Avenues for post-translational protein modification prevention and therapy

Abstract

Non-enzymatic post-translational modifications (nPTMs) of proteins have emerged as novel risk factors for the genesis and progression of various diseases. We now have a variety of experimental and established therapeutic strategies to target harmful nPTMs and potentially improve clinical outcomes. Protein carbamylation and glycation are two common and representative nPTMs that have gained considerable attention lately as favorable therapeutic targets with emerging clinical evidence. Protein carbamylation is associated with the occurrence of cardiovascular disease (CVD) and mortality in patients with chronic kidney disease (CKD); and advanced glycation end products (AGEs), a heterogeneous group of molecules produced in a series of glycation reactions, have been linked to various diabetic complications. Therefore, reducing the burden of protein carbamylation and AGEs is an appealing and promising therapeutic approach.

This review chapter summarizes potential anti-nPTM therapy options in CKD, CVD, and diabetes along with clinical implications. Using two prime examples—protein carbamylation and AGEs—we disucss the varied preventative and therapeutic options to mitigate these pathologic nPTMs in detail. We provide in-depth case studies on carbamylation in the setting of kidney disease and AGEs in metabolic disorders, with an emphasis on the relevance to reducing adverse clinical outcomes such as CKD progression, cardiovascular events, and mortality. Overall, whether specific efforts to lower carbamylation and AGE burden will yield definitive clinical improvement in humans remains largely to be seen. However, the scientific rationale for such pursuits is demonstrated herein.

1. Introduction

Proteins in the human body, during both health and disease, are exposed to a variety of chemical reactions capable of altering their active properties. Post-translational modifications (PTMs) of proteins are capable of altering protein charge, structure, and function in various physiological and pathophysiological conditions. The diversification of protein structures by PTMs provides an important means of protein regulation in human health. For example, phosphorylation is commonly required to activate the catalytic function of various enzymes (Walsh et al., 2005). Non-enzymatic post-translational modifications (nPTMs) of proteins are typically irreversible chemical reactions between electrophilic small molecules (endogenously or exogenously derived) and nucleophilic amino acid side chains or the N-terminus of proteins (Harmel and Fiedler, 2018). Without classical enzyme catalysis, nPTMs are usually spontaneous and not highly regulated. In recent years, multiple nPTMs have been found to play key roles in molecular aging and dysfunction (Gorisse et al., 2016), and have been implicated in various disease states such as chronic kidney disease (CKD) (Kalim et al., 2014), cardiovascular disease (CVD) (Verbrugge et al., 2015), diabetes mellitus (DM) (Singh et al., 2001), cancer (Sharma et al., 2019) and neurodegenerative disorders (Schaffert and Carter, 2020).

While discussion of the potential prevention and therapy of all known pathologic nPTMs is beyond the scope of any single review, there are clear areas that have been the primary focus of recent literature. Specifically, the nPTMs carbamylation (particularly in the context of kidney diseases) and glycation (particularly in the context of DM) have gained considerable attention. It is understandable that these entities have garnered such focus given the high prevalence of their commonly associated medical conditions in the general population and because therapeutic pathways appear primed for clinical study. Of note, one of the most common causes of CKD across the globe is DM (Alebiosu and Ayodele, 2005), and much of the morbidity and mortality associated with CKD and DM ultimately relates to cardiovascular causes (Sarnak et al., 2003). Thus, CVD, CKD and DM are often interlinked with each other, collectively acknowledged as cardio-renal-metabolic disease (Rangaswami et al., 2020). This review will look to potential avenues of carbamylation and glycation prevention and therapy across this broad axis. There are still many other PTMs, as noted above, with therapeutic potential in the early investigational stages and, therefore, considered beyond the scope of this review.

1.1. Overview of the carbamylation reaction

Carbamylation is defined as the non-enzymatic reaction between cyanate and amino or sulfhydryl groups of amino acids, peptides, or proteins (Delanghe et al., 2017). As shown in Fig. 1, urea or otherwise derived cyanate (which is in equilibrium with its tautomer and reactive form isocyanate), can spontaneously react with select sites on proteins. The net result of these reactions, which we are calling “carbamylation”, is the addition of a “carbamoyl” moiety (−CONH₂) to a functional group. It must be noted that several authorities accurately state that the reaction in question is more correctly deemed “carbamoylation” and, in fact, “carbamylation” denotes a different chemical reaction (the reversible interaction of CO₂ with α- and ε-amino groups of proteins) (Jelkmann, 2008). However, over the past decades, the biomedical literature has largely been calling the reaction involving a carbamoyl moiety, “carbamylation”, so we will continue in this manner for convention.

