Muscle protein turnover and low-protein diets in patients with chronic kidney disease

Abstract

Adaptation to a low-protein diet (LPD) involves a reduction in the rate of amino acid (AA) flux and oxidation, leading to more efficient use of dietary AA and reduced ureagenesis. Of note, the concept of ‘adaptation’ to low-protein intakes has been separated from the concept of ‘accommodation’, the latter term implying a decrease in protein synthesis, with development of wasting, when dietary protein intake becomes inadequate, i.e. beyond the limits of the adaptive mechanisms. Acidosis, insulin resistance and inflammation are recognized mechanisms that can increase protein degradation and can impair the ability to activate an adaptive response when an LPD is prescribed in a chronic kidney disease (CKD) patient. Current evidence shows that, in the short term, clinically stable patients with CKD Stages 3–5 can efficiently adapt their muscle protein turnover to an LPD containing 0.55–0.6 g protein/kg or a supplemented very-low-protein diet (VLPD) by decreasing muscle protein degradation and increasing the efficiency of muscle protein turnover. Recent long-term randomized clinical trials on supplemented VLPDs in patients with CKD have shown a very good safety profile, suggesting that observations shown by short-term studies on muscle protein turnover can be extrapolated to the long-term period.

CKD, low-protein diets, nutrition, protein

BACKGROUND

Despite the fact that protein restriction has been used for many decades in the treatment of patients with chronic kidney disease (CKD), there are still several unresolved issues regarding the metabolic effects of low-protein diets (LPDs). One major issue is our still incomplete understanding of the response of muscle metabolism to protein restriction in humans. Skeletal muscle is a highly adaptive tissue that responds to hormones, substrate supply and exercise with changes in protein metabolism and ultimately in muscle composition and size [1]. However, how and to what extent muscle protein metabolism adapts to decreased protein intake in humans is still largely unexplored. Studies performed using the nitrogen (N) balance have shown that healthy young subjects can stay on neutral or even slightly positive balance with protein intakes as low as 0.55–0.6 g/kg [2–5]. Studies performed using stable isotope amino acid (AA) kinetics have shown that adaptation to dietary protein restriction involves a reduction in the rate of AA flux and oxidation, leading to more efficient use of dietary AA and a decrease in ureagenesis [2–5]. Of note, the concept of ‘adaptation’ to low-protein intakes has been separated from the concept of ‘accommodation’, the latter term implying a decrease in protein synthesis, with the development of wasting, when dietary protein intake becomes inadequate, i.e. beyond the limits of the adaptive mechanisms [3, 4, 6]. In patients with CKD, an impaired ability to activate an adaptive response might impair N conservation when an LPD is prescribed [7].

The aim of this work is to review the mechanisms underlying the response of skeletal muscle to LPDs in CKD, while providing a summary of factors that increase the risk of protein wasting. The mechanisms underlying wasting syndrome include the loss of kidney metabolism and function as well as the activation of pathways leading to anorexia, acidosis, altered intracellular insulin signalling and inflammation, which proceed along with the progression of CKD. These factors overlap with those already operating in ageing and in comorbid conditions, such as diabetes and sepsis, and are likely to orchestrate the wasting syndrome.

RATES OF BODY PROTEIN TURNOVER ARE HIGH IN HUMANS

Conventionally, in humans, protein metabolism and requirements have been investigated by the N balance technique [8]. N balance is the net result of several metabolic reactions involved in the degradation and synthesis of protein, which does not give information on the processes of protein turnover. Rates of protein synthesis and degradation can be measured by determining AA incorporation into proteins or release from in vivo or isolated and perfused organs. Current techniques for the measuring of protein turnover in vivo in humans are based on the isotopic dilution of a tracer [8]. By the most common approach, protein turnover rates are evaluated by introducing a stable isotope-labelled AA into the free AA pool and by studying its dilution by unlabelled AAs deriving from the degradation of endogenous or dietary proteins, or its disappearance into oxidative pathways and synthesis. In skeletal muscle, the simultaneous measurement of mixed muscle protein synthesis and degradation is possible using primed, constant infusions of isotopically labelled AAs associated with the arteriovenous catheterization technique across the forearm, which is mainly made of muscle [8]. This technique has allowed us to understand whether protein loss is caused primarily by a change in synthesis or degradation.

The body of a 70-kg man contains about 10–12 kg of protein that is mainly stored in skeletal muscle [9]. This protein content is remarkably stable over time, despite protein degradation being high and ~3% of body proteins (as much as 250–300 g) being degraded daily (Figure 1). The stability of body protein content is guaranteed by the continuous large recycling into protein synthesis of AA deriving from protein degradation. This fast and continuous catabolism and resynthesis of body protein accounts for ~70–80% of the turnover of free essential AA in the basal, post-absorptive state. The efficiency by which AA is recycled from protein breakdown is so high (65% of AA deriving from protein degradation is reused for protein synthesis) that, given the nutrient intake that is common in Western countries, only 12–15 g of N are eliminated through urea excretion daily, reflecting the net catabolism of ~70–80 g of protein, which are replaced by protein intake (Figure 1) [1, 8].

