Microbiota metabolites: Pivotal players of cardiovascular damage in chronic kidney disease

&NA; In chronic kidney disease (CKD), cardiovascular (CV) damage is present in parallel which leads to an increased risk of CV disease. Both traditional and non‐traditional risk factors contribute to CV damage in CKD. The systemic role of the microbiota as a central player in the pathophysiology of many organs is progressively emerging in the literature: the microbiota is indeed involved in a complex, bi‐directional network between many organs, including the kidney and heart connection, although many of these relationships still need to be elucidated through in‐depth mechanistic studies. The aim of this review is to provide evidence that microbiota metabolites influence non‐traditional risk factors, such as inflammation and endothelial dysfunction in CKD‐associated CV damage. Here, we report our current understanding and hypotheses on the gut‐kidney and gut‐heart axes and provide details on the potential mechanisms mediated by microbial metabolites. More specifically, we summarize some novel hypotheses linking the microbiota to blood pressure regulation and hypertension. We also emphasise the idea that the nutritional management of CKD should be redesigned and include the new findings from research on the intrinsic plasticity of the microbiota and its metabolites in response to food intake. The need is felt to integrate the classical salt and protein restriction approach for CKD patients with foods that enhance intestinal wellness. Finally, we discuss the new perspectives, especially the importance of taking care of the microbiota in order to prevent the risk of developing CKD and hypertension, as well as the still not tested but very promising CKD innovative treatments, such as postbiotic supplementation and bacteriotherapy. This interesting area of research offers potential complementary approaches to the management of CKD and CV damage assuming that the causal mechanisms underlying the gut‐kidney and gut‐heart axes are clarified. This will pave the way to the design of new personalized therapies targeting gut microbiota. Graphical abstract Figure. No caption available.

A C C E P T E D M A N U S C R I P T complex, bi-directional network between many organs, including the kidney and heart connection, although many of these relationships still need to be elucidated through in-depth mechanistic studies.
The aim of this review is to provide evidence that microbiota metabolites influence non-traditional risk factors, such as inflammation and endothelial dysfunction in CKD-associated CV damage.
Here, we report our current understanding and hypotheses on the gut-kidney and gut-heart axes and provide details on the potential mechanisms mediated by microbial metabolites. More specifically, we summarize some novel hypotheses linking the microbiota to blood pressure regulation and hypertension. We also emphasise the idea that the nutritional management of CKD should be redesigned and include the new findings from research on the intrinsic plasticity of the microbiota and its metabolites in response to food intake. The need is felt to integrate the classical salt and protein restriction approach for CKD patients with foods that enhance intestinal wellness.
Finally, we discuss the new perspectives, especially the importance of taking care of the microbiota in order to prevent the risk of developing CKD and hypertension, as well as the still not tested but very promising CKD innovative treatments, such as postbiotic supplementation and bacteriotherapy.

Introduction
Cardiovascular (CV) damage is one of the main concerns in the clinical management of chronic kidney disease (CKD). Indeed, CKD patients are exposed to a 2-to 4-fold higher risk of A C C E P T E D M A N U S C R I P T cardiovascular diseases (CVD) than the general population and are more likely to die from CVD than from CKD [1,2]. CVD incidence is not sufficiently accounted for and supported by the traditional risk factors [3]. Quite the contrary, many emerging non-traditional risk factors, including endothelial dysfunction and inflammation, are being recognized as co-responsible in the aetiology of CV damage [4]; in addition there is increasing evidence that the microbiota should be considered a key player in this setting, since its metabolites influence CKD progression and the aforementioned non-traditional risk factors [3][4][5].
Moreover, novel indications suggest that the gut and even oral microbiota may play a role in blood pressure (BP) control and in the aetiology of hypertension [6,7]. Hypertension is a frequent comorbidity and a risk factor associated with CKD, also in light of the major role of the kidney in BP regulation. The new evidence for an association between hypertension and gut microbiota dysbiosis adds a new tile to the complex frame of the gut-kidney axis.
In this review we discuss the role of the microbiota and its bioactive metabolites as a key player in the social network represented by gut-kidney-heart axis, and we summarize the current understanding and hypotheses on the role of gut microbiota as a non-traditional risk factor for CV damage. In addition a review of some literature studies on the enterosalivary NO pathway where oral microbiota is a key actor is also provided. What is more, molecular details in support of the putative mechanisms of uremic toxins-induced increase in CV risk are given together with an overview of the future perspectives in microbiota modulation, both in terms of prevention and therapy. Indeed, in light of the high plasticity of the gut microbiota in response to diet, and in consideration of the fundamental importance of nutrition in CKD, the nutritional management and the pro-, pre-, syn-and postbiotic supplementation offer an enormous potential in terms of reduction of CV damage by targeting the gut.

