alternatively, there is compelling evidence to suggest that consumption of other dietary compounds alongside NO 3 − may have the capacity to influence response to this compound, such that an individual's background diet could determine (at least transiently) their status as a NO 3 − 'responder' or 'non-responder'. For example, consumption of glucosinolate-rich vegetables, such as those from the Brassica family like broccoli, cauliflower, and cabbage, proximal to consumption of NO 3 − -rich vegetables was shown to blunt the BP lowering response of the latter [130]. Interestingly, this appears to be related to a similar mechanism to which smoking attenuates the effect of NO 3 − . Specifically, during processes that result in plant cell membrane damage such as mastication, glucosinolates are exposed to the enzyme myrosinase, which catalyses the hydrolysis of glucosinolates into thiocyanate [161]. Although consumption of thiocyanate-rich vegetables leads to lower salivary and plasma thiocyanate concentrations compared with smoking, Dewhurst-Trigg et al. [130] showed that the BP-lowering effect of a NO 3 − -rich smoothie was attenuated by the presence of thiocyanate rich vegetables. In that study, thiocyanate did not seem to interfere with NO 3 − transport into the mouth (as evident by similar salivary NO 3 − concentrations when NO 3 − was consumed alongside vegetables that were both high and low in thiocyanate), suggesting that thiocyanate may influence other aspects of NO 3 − metabolism. Specifically, co-ingestion of thiocyanate synthesising vegetables and NO 3 − -rich vegetables lowered salivary NO 2 − concentration compared to ingestion of NO 3 − -rich vegetables alone. This suggests that some Brassica vegetables might transiently alter the oral microbiome, consistent with the antimicrobial effects of thiocyanate derivatives in the oral cavity [162].
A study by Hughan et al. [163] found that the coingestion of sodium NO 3 − alongside conjugated linoleic acid, an unsaturated fatty acid particularly abundant in dairy and meat products, attenuated the rise in plasma NO 3 − and NO 2 − concentrations and supressed the BPlowering and platelet-inhibiting effects that were apparent when supplements were administered in isolation. Mechanistically, co-consumption of conjugated linoleic acid altered the metabolic fate of ingested NO 3 − leading to the formation of conjugated linoleic acid nitration products, which do not appear to have the same vasodilatory and platelet inhibiting properties as NO 2 − and NO. Likewise, Bailey et al. [164] found that the ingestion of iodide, which is fortified in many foods [165] and known to compete for salivary NO 3 − uptake [159], lowered salivary NO 3 − concentration when co ingested with NO 3 − rich beetroot juice. However, the increase in salivary and plasma NO 2 − concentration, alongside the lowering of BP, were similar compared with NO 3 − alone. Finally, a possible interaction between dietary NO 3 − and sulphate was identified by Kuhnle et al. [105] who indicated that when estimated sulphate intake was low, higher dietary NO 3 − intake was associated with lower BP. Conversely, when sulphate intake was high, this association was reversed, such that greater NO 3 − intake was actually associated with higher BP. The mechanistic basis through which sulphate could modulate the BP lowering effects of dietary NO 3 − is presently unknown. Collectively, the evidence presented above indicates that the response to NO 3 − is unlikely to be uniform between individuals, and could also potentially differ within individuals based around malleable lifestyle factors such as habitual diet. Better understanding the factors that influence responsiveness to NO 3 − is crucial to maximise the efficacy of NO 3 − -based interventions and will facilitate the development of targeted interventions for individuals most likely to benefit from consumption of this compound. Given many of the factors which appear to moderate the effectiveness of NO 3 − impact the oral conversion of this compound into NO 2 − , future research could also explore the potential physiological effects of direct NO 2 − administration (for a recent example, see [166]), which does not require processing in the mouth and could theoretically elicit more consistent responses between individuals. Nevertheless, caution should be taken to ensure such a strategy does not increase formation of potentially carcenogenic nitrosamines [167].