Figure 1.

Schematic representation of chemical genesis of protein carbamylation and potential therapeutic targets.

As kidney function declines, circulating levels of urea and cyanate increase in the human body, and thus so does carbamylation load (Pietrement et al., 2013; Vanholder et al., 2018). An alternative source of cyanate is thiocyanate or cyanide which can derive from diet or smoking and be catalyzed by myeloperoxidase (MPO), an enzyme present in the granules of neutrophils, monocytes and macrophages, at inflammatory sites (Delporte et al., 2018; Wang et al., 2007). Yet in the face of declining kidney function, the increase in urea is felt to be the dominant driver of carbamylation (Kalim et al., 2016, 2013).

Carbamylated proteins undergo structural and functional changes, leading to modified biological activity (Jaisson et al., 2011). Accumulation of carbamylated proteins has been described as a hallmark of molecular aging, and such proteins are involved in the pathogenesis of several conditions including atherosclerosis (Jaisson et al., 2015; Speer et al., 2014; Sun et al., 2016), thrombus formation (Binder et al., 2017; Holy et al., 2016) and renal fibrosis (Gross et al., 2011) for a few examples. Protein carbamylation also plays a role in the generation of abnormal tau protein deposition involved in the development of Alzheimer’s disease (Farías et al., 1997).

1.2. Overview of the non-enzymatic glycation reaction

Glycation is defined as the non-enzymatic reaction between reducing sugars or sugar derivatives and amino groups of amino acids, proteins, lipids or nucleic acids (Simm, 2013). Advanced glycation end products (AGEs) are a heterogeneous class of molecules or compounds that are derived from this process (Vlassara, 1997; Vlassara and Palace, 2003). With initial observations made by the French chemist Louis-Camille Maillard, it is common to see reference to the relevant reaction as the Maillard reaction (Maillard, 1912). This complex reaction involves a series of steps in which reducing sugars undergo non-enzymatic reactions that produce reactive carbonyl compounds and the subsequent glycoxidation of proteins, lipids and nucleic acids. AGEs are ultimately yielded when the glycation targets are rearranged to form stable heterogeneous compounds (Singh et al., 2001). The receptor for AGEs (RAGE) is a multi-ligand pattern-recognition receptor expressed in different cell types (Hudson and Lippman, 2018) and interaction of AGEs with RAGE can activate complex intracellular signaling cascades. RAGE independent effects of AGEs include change in molecular conformation and clearance, alteration of enzyme activity, and interference with receptor recognition (Aronson and Rayfield, 2002).

Notably, as DM is the leading cause of CKD, it is not surprising that the two nPTMs most closely related to these disease processes (non-enzymatic glycation and carbamylation) can co-occur. In vitro and animal studies suggest these two reactions have a competitive relationship with each other as they often target the same protein sites (Nicolas et al., 2019, 2018). However, at more intensive carbamylation states, competitive glycation appears less significant (Nicolas et al., 2019, 2018).

2. The relation between nPTMs and adverse clinical outcomes

As reviewed elsewhere in this series and briefly noted here, carbamylation and non-enzymatic glycation have been clearly and independently associated with adverse clinical outcomes in humans. A detailed accounting of all clinical association studies related to these nPTMs is beyond the scope of this review and it is important to note the difficulty in assigning weight to such diverse studies with varying sample sizes, study design, and even nPTM assay methods. Nevertheless, Table 1 highlights select consistent examples that provide compelling evidence that these nPTMs are associated with adverse clinical outcomes.

Table 1.

Patient-oriented studies linking nPTMs to adverse clinical outcomes

PTM	Biomarker (Measurement method)	Study design	Study population	Associated clinical outcomes	References
Carbamylation	Protein-bound homocitrulline (HPLC–MS/MS)	Case-control	550 patients undergoing cardiac catheterization	MACE	(Wang et al., 2007)
		Cohort	347 patients on HD	Mortality	(Koeth et al., 2013)
		Cohort	111 patients with chronic systolic heart failure	Mortality; Cardiac transplantation	(Tang et al., 2013)
	Carbamylated albumin (HPLC–MS/MS)	Cohort	187 patients on HD	Mortality	(Berg et al., 2013)
		Cohort	158 patients on HD	Erythropoietin resistance; mortality	(Kalim et al., 2013)
		Case-control	366 patients on HD	Mortality	(Kalim et al., 2016)
		RCT post-hoc analysis	1,161 patients on HD with diabetes	Cardiovascular mortality; Sudden cardiac death	(Drechsler et al., 2015)
		Case-control	300 patients with CKD	CKD progression; Mortality	(Kalim et al., 2020)
Glycation	Pentosidine (ELISA)	Case-control	141 patients with heart failure	Cardiovascular mortality; Re-hospitalization	(Koyama et al., 2007)
	Pentosidine (HPLC–MS/MS)	Cohort	746 patients with CKD	Mortality	(Machowska et al., 2016)
	CML (ELISA)	Cohort	81 patients with CAD	CAD progression	(Basta et al., 2008)
		Cohort	154 patients on HD	Mortality	(Wagner et al., 2006)
	Skin AF (AF reader)	Case-control	109 patients on HD	CVD Mortality	(Meerwaldt et al., 2005)