The microbiota issue in CKD: the gut-kidney axis
Increasing evidence is demonstrating the presence of a variety of relationships between the gut microbiota and extra-intestinal organs, including the kidney, heart and brain, often referred to as "axes" [5,[7][8][9]. A complex, bi-directional relationship between the gut microbiota and the kidney A C C E P T E D M A N U S C R I P T does exist. This is true in physiological conditions, with the microbiota influencing metabolism and immunity and the kidney dealing with microbial metabolites, but becomes crucial in pathology [5,8].
Indeed, during manifest CKD, especially in advanced stages, the kidney is not anymore able to deal with microbial-derived uremic toxins (UTs) that start to accumulate in the blood triggering a cascade of inflammatory and oxidative reactions [5] and inducing pro-fibrotic effects [10][11][12] likely to be responsible for renal damage progression. Several studies in CKD patients have investigated the relationship between increased serum levels of microbial-derived UTs and CKD progression [13,14], even though these associations have not yet been conclusively established [15]. On the other hand, it is the same uremic status that induces a shift in the gut microbial metabolism and composition [16], and the resulting dysbiosis is often worsened, especially in advanced CKD, by the current nutritional management which is characterized by a drastic fiber restriction that favours the proteolytic imbalance [17].
Some studies have been focused on the characterization of gut microbiota dysbiosis in CKD.
Studying microbiota in haemodialysis (HD) patients, Hida et al. found an increase in enterobacteria and enterococci, accompanied by a decrease of bifidobacteria compared with controls [18]. In a more comprehensive study conducted on end-stage renal disease (ESRD) patients, the group of Vaziri highlighted an abundance of 190 bacterial operational taxonomic units from Brachybacterium, Catenibacterium, Enterobacteriaceae, Halomonadaceae, Moraxellaceae, Nesterenkonia, Polyangiaceae, Pseudomonadaceae, and Thiothrix families, compared with healthy subjects [16]. In a metabolomic characterization study by the same group, an expansion of bacterial families possessing urease, uricase, and indole and p-cresol forming enzymes, and a reduction of families possessing butyrate-forming enzymes was observed in ESRD [19]. Overall these results indicate a shift towards the proteolytic metabolism which results in increased UTs production.

The cellular toxicity of protein-bound uremic toxins
The impact of protein-bound UTs on vital processes and on clinical outcomes in CKD has long been neglected, thus delaying the development of intervention strategies suitable to reduce their plasma levels to alleviate their potential toxic effects.
Due to their reduced renal clearance, they progressively accumulate in CKD patients: many biological and toxic effects have been attributed to them [20].
Actually, colonic microbes produce a large number of uremic solutes. Larger populations of CKD patients with colectomies and the use of complementary LC-MS-based analysis methodologies would be required in order to obtain a full profile of the microbial-derived compounds retained in uremia. In fact, most uremic solutes can go unidentified because they cannot be found in standard metabolomic databases, as shown in a study comparing plasma from HD patients with and without colons, which has identified 5 of 30 colon-derived uremic solutes, including IS, PCS, indoxyl glucuronide, indoleacetic acid, p-cresyl glucuronide, phenyl sulfate, phenyl glucuronide, phenylacetic acid, phenyl acetyl glutamine, hippuric acid and 3-carboxy-4-methyl-5-propyl-2furanpropionic acid [25]. The high affinity of some UTs with serum proteins, such as albumin, makes them difficult to eliminate with standard dialysis when their serum levels increase, as it happens in CKD. Their accumulation, beyond their intrinsic toxicity, results in organ damage. Here we summarize the molecular basis of the cellular toxicity of the protein-bound UTs IS, PCS, and of the less investigated Indole Acetic Acid (IAA), all potentially contributing to an increased CV risk in CKD. Table 1 summarizes their synthesis and excretion pathways.
IS has been measured and observed to increase in the blood of patients with CKD since the 60s [26]. IS has a molecular weight of 213 Da and is a small solute deriving from the metabolism of dietary tryptophan, which is converted into indole by the tryptophanase present only in colon resident microbes. Indole is absorbed in the colon and then enters the systemic circulation where further oxidation and sulfation reactions in the liver form IS [5]. IS is taken up by organic anion transporters (OAT1 and OAT3) [27] placed on the basolateral membrane of tubular cells to be secreted in the urine. The interest in PCS as such is relatively more recent [28], since the volatile p-cresol started to be investigated, instead of its sulfate ester, in the blood of uremic patients in the 70s [29]. PCS, like IS, has a low molecular weight (188 Da) and originates exclusively in the colon from dietary nutrients deriving from the conversion of tyrosine and phenylalanine to p-cresol by anaerobic intestinal bacteria [5]. P-cresol is then absorbed in the gut and sulfated in the enterocytes and in the liver [30]. As IS, PCS is filtered by glomerular capillaries to the tubular cells through OAT-1 and OAT-3 to end-up in urine.