Research focus 5: oral microbiota and oral health
Once in the oral cavity, NO 3 − is reduced to NO 2 − during the anaerobic respiration of facultative and obligate bacteria which are particularly abundant on the dorsal surface of the tongue [168]. The oral microbiome collectively comprises over 700 individual species or phylotypes of bacteria that are organised in a series of complex interdependent communities [169]. To date, 14 species of bacteria have been identified as NO 3 − reducers, the majority of which are from the genera Veillonella, Prevotella, Neisseria, and Haemophilus [170]. A greater relative abundance of these bacteria on the tongue has been shown to augment the rate and magnitude of salivary NO 2 − production following the ingestion of NO 3 − rich beetroot juice [171]. Conversely, disruption of the oral microbiome by antibacterial mouthwash causes a transient loss of viable NO 3 − -reducing bacteria [172] and severely blunts the generation of NO 2 − in the saliva [173]. Strong antibacterial mouthwash has also been shown to increase BP, likely due to suppression of NO production from the NO 3 − -NO 2 − -NO pathway [174][175][176]. These data confirm the essential role of the oral bacteria in NO homeostasis and support the hypothesis that oral and systemic health are inextricably linked [177].
The mouth is continually exposed to the external environment and is regularly subjected to brushing, flossing, and nutrient intake, all of which may influence the physiological conditions inside the oral cavity and alter the composition of the bacterial milieu [178]. Ageing is known to cause a reduction in salivary flow rate [179] and has been reported to alter the composition of the oral microbiome in some [180,181] but not all [182] studies. Other factors may also be expected to influence the abundance and activity of oral bacteria, including exercise, diet, oral and systemic diseases, haemodialysis [183] and peritoneal dialysis [184] and medication (particularly antibiotics). In particular, the ingestion of NO 3 − -rich beetroot juice has been shown to increase salivary pH and cause meaningful alterations to the oral microbiome in favour of oral health [182,185]. Given the multitude of potential modifiers, it is perhaps unsurprising that there is profound between-individual variation in the abundance of NO 3 − -reducing bacteria [121].
Of note, these authors also reported significant withinindividual week-to-week variability in the abundance of these bacteria and the magnitude by which plasma NO 2 − increased following the ingestion of NO 3 − -rich beetroot juice. This was despite participants standardising their diet, physical activity, use of mouthwash, teeth brushing, and tongue cleaning between visits. The unpredictability in how different individuals respond to NO 3 − supplementation and how the same individual responds across repeated visits poses a particular challenge for researchers who wish to explore the therapeutic effects of this dietary intervention.
While recent advancements in genomic sequencing techniques have greatly enhanced our understanding of human bacterial interactions in the context of NO homeostasis, several important questions remain unanswered. To date, the majority of the research exploring links between the oral microbiome and health outcomes has only reported the relative abundance of phyla, genera, or species. Although this quantifies the proportional makeup of the community structure it does not reveal the metabolic activities of individual bacterial species [186] which may vary depending on substrate availability, metabolite expression from neighbouring microbes and host cells, and the impact of environmental conditions [187]. Future research should deploy meta-transcriptomic analysis to determine how factors such as diet, medication, physical activity, ageing, and disease influence NO 2 − and NO 3 − reductase gene expression of the oral bacteria. Furthermore, data from epidemiological studies and short-term intervention trials seem to support the notion that increasing habitual dietary intake of NO 3 − can improve markers of oral health and reduce the incidence of caries [185,188,189]. It remains to be established whether dietary NO 3 − supplementation may also be an effective treatment method for those already suffering from oral diseases such as chronic periodontitis.