Abbreviations: AF, autofluorescence; CAD, coronary artery disease; CKD, chronic kidney disease; CML, Nε-carboxymethyl-lysine; ELISA, enzyme-linked immunosorbent assay; HD, hemodialysis; HPLC–MS/MS, high-performance liquid chromatography with tandem mass spectrometry; MACE, major adverse cardiac event; PTM, post-translational modification; RCT, randomized controlled trial.

2.1. Protein carbamylation and adverse clinical outcomes

The motivation for clinical outcome studies related to carbamylation has strong mechanistic underpinnings. Classic experiments have shown that when certain proteins are incubated with urea, a modified form of the protein accumulates that is less positively charged, and urea exposure results in lysine within the protein becoming irreversibly modified into N (6)-carbamoyl-L-lysine (what is known as homocitrulline). This carbamylation reaction yields a relatively stable moiety which does not readily dissociate or oxidize and proteins can accumulate these modifications on their N-terminal α–amino groups or the ε-amino groups of lysine side chains throughout their lifespan. A major chemical effect of carbamylation on protein amino groups is the neutralization of a positive charge which in turn alters ionic protein-water interactions on the protein surface (Beswick and Harding, 1987; Jaisson et al., 2018; Wistow et al., 1983). This disturbance can destabilize secondary and tertiary protein structures resulting in subsequent conformational changes. Dozens of studies have now shown results implicating carbamylation in changes of protein charge (Bobb and Hofstee, 1971; Legendre et al., 1998), conformation (Beswick and Harding, 1987; Monhemi and Tabaee, 2020; Nowicki and Santomé, 1981), and stability (Fazili et al., 1993), with consequent alterations in enzyme and hormone activity (De Furia et al., 1972; Oimomi et al., 1987; Shaw et al., 1964; Van Lente et al., 1986; Veronese et al., 1972), binding properties (Dengler et al., 1992; Lee and Gibson, 1981; Weisgraber et al., 1978), receptor-drug interaction (Erill et al., 1980; Smyth, 1967), and cellular expression and responses (Balion et al., 1998; Garnotel et al., 2004; Ha et al., 2010; Jaisson et al., 2007; Lane and Burka, 1976; Maddock and Westenfelder, 1996). These studies are just a few examples which highlight plausible mechanisms by which hyper-carbamylation could contribute to adverse clinical outcomes.

A few examples of specific pathophysiological pathways shown to accelerate due to carbamylation include atherosclerosis, vascular calcification, and fibrosis of kidney parenchyma. Carbamylation of low density lipoprotein (LDL) can enhance its atherogenic properties partly by decreasing its binding to the LDL-receptor (Wang et al., 2007) and by preventing its clearance from circulation, a phenomenon seen in both animal models of chronic renal failure (Hörkkö et al., 1992; Shapiro, 1993), as well as in uremic patients (Apostolov et al., 2010; Hörkkö et al., 1995, 1994; Shah et al., 2008). Carbamylated LDL has also been shown to readily bind macrophage scavenger receptors, facilitating foam cell formation and inflammatory signaling (Apostolov et al., 2009, 2007; Wang et al., 2007). Moreover, carbamylated LDL can induce endothelial cell apoptosis (Apostolov et al., 2007; Ok et al., 2005) and stimulate vascular smooth muscle proliferation (Asci et al., 2008) while carbamylated collagen increases release of matrix metalloproteinases by monocytes, possibly stimulating remodeling of the extracellular matrix (Garnotel et al., 2004; Jaisson et al., 2006).