Even though deriving from different microbial metabolic pathways, IS and PCS are both actively secreted by kidney and tightly bind to albumin when entering the circulation, in healthy individuals as well as in CKD patients [31]. IS and PCS are bound to albumin in a reversible manner for approximately 97% and 95%, respectively, so that a rapid equilibrium between the bound and free fraction is established [31]. They share the same albumin-binding site for which they are competitive binding inhibitors [21]. Because of their binding properties, IS and PCS cannot efficiently be removed by HD and accumulate in ESRD patients thus exacerbating their potential toxic effects, especially at the CV level [32].
Lots of studies suggested the toxicity of IS and PCS in different type of cells, especially in renal and endothelial cells, although a very interesting review by Vanholder advises to use caution when it comes to interpreting the results of cell studies and pay special attention to the use of realistic concentration of IS, PCS and albumin [20]. The most studied in vitro effect of protein-bound UTs is the induction of oxidative stress in endothelial cells. Dou et al. [33] first demonstrated that IS and pcresol inhibit the endothelial proliferation in HUVEC and decrease endothelial wound repair. In the presence of albumin, at concentrations found in human plasma, IS and p-cresol decrease endothelial wound repair only at higher doses, comparable with the increase in the free non protein-bound fraction of these solutes in the experimental medium. This study suggests a role for these solutes in the endothelial dysfunction observed in uremic patients. The same authors found that IS induces a significant production of reactive oxygen species (ROS) in HUVEC [34], as confirmed by other researchers [22,31], increases NAD(P)H oxidase activity, and strongly decreases levels of glutathione, one of the most active antioxidant systems of the cells [34]. IS has been shown to induce a dose-dependent inhibition of endothelial cell proliferation [22] and an increase in cell senescence through the ROS-NF-κB-p53 pathway [22,35]. Furthermore, two studies [23,36] have investigated the effect of IS on oxidative stress and fibrosis in human proximal tubular cells, rat kidneys and heart. Their results suggest that IS (i) induces oxidative stress in renal cells and in CKD rats; (ii) worsens cardiac fibrosis and cardiomyocyte hypertrophy with enhanced oxidative stress and (iii) reduces anti-oxidative defence in hypertensive rats. In this regard, it is worthy pointing out that an in vivo study seems to confirm the involvement of IS in oxidative stress processes as it has been able to demonstrate the association of IS serum levels Inflammation-associated tubule-interstitial injury is a common pathological mechanism for kidney injury: a recent study [41] has shown that PCS and IS are able to stimulate significant cellular has a higher reduction rates by HD (44%) compared to IS [31]. Literature data show that UTs derived from tryptophan, such as IS and IAA, bind the transcription factor aryl hydrocarbon receptor (AhR) whose activation is associated with an increase in CVD risk [53]. Indeed, the same group observed that IS and IAA induced an increased expression of tissue factor in endothelial and peripheral blood mononuclear cells, by AhR activation [54]. Moreover, they studied the proinflammatory and pro-oxidant effects of IAA in endothelial cells, finding an increased endothelial ROS production by IAA at uremic concentration, as well as the activation of pro-inflammatory enzyme cyclooxygenase-2 mediated by AhR [55]. IAA was also found to increase the endothelial expression of the inflammatory genes IL-6, IL-8, ICAM-1, and MCP-1 [55]. Furthermore, in vitro experiments have shown that IAA, like IS, induces the apoptosis of endothelial progenitors cells in CKD [56], although the causative mechanisms are still unknown. Interestingly, beside the potential role of AhR in the atherosclerotic process [57], this receptor could also be involved in the imbalance between vascular damage/regeneration in CKD [58]. However, when assessing IAA actual toxicity, it is worth considering that the data reported are sometimes inconsistent [25] and need further insights.
A final remark about kynurenine, which derives from the main metabolic pathway of tryptophan, and its correlation with CKD severity and chronic inflammation [59]. The binding of kynurenine with AhR [60] is likely to be involved in the same events that lead to increased CV risk, as mentioned before [53]. Unlike IS and IAA, kynurenine is significantly reduced by HD [61].