NO 3
− is increasingly recognised as a beneficial ion that protects against chronic disease, yet, as noted in Section 3 of this review, historically, it was considered a food contaminant with adverse health effects, particularly increased risk of certain cancers and methaemoglobinaemia [1,88]. While the aforementioned WHO ADI for NO 3 − of 3.7 mg/kg of body mass remain in place today, the discovery of multiple positive health effects of NO 3 − have prompted a re-examination of these claims.
In 2004 the WHO reaffirmed their restrictions on NO 3 − intake yet, in 2008, a panel of experts from the European Food Safety Authority, concluded that the epidemiological evidence did not support an association between NO 3 − and cancer risk [190]. Similarly, in 2010, the International Agency for Research on Cancer confirmed that there was inadequate evidence to suggest NO 3 − from food or water was carcinogenic in humans [191]. Evidence that NO 3 − might cause infant methemoglobinemia, which was first mooted in the 1940s [192], has also been questioned. Indeed, an investigation conducted on behalf of the WHO in 2004 found no exposure-response relationship between dietary NO 3 − and methemoglobinemia in infants [193]. It is also worth noting that although some studies report mild adverse symptoms with high NO 3 − intake such as nausea and sickness, to the authors knowledge, no serious adverse events have ever been reported in clinical trials administering NO 3 − [1,93]. Notwithstanding, the available evidence does not rule out the possibility that prolonged consumption of NO 3 − above the ADI could harm health. Currently, at least with short to medium term intakes, research suggests that doses exceeding the ADI are needed to optimise vascular health or exercise performance [17,48]. Because most human trials have only examined the acute health effects (< 4 weeks) of increased NO 3 − intake, the longterm safety of consuming NO 3 − in amounts that exceed the ADI is not well understood. At present, epidemiological studies provide the strongest evidence that prolonged, high intakes of NO 3 − are safe. Indeed, these indicate that rather than being harmful, dietary NO 3 − intake is inversely associated with cardiovascular disease risk [102,194] and certain cancers [117]. Furthermore, diets and dietary patterns high in fruits and vegetables are linked to greater longevity [195,196], protection against type 2 diabetes [197] and chronic obstructive pulmonary disease [198], and improved cardiovascular [92,199,200] and cognitive health [201,202]. This suggests that higher intake of dietary NO 3 − , at least through plants, is more likely to be associated with health benefits than adverse effects.
Some animal studies have explored the longer-term effects of high dietary NO 3 − intake on health. In a study in rats, 10 weeks of a low sodium NO 3 − dose (0.1 mmol/kg/ d), which the authors suggest is equivalent to amounts achievable in the human diet, reduced BP, whereas a much higher dose (1 mmol/kg/d), elevated BP [203]. Interestingly, this study found that the high NO 3 − dose down-regulated eNOS activity, not only suggesting a crosstalk between the canonical and NO 3 − -NO 2 − -NO pathway, but also that any vascular benefits afforded by NO 3 − supplementation could wane over time. Nonetheless, these findings were not supported by a more recent animal study from the same group. Hezel and colleagues [204] fed mice the human equivalent of 350 mg/d or 26 mg/d of sodium NO 3 − . After 17 months, mice consuming the high NO 3 − diet did not have elevated BP or any other adverse health effects, despite the fact the dose exceeded the WHO recommended ADI for an adult under~95 kg. On the contrary, the high NO 3 − diet decreased plasma insulin and modulated inflammation, findings consistent with the metabolic benefits observed in acute human studies [205]. These effects need to be verified in humans but support the notion that prolonged increases in NO 3 − intake are not harmful to health.
It is important to note that any carcinogenic risk attributed to NO 3 − intake could be mitigated by the intake of antioxidants such as vitamin C or (poly) phenols, which are present in most fruits and vegetables. Studies have shown that vitamin C and E are effective inhibitors of nitrosamine formation [167]. In addition, (poly) phenols, which are abundant in commonly consumed NO 3 − sources such as spinach and beetroot [206], can also abrogate nitrosamine formation [73]. Thus, increasing NO 3 − intake through a greater vegetable intake may significantly lessen the risk of any NO 3 − induced nitrosamine formation. This could partly explain why diets high in vegetables are associated with a reduced and not heightened risk of cancer.
Health concerns have also been raised over the high oxalate content of NO 3 − -rich vegetables [207,208]. Oxalates are present in several foods, but particularly high in spinach, beetroot, and rhubarb [208,209]. Intake of these foods increases urinary oxalate excretion, a risk factor for renal stone formation [209][210][211], thus, it is currently recommended that foods rich in dietary oxalates are consumed in moderation [208,210]. However, the link between dietary oxalates and kidney stone formation remains equivocal. Although consuming oxalate rich foods increases oxalate excretion, a large prospective study (> 190,000 participants) found only modest non-significant associations between dietary oxalate intake and kidney stone risk, concluding that dietary oxalate intake is not a major risk factor for the formation of kidney stones in younger or older adults [212]. Furthermore, the Dietary Approaches to Stop Hypertension (DASH) diet, which is high in oxalates and NO 3 − -rich vegetables [1], was recently shown to increase urinary oxalate excretion but reduce the risk of kidney stone formation in~260 patients [213]. The authors attributed these findings to the high calcium and magnesium content of the diet limiting oxalate absorption. This is supported by previous research showing that oxalates from beetroot have low bioavailability (< 1%), owing to their high calcium content [209]. While more prospective human trials are needed, evidence that oxalate rich vegetables increase the risk of kidney stone formation is limited.
To summarise, claims that dietary NO 3 − promotes cancer or methemoglobinemia, or that dietary oxalates cause kidney stones are weak and unsubstantiated.
Rather, there is compelling evidence that dietary NO 3 − has salutary health effects and warrants consideration as a long-term therapeutic treatment strategy to manage vascular and metabolic health. Notwithstanding, longerterm studies in humans are lacking and thus it cannot be ruled out that a prolonged increase in NO 3 − intake, above the WHO recommended ADI, may have adverse effects for some individuals. Thus, it is incumbent that researchers examine the long-term safety of increasing dietary NO 3 − consumption in a range of contexts and populations. This research will be vital for convincing the public and regulators that NO 3 − consumption is safe and that current recommendations to limit dietary NO 3 − intake should be re-considered.

Conclusions
This article has briefly outlined the current state of knowledge around the potential health effects of dietary inorganic NO 3 − . Six key areas worthy of future research were identified to enhance understanding of the potential role of NO 3 − in improving population health. As such, it is hoped that this article will help direct researchers to further explore the role of NO 3 − as a potential health-promoting dietary component.

Authors' contributions
This review was conceived by OMS and TC, and designed by OMS, CE, AIS, MS, SJB and TC. OMS, CE, AIS, MS, SJB and TC drafted and critically revised the manuscript. OMS created the schematic. All authors approved the final version of the manuscript prior to submission.

Funding
No specific funding was received to support the writing of this review.
Availability of data and materials Not applicable.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.