Carbamylated proteins at concentrations present in uremia have similarly been shown to activate glomerular mesangial cells to a profibrogenic phenotype and stimulate collagen deposition (Shaykh et al., 1999). Cyanate exposure strongly inhibits the collagenase activity of human and rat mesangial cell MMP-2 (Gross et al., 2011). In general, it is suggested that carbamylation of collagen or enzymes involved in extracellular-matrix remodeling can disrupt the balanced remodeling of extracellular proteins and enhance fibrosis (Jaisson et al., 2007; Kraus et al., 2001). Additionally, carbamylated albumin results in renal tubular cell damage and peritubular fibrosis via stimulation of TGF-β, EGF, NF-κB and endothelin-1 in the amphibian kidney (Gross et al., 2011). Carbamylation of high-density lipoprotein (HDL) can impact migration, angiogenesis, and proliferation in endothelial cells (Holzer et al., 2011; Sun et al., 2016), while other studies suggest cyanate itself has direct pathologic effects on endothelial cells (El-Gamal et al., 2012; Xiao et al., 2001). Similar mechanistic links to endothelial dysfunction can be seen through carbamylation’s contribution to oxidative stress. For example, S-carbamylation of free cysteine and glutathione sulfhydryl groups as well as of hydrogen sulfide interferes with their antioxidant functions (e.g. free radical scavenging) (Praschberger et al., 2013; Schreier et al., 2011). Most recently, carbamylation’s mechanistic role in vascular disease and calcification has been elucidated through in vitro and in vivo studies of carbamylated sortilin, elastin, uromodulin, and mitochondrial proteins all resulting in increased vascular pathology (Alesutan et al., 2021; Doué et al., 2021; Jankowski et al., 2021; Mori et al., 2018).

There are a few biomarkers that have been used to assess and quantify carbamylation burden, including homocitrulline (Jaisson et al., 2016, 2012), carbamylated albumin (Berg et al., 2013), carbamylated hemoglobin (Balion et al., 1998) and carbamylated LDL (Apostolov et al., 2005). Most of the biomarkers are quantified using high-performance liquid chromatography (HPLC) with online tandem mass spectrometry (MS/MS). HPLC-MS/MS confers robust measurement with high precision and reproducibility. For example, the assay of carbamylated albumin quantified using this method has excellent linearity with a coefficient of variation of 4.2% (Berg et al., 2013). However, their use in routine clinical practice is currently limited by time-consuming techniques and expensive equipment (Jaisson et al., 2018). If the clinical utility of carbamylation measures eventually becomes established, the development of efficient, reliable, and cost-effective assays will need to follow.

With such mechanistic underpinnings and available biomarkers, several patient-oriented studies (Table 1) have unsurprisingly shown that carbamylation is an independent risk factor for adverse clinical outcomes including CVD, CKD progression, and mortality in patients with CKD as well as normal kidney function (Kalim et al., 2020, 2016; Tang et al., 2013; Wang et al., 2007). Moreover, carbamylation was associated with CVD and mortality in end-stage renal disease (ESRD) as well (Berg et al., 2013; Drechsler et al., 2015; Koeth et al., 2013).

2.2. AGEs and adverse clinical outcomes

Similarly, the accumulation of AGEs alters the structure and function of proteins, contributing to the pathogenesis of metabolic diseases such as diabetes through a variety of mechanisms (Chaudhuri et al., 2018). As an example, after AGEs bind to RAGE, NF-kB and MAPK signaling pathways are activated, leading to inflammation and endothelial dysfunction (Sadik et al., 2012). AGEs are also implicated in the pathogenesis of neurodegenerative disease (Münch et al., 2012); for example, AGE formation has been found at the periphery of Lewy bodies in patients with Parkinson’s disease (Castellani et al., 1996).

Although markers for glucose control such as hemoglobin A1c are well-established and commonly used in clinical practice (Speeckaert et al., 2014), the biomarkers used to indicate and assess the burden of AGEs based on downstream products of glucose metabolism are only emerging (Chaudhuri et al., 2018). Several patient-oriented studies (examples in Table 1) have shown that AGEs measured by various biomarkers are independent risk factors for adverse clinical outcomes including CKD progression, CVD, and mortality in patients with or without diabetes mellitus (Koyama et al., 2007). Tissue AGE accumulation reflected by skin autofluorescence was associated with vascular calcification (Wang et al., 2014) and predicts mortality in CKD (Machowska et al., 2016) and ESRD patients (Meerwaldt et al., 2005; Wagner et al., 2006).

3. Prevention and Therapy

The identification and characterization of these harmful, mechanistically relevant, nPTMs, set them up as promising therapeutic targets in various disease states, especially CVD, CKD, and DM. There is an increasing body of evidence suggesting that protein carbamylation and AGEs may serve as potential preventative and therapeutic targets to reduce the morbidity and mortality from the relevant cardio-renal-metabolic diseases.