How gut microbiota metabolites influence CV risk in CKD
The gut microbiota influences human physiology through its metabolites. The protein-bound UTs PCS, IS, IAA, but also other microbial metabolites such as Trimethylamine-N-Oxide (TMAO), Nitric Oxide (NO) and Short-Chain Fatty Acids (SCFA) represent the main molecules to which many experimental and clinical studies have ascribed a direct effect or an association with CV risk. They could be considered the mediators through which mainly the gut (but also the oral microbiota), influences CV risk in CKD. Table 2 summarizes the results of the main experimental and clinical studies that support the role of gut microbiota metabolites IS, PCS, TMAO and IAA in CV complications of CKD.
As already reported, the strong and evident pro-oxidative [31,36], pro-inflammatory [41] and profibrotic effects [10][11][12]62] of IS and PCS shown in in vitro and in vivo experiments as well as in the in vivo induction of aortic calcification by IS [63] seem to provide a molecular basis for their role as co-promoters of CV and renal damage progression in CKD. In addition, it is worth considering that protein-bound UTs are not efficiently removed by dialytic treatment and that this contributes to additionally increased CV risk in ESRD [32,64]. Well-established research findings have ascribed adverse clinical outcomes to these UTs, as reported summarized in a recent meta-analysis which has investigated IS and PCS serum levels as biomarkers of CV events and all-cause mortality in CKD patients [24]. The authors of this study selected 11 out of 155 articles, with a total of 1,572 patients, using strict inclusion criteria and considering the risks of all-cause mortality and CV events as primary outcome measures. The results indicated that both elevated free PCS and IS were significantly associated with increased mortality in CKD patients, while elevated free PCS levels, but not IS levels, were significantly associated with an increased risk of CV events in CKD patients [24]. The latest literature studies are mainly focused on dialysis patients and on the removal strategies of these protein-bound UTs. It can easily be inferred that the toxicity of elevated serum levels of IS and PCS is exacerbated in these patients [64]. Indeed, in a recent national multicenter prospective cohort study including 394 incident dialysis patients [64] the study of the free serum levels of 4 uremic metabolites (PCS, IS, hippurate and phenylacetylglutamine) has shown a correlation between higher PCS levels and a greater risk of CV mortality and first CV event. Moreover, patients with the highest levels of the four combined solutes had a greater risk of CV mortality. However, it is worthwhile reporting that a very recent study on a subgroup of 1,273 HD patients from the HEMO trial has revealed no association between IS, PCS, phenylacetylglutamine and hippurate and cardiac death, sudden cardiac death and first cardiovascular event [65], although the association between IS and PCS and higher risk of cardiac death and sudden cardiac death has been shown to be significant in patients with low serum albumin levels. There are two points, raised by the same authors that recommend some caution when interpreting these results: the first point being that in the study in question the total levels of PCS and IS were considered, while the free levels of UTs may be better indicators of their potential toxicity, as tissues are exposed to them. The second point is that, given the strong scientific rationale supporting the cardiotoxicity of IS and PCS, their role in uremic toxicity and CV risk cannot be excluded and that only a RCT could provide a strong demonstration [65].
TMAO is a downstream metabolite of dietary choline, phosphatidyilcholine and L-carnitine, produced by the gut microbiota after their conversion into TMA and oxidized to TMAO in the liver [66]; high levels of its dietary precursors lead to an increase in TMAO circulating levels [67]. TMAO is normally cleared by the kidneys and, as a result, it is elevated in CKD patients [68]; unlike the other UTs, it has no high affinity with serum albumin, being efficiently removed by the dialytic treatment [69]. Accumulating evidence in the literature indicates a role for elevated TMAO levels as an indicator of increased risk of all-cause mortality in CKD patients [68]. Very recently, the group of Heianza has conducted a systematic review and meta-analysis of prospective studies on the association of TMAO and its precursors L-carnitine, choline and betaine with major acute cardiovascular events (MACE), demonstrating that elevated levels of these metabolites are associated with MACE and all-cause mortality, independently of other traditional risk factors [66].
TMAO could modulate the CV risk by many potential mechanisms. An interesting study examined the gut microbiota-dependent metabolism of L-carnitine to produce TMAO in both rodents and humans, demonstrating a role for TMAO in atherosclerosis pathogenesis [67]. The authors showed that one pro-atherosclerotic mechanism observed for TMAO is the reduction of reverse cholesterol transport (RCT) through gut microbiota-dependent mechanisms in vivo, but the molecular mechanisms whereby TMAO inhibits RCT are not entirely clear [67]. Furthemore, other studies focused on the involvement of TMAO in atherosclerotic plaque development in animal models, showing a critical role for TMAO in increasing macrophage cholesterol accumulation and foam cell formation [70], enhancing human platelets hyper-reactivity, and enhancing in vivo thrombosis [71] or by a pro-inflammatory action [72].
As already mentioned, the microbial-derived metabolite IAA is another potential link with increased CV risk in CKD. Although less studied, its pro-inflammatory and pro-oxidant effects potentially explain the association with higher mortality and CV events observed in CKD patients; anyway, these effects remain to be explored in vivo [55].
Another microbial-dependent effect influencing CV risk is linked to the modulation of NO availability. Amongst its pleiotropic effects, NO acts as a paracrine vasodilator. It is an important regulator of vascular health and BP, and a reduced NO endogenous production results in endothelial dysfunction and increased CV risk [73]. NO is enzymatically produced by NO synthases in different cell types, particularly in endothelial cells, starting from L-arginine as precursor [74].
Alternatively, NO can be produced by the oral microbiota as a "back-up" system, namely the "enterosalivary nitrate-nitrite-NO pathway" [73]. Following intestinal absorption, dietary nitrate is concentrated in saliva and, by means of a chain of chemical reactions, is therefore reduced to nitrite by bacteria living in the mouth; a portion of the swallowed nitrate is further converted in the stomach into NO and other nitrogen species [74].
Dietary sources of nitrate include green leafy vegetables and beetroot [74]. Indeed, many trials explored the effects of orally administered nitrate on BP regulation and vascular function. They overall demonstrated beneficial and comparable effects of dietary NO sources (beetroot juice) or nitrate salts supplements on BP, vascular endothelial function and platelet aggregation, in both healthy conditions and in pathological contexts such as hypertension, diabetes, and CV risk [74,75]. On the other side, the observation that the use of antibacterial mouthwashes is accompanied by an elevation of systolic BP provides a demonstration of the important and physiological role of oral microbiota metabolism in maintaining vascular health [76]. Given that microbiota plays an important role in the nitrogen metabolism, an intriguing theory has been elaborated by Briskey and coworkers, suggesting that a balanced microbiota is essential to guarantee a fine regulation between the reduction/oxidation system in the human body and that its imbalance could contribute to CKD etiology [4]. The authors explain that nitrification and denitrification bacterial reactions, well studied in agricultural context, are largely unexplored in the context of human physiology, although they likely occur in the human body by oral and gut microbiota, and contribute to the regulation of the nitrogen cycle similarly to what happens in natural ecosystems. Oral microbiota composition, nitrate, nitrite, ammonia and oxygen concentrations are all factors influencing each other in a cause-consequence relationship still to be extensively studied, but certainly needing investigations.
Additionally, novel insights come from a suspected action of the gut microbiota in the BP regulation and its involvement in hypertension and salt sensitivity [77]. Like in gut-kidney axis, probably also in gut-heart axis a bidirectional relationship exists. Provided that the kidney plays the major role in BP regulation, some pioneering evidence suggests that hypertension is somehow linked and/or regulated by microbiota and its metabolites, although this research topic is quite new. Some human and animal studies indeed showed that hypertension and atherosclerosis associate with a decrease in microbiota diversity and richness, alteration in its composition and in the microbial metabolic profile [6]. This is a new field, comprising some studies carried on in animal models and a few association studies, while researches focusing on the mechanistic explanation of the microbiota effect on BP are currently lacking. Anyway, some hypotheses have been proposed. Gut level, such as Olfr78, Gpr41 or Gpr43, antagonistically results in raised or decreased BP [78]. In addition, another SCFA-mediated effect has been hypothesized by Santisteban et al., acting on the modulation of the brain-gut-bone marrow axis, in which an enhanced sympathetic activity leads to hematopoietic stem cell mobilization, gut inflammation and raised BP [9]. Another link with CV system, still to be demonstrated, could be mediated by sulfur amino acids microbial metabolism leading to H2S production, suspected to be linked to BP regulation [79].
Moreover, evidence from animal studies suggests that the gut microbiota could be involved in determining salt sensitivity, with a mechanism still to clarify but probably mediated again by SCFA [77]. Indeed, some murine studies showed differences in the fecal microbial composition between salt-sensitive (SS) and salt-resistant (SR) mice, and it was demonstrated that transplantation of SR cecal content to SS was able to confer a sustained increase in BP and a shorter lifespan, shedding a new light on the possibility that salt-sensitivity and hypertension could be transmissible by means of the gut microbiota [7,77].

The impact of nutritional management on CV risk
The protective effects of DASH diet or Mediterranean diet on the risk of incidence of CKD are well assessed in the literature [80,81]. The renal and CV protective nature of these two dietary patterns have been ascribed to the anti-proteinuric and anti-hypertensive effects mainly due to the limitation of salt intake and to the higher content of plant-based food [80][81][82][83].