3.1. Prevention and therapy targeting carbamylation

Compelling evidence from animal data and epidemiological studies discussed above have suggested carbamylation may play an important role in a plethora of biochemical alterations with clinical manifestations in renal disease. Given carbamylation is established as an independent mechanistically relevant risk factor for adverse clinical outcomes that is readily quantifiable, it is reasonable to hypothesize that targeting carbamylation in CKD and ESRD may improve clinical outcomes. There are a variety of theoretical and practical strategies proposed to reduce the rate and burden of carbamylation (Fig. 1). These pathways include reducing the source of carbamylation’s cyanate—urea—by more intensive dialysis treatment, kidney transplantation, or reducing dietary protein intake; increasing free amino acid scavengers in circulation (which can react with cyanate to prevent the same reaction from occurring on a protein instead); or alternative therapies to block carbamylation’s chemical reaction from occurring. Notably, to date, the common limitations of studies pursuing such strategies include relatively small sample sizes and the lack of hard clinical outcomes (e.g. mortality or CKD progression). In the future, large, randomized control trials to study clinically meaningful outcomes will be needed.

3.1.1. Urea reduction

3.1.1.1. Dialysis intensification and kidney transplantation

Initiation of dialysis in patients with advanced CKD has been shown to reduce protein carbamylation substantially from pre-dialysis advanced CKD levels. This is felt largely due to enhanced urea clearance via dialysis as well likely an improvement in overall uremic symptoms and thus improvement in nutritional status and amino acid balance (Kalim et al., 2018, 2016). However, there is still an excessive carbamylation burden observed in ESRD patients on routine dialysis treatment, raising the question of whether an intensified dialysis treatment prescription could further reduce carbamylation. A prospective study with 53 ESRD patients investigated the impact of in-center nocturnal extended duration dialysis on the burden of carbamylation and cardiac structure and found that compared with standard dialysis (3–4 hours per session, 3 times per week), extended dialysis (7–8 hours per session, 3 times per week) was associated with reduction of carbamylation levels as assayed by carbamylated albumin at 1 year. Such carbamylation reduction correlated with urea reduction and was associated with reduced left ventricular mass measured by cardiac MRI (Perl et al., 2016). Challenges to the approach of intensified dialysis included patient reluctance to undergo additional hours of treatment each week which is often viewed as a significant adverse impact on quality of life. However, with the changing landscape of favoring home dialysis over in-center dialysis under the Advancing American Kidney Health initiative (Himmelfarb et al., 2020), the implementation of extended duration dialysis could become easier.

Looking beyond dialysis, kidney transplantation is the treatment of choice for most patients with ESRD. A successful kidney transplant improves quality of life and reduces the mortality risk for the majority of patients when compared with maintenance dialysis (Wolfe et al., 1999). The impact of kidney transplant on carbamylation has not been robustly studied, but given carbamylation’s close correlation to eGFR, it stands to reason that as eGFR is restored post-transplant, carbamylation load would decrease. This was briefly studied by Han et al, who showed the carbamylated hemoglobin concentration of 6 patients decreased by 19.7% within 2 to 3 weeks after kidney transplantation (Han et al., 1997). However, the true in vivo kinetics were less clear as there was also an average hemoglobin increase of 25% which may have confounded what the direct effects of restored solute clearance had on carbamylation.

3.1.1.2. Dietary modification

Another strategy to reduce urea in pre-dialysis CKD patients is dietary modifications. A Mediterranean diet and a very low protein diet have been shown beneficial in the reduction of urea burden and decreasing carbamylation. In a randomized crossover controlled trial, 60 patients with advanced CKD were assigned to very low protein diet supplemented with keto-analogues or Mediterranean diet or “free diet”; compared to a “free diet”, both the very low protein and Mediterranean diets were associated with decreases in serum homocitrulline levels and such reduction correlated with urea reduction (Di Iorio et al., 2018). Keto-analogues are a very attractive carbamylation therapy as they are amino acid precursors that utilize an amino group from urea to form amino acids (thus reducing urea/ cyanate and increasing carbamylation scavengers simultaneously). However, dietary modification is challenging for many patients to maintain, and strong motivation is required. Moreover, competing dietary demands from comorbid conditions such as diabetes, limited medical practitioner knowledge of specific dietary recommendations, and complicated psycho-socio-economic, behavioral, and cultural factors all can pose challenges to routine dietary interventions. Nevertheless, the evidence suggests that a low-protein diet supplemented with keto-analogues can significantly reduce carbamylation burden. It is not clear what the relative contribution of the low protein diet vs. the keto-analogue supplementation makes to reducing carbamylation and teasing apart these two interventions is an area for future research as is linking such intervention to clinical outcomes. To that end, a contemporary randomized clinical trial in CKD patients recently showed considerable improvements in the outcome of 50% reduction in eGFR or reaching ESRD with a very low protein diet supplemented with keto-analogues (compared to just a low protein diet), yet the underlying mechanisms for this result were not clear and could not be attributed to changes in nutritional parameters alone (Garneata et al., 2016). Perhaps, those who benefited responded to lower urea, restored amino acid concentrations, and subsequently reduced protein carbamylation. Effect sizes might be greater and more consistent if such studies included only subjects with accelerated protein carbamylation were targeted.