In renal insufficiency, the total body sodium pool increases, causing hypertension and edema. This is due to the glomerular filtration rate (GFR) reduction, to hyperactivity of the sympathetic system and RAAS. A high salt intake increases the amount of urinary protein excretion [84], a major risk factor for the progression of both kidney and CV damage. Conversely, reducing salt intake reduces urine protein excretion by about 20% [85]. Salt restriction has been widely used in renal patients, often representing the only dietary intervention in the clinical practice. Low sodium diet is effective in lowering BP and proteinuria, and these effects are additive to the inhibition of RAAS and [86]. On the contrary, unrestricted sodium diets impair and blunt the favorable effects of RAAS inhibitors on proteinuria and CKD progression, both in diabetic and in non-diabetic cohorts [87,88]. In advanced stages of CKD, salt restriction is mandatory to reduce the risk of fluid retention, and consequently of edema, heart failure or uncontrolled hypertension. So, consensus exists about the usefulness of salt restriction in CKD patients [89], although often difficult to achieve in clinical practice [90]. Instead, the history of plant-based dietary manipulation in CKD patients is quite various and often contradictory.
The DASH or Mediterranean diets are healthy dietary patterns proven to be useful as a prevention strategy for CKD in a population-based setting. Moreover, specific renal diets aimed at controlling the protein, sodium and phosphorus loads are needed in advanced stages of renal insufficiency.
The nutritional treatment in CKD is necessary to slow down the pathology progression, to control its comorbidities and to prevent protein energy wasting [91,92]. The benefits of a controlled diet in CKD come from the reduction of the retention of UTs, sodium, phosphate and fixed acids [93], mainly derived by the catabolism of dietary proteins [94]. For this reason, nutritional treatments in renal patients have historically been focused on dietary protein restriction [91].
The new advances in our understanding of the central role of the gut microbiota in renal pathology and CV damage should be considered in the design of innovative renal diets, aimed at a new goal: modulating the gut microbiota, a non-traditional and modifiable risk factor. Dietary manipulation has indeed the potential to affect intestinal microbiota composition, metabolism and activity, and to control the CV risk by decreasing the production of the UTs described above.
Gastrointestinal bacteria can ferment complex carbohydrates (saccharolytic pathway) or use amino acids as an alternative fermentation substrate (proteolytic pathway): changes in the microbiota cause changes in metabolites production, which in turn can affect human pathophysiology [8]. In particular, an unbalance towards proteolytic metabolism may be deleterious. In this condition, an increase in the production of many toxic metabolites, namely amines, indoles, phenols, hydrogen sulfide occurs. These are the reasons why a gut is considered to be "healthy" when its metabolism is mainly saccharolytic. In a chronic renal failure setting, urea concentration increases in extracellular as well as in the intestinal fluids. This alters colonic pH and promotes the overgrowth proteolytic microbial species, leading to an increased generation of UTs with a negative impact on CV risk [8]. In this context, the adoption of a targeted diet can shift the saccharolytic/proteolytic balance in a favorable way. Many plant-based food belonging to Mediterranean or DASH dietary schemes carry a considerable amount of fiber with prebiotic activity. This fiber remains undigested until the terminal gastrointestinal tract, where it selectively promotes the growth of probiotic, saccharolytic bacteria, leading to a decrease in protein-bound UTs [8,95,96].
The current nutritional guidelines in CKD do not take into consideration microbiota modulation. In fact, in order to control potassium intake, they often foresee a dramatic restriction of many fiberrich foods such as fruits, vegetables and legumes, thus worsening potential constipation conditions. Actually, the risk of hyperkalemia is increased in CKD patients. This is due to the reduced GFR, though a very effective increase in K excretion by remaining nephrons occurs even in advanced CKD stages [97]. The main concern is the use of RAAS inhibitors, b-blockers, heparins that are frequently used, for their proven efficacy, in CKD and heart failure patients. In these settings, avoidance of high potassium intake is mandatory as a part of the strategy for hyperkalemia prevention. The upcoming new potassium binders may be a potential good option for less limitation of dietary potassium load, certainly to be further studied with randomized controlled trials aimed to define optimal management strategies [98].
But two aspects should be considered. First, vegetables contribute to an overall alkalizing effect: most fruit and vegetables are known to produce alkali when metabolized, contributing to acid neutralization [83,99]. Second, fiber intake has a positive effect on constipation. Constipation has recently been demonstrated to be a risk factor for CKD development and progression [100], for an unknown mechanism probably linked to uremic toxins retention and gut dysbiosis. This two effects of a plant-based diet can potentially and favorably affect potassium metabolism reducing the risk of hyperkalemia. Even in dialysis patients, the risk of hyperkalemia is linked to constipation and not to the amount of dietary potassium load [101]. Moreover, it is important to consider that phosphate from vegetable sources is poorly absorbed by the human intestine, thus limiting the effective body burden. So, in combination with protein restriction, that has already been demonstrated to be beneficial by reducing the proteolytic fermentation at colonic level, and beyond the expected reduction in terms of onset and severity of uremic symptoms and complications [83,102,103], fiber intake should be encouraged in renal diets.
Moreover, the introduction of plant food results in beneficial effects also at CV levels: the DASH diet reduces BP at all sodium levels thanks to its content of vegetable matrices [104] and, as already reported above, an increased intake of green leafy vegetables and beetroot, rich in nitrate, favors oral microbiota-mediated NO generation [105]. This becomes particularly critical in HD patients, notoriously exposed to higher CV mortality. Here, the dialytic treatment per se seems to be responsible for a persistent depletion of circulating nitrite and nitrate, leading to a chronic reduction of steady-state NO levels [106]. Given the important action of their balance as an endothelium-independent redundant system for NO supply, and the experimental demonstration that their insufficient dietary intake promotes or accelerate disease, the role of nitrate and nitrite as dietary nutrients and not as harmful compounds, as they are currently considered, has been recently suggested for its benefits [75].
In conclusion, fiber intake provided by plant-based foods, combined with sodium and protein restriction should be encouraged in the dietary management of CKD. This kind of management would result in a combined favorable action at CV level through the modulation of glomerular hemodynamics, acid-base metabolism, effective phosphate load, lipid profile, systemic hypertension and NO generation, bowel transit, microbiota composition and metabolism, UTs levels, all affecting CV risk (Fig. 1).

Focus on microbiota -Perspectives: prevention and treatment Prevention: lifestyle changes to prevent CKD pathologies by taking care of microbiota
There are some clues indicating that microbiota could play a role in the aetiology of CKD and hypertension [7]. Indeed, a dysbiotic microbiota could potentially represent a risk factor for the development of CKD after acute injury [5] or following a dysregulation of the nitrification/denitrification balance [4]. Moreover, reduced bowel transit and severe constipation have been associated with an increased risk of CVD, incident CKD/ESRD and progressive eGFR decline [100]. All these pieces of evidence encourage the adoption of preventative life-style measures to keep a healthy and balanced microbiota, especially in the presence of other risk factors. Actually, the prevention could start earlier, before birth. There are proofs that low-birth weight and unbalanced nutrition in pregnant women account for risk factors for the development of CKD, CVD and metabolic disease in adult age [107]. In particular, since nephrogenesis is concentrated in the last trimester of pregnancy, pre-term birth and low-birth weight represent a risk factor both for developing CKD after an acute insult, and for hypertension, due to the low nephron number [107,108]. A major interaction and communication between different medical branches (gynaecology, neonatology and nephrology) in order to encourage a balanced nutrition during pregnancy, together with natural delivery and breastfeeding, beneficial for new-borns microbiota development, should be promoted as a preventative strategy for keeping under control the risk of developing CKD and hypertension [107].

Treatment: postbiotics and bacteriotherapy
The ability of the gut microbiota to respond to environmental changes offers the potentiality to act on it in order to control the progression of CKD and to reduce CV risk factors. The above dietary management could be complemented by the integrating a probiotic, prebiotic, synbiotic or the novel postbiotic approach. Indeed, some trials have tested the use of probiotics, prebiotics or synbiotics in CKD, reporting positive effects on uremic toxins, blood urea nitrogen and uric acid reduction, although none of these approaches has been demonstrated to be able to decrease UTs levels back to physiological ones [109]. Apart from this, the other limitations of these studies are the small sample size, the use of surrogate outcomes and the short study duration. It would be interesting to test on a longer timeframe if this approach could be effective in slowing down CKD progression and reducing CV events. In CVD context, the use of probiotics beverages and foods has been tested in some trials, demonstrating its efficacy in lowering systolic and in some case also diastolic BP, both in healthy and in hypertensive subjects [7], while the protective activity of prebiotics on hypertension and CV risk is mainly indirect, given the ability of some fiber-rich foods to improve the lipid and glycaemic frame [79,110]. Nevertheless, in a very recent study our group has demonstrated an improvement in endothelial function measured by FMD after beta-glucans supplementation in healthy subjects [96].