3.1.2. Amino acid supplementation

As mentioned above, free amino acids can compete with proteins to react with cyanate and thus reduce protein carbamylation by serving as “scavengers” (Berg et al., 2013). However, patients with advanced CKD and ESRD are often deficient in free amino acids due to various reasons, including poor nutritional status and loss of amino acids in the dialysate (Duranton et al., 2014; Ikizler et al., 1994). There are different means of repleting amino acids, including intravenous, intraperitoneal, and oral/ dietary approaches. Further work is needed to determine the best dose, route, and type of amino acids to reduce carbamylation and maximize the efficacy, feasibility and safety for CKD patients at different stages.

3.1.2.1. Intravenous amino acid supplementation

In a prospective pilot study, 12 hemodialysis patients who received 250cc of intravenous (IV) amino acids which contained 14g of essential amino acids 3 times per week post-dialysis over 8 weeks had significantly reduced carbamylated albumin levels at 4 weeks and 8 weeks, compared to 11 hemodialysis control patients (Kalim et al., 2015). Importantly there were no safety concerns with this approach and notably the dosing did not result in higher blood urea nitrogen levels in the treated individuals (a concern when providing significant amounts of exogenous nitrogen). Such a strategy is appealing as it could potentially also improve proteinenergy malnutrition, another common morbidity in ESRD populations. Another trial to study the effect of IV amino acids in ESRD patients, CarRAAT-2 (Trial of Carbamylation in Renal Disease-Modulation With Amino Acid Therapy; NCT02472834) has been completed with results anticipated soon.

3.1.1.2. Intraperitoneal amino acid supplementation

In a secondary analysis from the IMPENDIA trial (a randomized trial studying the effect of low-glucose peritoneal dialysis solution in DM) (Li et al., 2013), patients maintained on peritoneal dialysis were treated with either icodextrin and intraperitoneal amino acids, or dextrose only peritoneal dialysis solutions. The intraperitoneal amino acid group did not show reduced carbamylated albumin levels (Trottier et al., 2018). However, the study was limited by the small sample size and lack of generalizability due to the inclusion of diabetic patients only. Overall, this study suggested that intraperitoneal amino acid administration might not be the best approach to mitigate carbamylation. An additional finding from this study was that compared to a cohort of similar hemodialysis patients, urea levels on average were higher in the peritoneal dialysis group and this correlated with greater carbamylation levels. It was postulated that perhaps any benefits or advantages of peritoneal dialysis over hemodialysis may be blunted by a greater urea-driven carbamylation load in the peritoneal dialysis group.

3.1.2.3. Oral amino acid supplementation

We can also give oral protein and/or amino acid supplementations (or keto-analogue supplementation as detailed above), both in non-dialysis CKD patients and dialysis patients. For non-dialysis CKD patients, it is not easy to administrate IV amino acids routinely. Therefore, an oral option becomes very appealing. However, the study of oral nutritional and amino acid supplementation has focused mainly on protein and energy homeostasis in dialysis patients (Lacson et al., 2012; Weiner et al., 2014), with no insights into the effects on carbamylation. While such studies have shown mortality benefits in dialysis patients, similar to the benefits seen in low protein diets supplemented with keto-analogues (Garneata et al., 2016), the mechanisms of such benefits remain unclear. As above, effect sizes may be greater and more consistent across studies if therapies targeted carbamylation levels. Additional studies investigating the effect of oral amino acid and/or keto-analogue supplementation on carbamylation burden and clinical outcomes in CKD patients are needed to clarify this.

3.1.3. Investigative paths to attenuate carbamylation reactions

Carbamylation occurs naturally (at low levels in the absence of CKD or other drivers) but can accumulate on proteins with long half-lives. The accumulation rate of carbamylation over time in the skin of different mammalian species with different life expectancies is inversely correlated with longevity (Gorisse et al., 2016). This suggests that some animals may harbor potential protective mechanisms against carbamylation that are yet to be identified. Similar speculation of an ability to guard against carbamylation’s harmful effects has also been pondered in the context of elasmobranch physiology (sharks, rays, and skates). Relative to humans, these animals have extremely high blood urea concentrations (350–500 mM) to maintain osmotic balance with a marine environment yet have several homologous proteins to humans that do not appear to malfunction in such a milieu. How these animals handle high urea levels without enduring adverse effects is not entirely known (Delanghe et al., 2017; Yancey et al., 1982). The underlying mechanism is still speculative and thought to be related to a coexisting protective effect of methylamines including trimethylamine N-oxide (TMAO), betaine, glycerylphosphorylcholine and sarcosine, which can offset the harmful effects of urea on proteins (Ballantyne, 2016; Lee et al., 1991). This is an important area for future research as the understanding of the underlying mechanism carries large clinical implications for any attempts to reduce carbamylation either nutritionally or pharmacologically in CKD patients with high urea levels.

Experimental studies have also shown various agents successful at blocking carbamylation, but these are yet to be assessed in humans. Therapies including ascorbic acid, α-tocopherol, lycopene, and flavonoids all inhibited LDL carbamylation in vitro (Ghaffari and Shanaki, 2010a, 2010b). Eicosapentaenoic acid supplementation in mice decreased mitochondrial protein carbamylation (Johnson et al., 2015), and triclosan similarly inhibited carbamylation (Bright et al., 2018). Aspirin, ibuprofen, and bendazac have all been shown to inhibit carbamylation events that commonly pathologically occur on lens proteins (Crompton et al., 1985; Lewis et al., 1986; Plater et al., 1997; Rao and Cotlier, 1988). Again, these studies are either in vitro or animal studies, and have not been translated to clinical applications and serve as areas for future investigation.

3.2. Prevention and therapy targeting AGE

Chronic hyperglycemia, characteristic of DM, promotes the formation of AGEs and thus, like reducing urea levels to prevent carbamylation, optimal glycemic control is a cornerstone of preventing accelerated AGE formation. Moreover, several medications including antidiabetic medications, blood pressure medications, and statins appear to block Maillard reactions to varying degrees and thereby reduce the development of AGE in addition to their primary respective function (Jud and Sourij, 2019). However, despite the wide availability and variety of traditional diabetes drugs and the emergence of novel antihyperglycemic agents such as SGLT2 inhibitors and GLP-1 agonists (Brown et al., 2021), the morbidity caused by diabetes including retinopathy, nephropathy, and accelerated CVD is still unacceptably high (Centers for Disease Control and Prevention, 2020). Therefore, therapies to more specifically target AGEs and prevent and reduce diabetic-related end-organ damage are crucial. Since AGE formation is complex and involves many steps, AGE therapy is challenging (Dariya and Nagaraju, 2020). Possible pathways include AGE inhibitors, AGE cross-link breakers and RAGE antagonists. There are also dietary factors and phytochemicals that target AGE pathways through various mechanisms. Many of the following therapies are still limited by a lack of proven clinical efficacy or the presence of safety concerns.

3.2.1. AGE inhibitors

AGE inhibitors prevent AGE formation and cross-linking. Aminoguanidine, a guanidine derivative, acts as a dicarbonyl scavenger to prevent the formation of AGEs (Thornalley, 2003). It was studied in the setting of diabetic nephropathy in the ACTION I trial (A Clinical Trial In Overt Nephropathy of Type 1 Diabetics) where subjects in the aminoguanidine group had slower estimated glomerular filtration rate (eGFR) decline, reduced proteinuria, and improved lipid profiles (Bolton et al., 2004). However, the subsequent ACTION II trial (Aminoguanidine Clinical Trial in Overt Type 2 Diabetic Nephropathy) was stopped early due to futility and safety concerns with the intervention including flu-like symptoms, ANCA vasculitis, and elevated liver enzymes (Freedman et al., 1999). Pyridoxamine, an amine of vitamin B6, acts as a chelator of redox metal ions catalyzing part of Maillard reactions (Voziyan et al., 2003). It was also studied in the setting of diabetic nephropathy. A double-blind, randomized, placebo-controlled trial enrolling 317 patients with type 2 diabetic nephropathy failed to show reno-protective effects of pyridoxamine dihydrochloride (Lewis et al., 2012).