A new perspective comes from the postbiotic approach, that is the possibility of intervention by nutritional supplementation with beneficial microbial end-products with a potent immunomodulating activity [111], especially in subjects with altered intestinal barrier integrity and ongoing intestinal inflammatory status [112]. Animal research has experimented the postbiotic approach as a safer alternative to antibiotics in poultry industry, with encouraging results on increased in vitro inhibitory activity against pathogens [113], chickens weight gain, feed efficiency, microbial composition, ileal inflammatory profile and volatile fatty acid production [114,115]. In humans, this approach is far  [111]. In a mouse model of obesity-induced insulin resistance, a muramyldipeptide derived from bacterial cell wall showed insulin-sensitizing effects [116], while a Lactobacillus-derived postbiotic showed efficacy in counteracting the inflammatory response associated to intestinal bowel disease or to Salmonella infection of healthy tissue [112]. Postbiotics have been suggested as a promising therapeutic option in children necrotizing enterocolitis [117].
Bacteria-derived peptides like Serine-Threonine peptide (STp) have been shown to restore in vitro a mature profile of gut dendritic cells in ulcerative colitis [118], while Lactobacillus reuteri DSM 17938 cell-free supernatant has been shown to modulate the gut dendritic cell phenotype and function [119]. Additionally, a Lactobacillus fermentum BGHV110 strain (HV110)-derived postbiotics have been able to stimulate in vitro a protective hepatic cells autophagy in response to high paracetamol doses [120]. The postbiotic approach has never been tested in CKD and certainly needs to be clinically validated. SCFAs, for example have the potential to exert their beneficial effects on colonic microenvironment, epithelial integrity, limitation of growth of potentially pathogenic species, and inflammation [121].  [122]. There are some findings supporting the possibility to expand the field of application of this procedure to other clinical settings involving an alteration of the gut microbiota, including CKD and CV damage, but they would need extensive clinical research [123]. Contexts in which the mucosal immunity plays a key role in the aetiology and progression of the renal pathology and microbiota composition seem to be associated with pathology progression, such as IgAN [124][125][126] and lupus nephritis [127], could represent the ideal candidates for future FMT protocol settings [128]. The association between IgAN and the occurrence of inflammatory bowel disease [129] and the tuning of gut permeability and of the immune responses associated to lupus nephritis by microbiota manipulation [130] suggest that future efforts in this direction could be successful.

Conclusions
In conclusion, there is an important body of literature highlighting a role of the gut and oral microbiota in both aetiology and in progression of CKD, hypertension and CV damage. Certainly, the main gap to be filled in the literature concerns the elucidation of the underlying mechanisms and of the cause-effect relationships: is a dysbiotic microbiota one of the concurrent causal agents of hypertension, CV damage and progressive kidney failure or do these pathologies influence the composition and function of the microbiota, leading to dysbiosis? And, in both cases, what are the causal mechanism leading to the development of the diseases/dysbiosis? In the context of CKD some aspects have been clarified. It is by now well-established that the uremic state is responsible for an alteration of colonic environment, leading in turn to the modification of the microbiota composition and the increase in uremic toxin production that, together with the main mechanism, i.e. their reduced renal clearance, result in increased UTs circulating levels. However, the opposite relationship remains to be explored: if and how an altered microbiota can be responsible for inducing kidney loss of function.
Even so, research on the association between hypertension and a specific, dysbiotic microbial setting is still in its infancy and certainly needs to be expanded, even if there are some clues pointing towards a bidirectional communication of the gut-heart axis. In this field the need is felt to move from studies on animal models to human studies, and to switch from pure association studies to the investigation of the basal cause-effect relationships linking CKD, hypertension and microbiota.
Actually, in CKD the role of the microbiota in concurring to CV damage is emerging in the literature through a dual mechanism: firstly through altered function and secondly through increased production of harmful microbial metabolites. Importantly, microbiota and microbial derived UTs are emerging as potential modifiable risk factors of CV damage, provided that the combined action of diet and supplementation is capable of reducing UTs to the levels found in non-CKD subjects. So far no trial has been successful in this, albeit a reduction in UTs production rate is expected to be beneficial under any circumstances. In this regard, the microbiota represents a strategic factor to be considered in the clinical management of CKD patients.
Even so, new insights into the mechanistic aspects of the intricate pathophysiologic network of microbiota, kidney and heart will allow for designing specific, microbial-oriented therapies, to be integrated or associated to nutritional treatments. This combined approach could represent an effective strategy to i) postpone the need for dialysis, ii) support pre-emptive transplantation much more in the years to come, and iii) improve the life expectancy and the overall quality of life of our patients.

Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.  Colon bacteria: chemical deamination, transamination, and decarboxylation of tyrosine and phenylalanine  p-cresol Liver and enterocytes: sulfation of p-cresol PCS PCS is transported through OAT1 and OAT3 by proximal renal tubular cells and secreted in the urine IS [5,27] Dietary proteins (Tryptophan)