3.2.2. AGE cross-link breakers

AGE cross-link breakers target the cross-links formed between AGEs and other extracellular molecules. Alagebrium, a thiazolium derivative, is the most extensively studied AGE cross-link breaker (Toprak and Yigitaslan, 2019). In a multi-center randomized trial conducted in the US, 93 elderly patients with hypertension and vascular stiffening were enrolled, and the alagebrium group was shown to have improved total arterial compliance compared to placebo (Kass et al., 2001). In another single-arm trial, 23 patients with heart failure with preserved ejection fraction treated with alagebrium showed reduced left ventricular mass and a trend towards improvement in left ventricular diastolic function assessed by E/e’ at 16 weeks compared to baseline (Little et al., 2005). In contrast, in another well-designed randomized, double-blind, placebo-controlled study enrolling 120 patients with heart failure with reduced ejection fraction, the alagebrium group did not show improvement in overall exercise tolerance or systolic dysfunction at 36 weeks (Hartog et al., 2011). Overall, alagebrium seems well tolerated in all the studies without major side effects. Unfortunately, further clinical trials have been terminated due to financial difficulties (Toprak and Yigitaslan, 2019). Currently, there are ongoing trials to evaluate the efficacy and safety of other AGE cross-link breakers in patients with diabetes and heart failure, such as TRC 041266 (NCT04507347).

3.2.3. RAGE antagonists

Blocking RAGE signaling has been shown to reduce downstream inflammation and progression of diabetic vascular complications (Hudson and Lippman, 2018). Soluble RAGE (sRAGE), a recombinant form of RAGE comprising the extracellular region, has been demonstrated to reduce the size and complexity of atherosclerotic lesions (Park et al., 1998), diminish albuminuria with improved renal function (Wendt et al., 2003), and enhance wound healing (Goova et al., 2001) in diabetic rats. However, due to a large size, it is difficult and expensive to produce at therapeutic levels, making it less ideal in clinical use (Kim et al., 2021). Small-molecular RAGE inhibitors which are designed to target the extracellular ligand-binding site of RAGE or block the intracellular signaling pathway seem a more feasible choice. One of them, TTP 488, also known as PF-04494700 or azeliragon, has been tested in multiple clinical trials of patients with Alzheimer’s disease. In 2007, a phase 2 trial recruited 399 patients with mild to moderate Alzheimer’s was terminated early due to safety signals in the high dose group and concern for futility in the low dose group (Burstein et al., 2014). However, subsequent analysis on the follow-up data suggested a possible slower cognitive decline in the low dose group (Galasko et al., 2014). Subsequently, a Phase 3 trial of TTP 488, STEADFAST (Evaluation of the Efficacy and Safety of TTP488 in Patients With Mild Alzheimer’s Disease; NCT02080364), failed to meet its co-primary endpoint of cognitive assessment scores.

3.2.4. Diet and phytochemicals to reduce AGEs

Food is an exogenous source of AGEs. Frying or cooking with high temperature makes food more susceptible to Maillard reactions and thus higher AGE concentrations (Santos and Penha-Silva, 2021). In a randomized crossover dietary interventional trial with 62 healthy adults, consuming a diet prepared at high temperature increased insulin resistance and cholesterol and triglyceride levels compared with a diet prepared through steaming (Birlouez-Aragon et al., 2010). In a meta-analysis of 13 randomized clinical trials, a low-AGE diet improved cardiometabolic parameters, including reduction in insulin resistance, total cholesterol, and low-density lipoprotein (Sohouli et al., 2021). There are many medicinal plants and their derived phytochemicals that have been shown to have anti-AGE activities based on in vitro or animal studies (Parveen et al., 2021). Some examples include apigenin (Zhou et al., 2019), curcumin (Lin et al., 2012), and epigallocatechin 3-gallate (Sampath et al., 2017). They pose future opportunities in the multifaceted therapeutic approach to prevent and treat AGE-mediated diseases. Many additional excellent reviews on this specific topic exist (Guilbaud et al., 2016; Tessier and Birlouez-Aragon, 2012).

4. Conclusions

This review discussed the status of multiple preventative and therapeutic strategies targeting at nPTMs in various diseases. Decades of basic science research with mechanistic studies and animal models have contributed a rich literature suggesting the pathophysiological roles of nPTMs in different disease states. In more recent years, epidemiological studies have shown the mechanistically informed adverse associations of nPTMs to clinical outcomes. Now, a variety of therapeutic approaches aimed at reducing nPTMs are being translated into potential clinical strategies to prevent and treat several different diseases processes. Moving forward, we next need more in human clinical trials with clinical endpoints to say for certain whether nPTM therapies can deliver on their promise to improve patient outcomes.