FIGURE 1

Protein turnover at normal (1 g/kg) protein intake in the post-absorptive state in a 70-kg man. The maintenance of body protein is performed through extensive reutilization of AAs released during protein degradation.

It is interesting that the rates of protein turnover are high, about four to five times the amount of protein ingested in the diet. The maintenance of high protein turnover rates can serve several purposes. On the one hand, high protein rates are needed to sustain growth in the infant. On the other hand, high protein turnover allows peripheral tissues to supply enough substrates to central organs during fasting. During life, rates of protein turnover decline, but they can be increased substantially in response to trauma or sepsis, conditions that need an increase in protein synthesis as a part of a reparative process [8].

Dietary protein has no nutritional value unless it is hydrolysed by proteases and peptidases to AA, dipeptides or tripeptides in the lumen of the small intestine (Figure 1). Thus the content, digestibility and relative proportions of AA in ingested protein are the determinants of their nutritional value. Proteins are digested in the gastrointestinal tract of the small intestine. Many AAs are metabolized by luminal bacteria and some are oxidized in the intestine; ~90% of glutamate ingested in the diet is utilized by the small intestine and only 5–10% of dietary glutamate enters the portal vein. AA metabolites are excreted in faeces and urine. In healthy subjects, some AA may be toxic, as is seen in various genetic metabolic disorders. Other AAs, e.g. glutamine, seem to be readily tolerated at high doses with no adverse effects. As shown in early studies in human subjects fed with a protein-adequate diet, an excess of a specific AA results in an increase in its oxidation, but not necessarily the oxidation of other AAs. In contrast, when an AA (particularly an AA not synthesized by animal cells) is deficient in a diet, the oxidation of other AAs is increased progressively with the increasing dietary intake of AAs or protein (see Watford and Wu [10] for a review).

In CKD, some of the potential toxins produced by bacteria include hydrogen ions, phenols, indoxyl compounds, guanidines and advanced glycation end products [11]. The generation of many of these products is reduced by restriction of dietary protein intake. Thus protein restriction may be useful to decrease uraemic toxicity.

MUSCLE MASS IS MAINTAINED DESPITE HIGH- OR LOW-PROTEIN INTAKE THROUGH CHANGES OF WHOLE BODY AND MUSCLE PROTEIN TURNOVER IN HUMANS: ‘ADAPTATION’ VERSUS ‘ACCOMMODATION’

Humans are able to maintain lean body mass and body protein balance over a broad range of dietary intakes of protein. Healthy young subjects can maintain neutral or slightly positive balance with protein intakes as low as 0.55–0.6 g/kg [12]. The Food and Agriculture Organization/World Health Organization and Food and Nutrition Board of the National Academy of Sciences have added 33% to this figure to obtain a safe intake of 0.8 g/kg/day [12]. However, there are insufficient data to establish a tolerable upper level in healthy subjects. If protein intake increases from 50 to as much as 300 g of protein per day in very-high-protein diets, after an initial increase in AA oxidation and protein synthesis, a new balance is obtained with no further increase in the anabolic response. Healthy children 1–3 years old can tolerate a dietary intake of 5 g/kg/day [10].

When protein intake decreases (but it is still adequate), the human body responds with integrated and adaptive metabolic changes involving a reduction in AA oxidation and protein degradation with more efficient use of AA deriving from protein degradation (Figure 2) [13–15]. The term ‘adaptation’ has been proposed to describe the physiological changes in protein metabolism that follow variations in protein intake that occur at or above the minimum required level. During adaptation to an LPD, protein synthesis is unchanged or declines to a parallel degree of protein degradation, to maintain body composition. However, these kinds of changes require time to be fully entrained, as shown by studies of the effects of LPD on AA oxidation and N excretion [16].

FIGURE 2

Adaptation to a low-protein intake in a 70-kg man: decrease of protein degradation and oxidation with maintenance of protein synthesis (body weight and composition stable).

The term ‘accommodation’ expresses the metabolic changes that permit the body to maintain N equilibrium, but only by means of a loss of cell mass and functional impairment, with adverse health implications (Figure 3). ‘Accomodation’ will occur when intake is lower than minimum requirements, i.e. a value that is presently not known but seems to be between 0.40 and 0.50 g/kg/day at the lower end of the reported distribution of requirements in normal young subjects [10, 15, 16]. In this case, in spite of an adaptative decrease in AA oxidation and protein degradation, protein synthesis also declines, with reduced recycling of AA deriving from protein degradation (Figure 3). ‘Accomodation’ is observed in people with protein energy wasting due to starvation. In this setting, zero N balance is obtained at the cost of reduced metabolic expenditure, muscle atrophy and functional impairment.

FIGURE 3

Accommodation to a low-protein intake causes decreased protein synthesis (either absolute or relative versus protein degradation).

Preservation of muscle protein synthesis is a necessary component of the ‘adaptation’ to low protein intake. However, it is unclear what is the minimal protein intake that may maintain skeletal muscle protein synthesis in healthy subjects. Hursel et al. [5] observed that a 10–12% decrease in whole-body protein synthesis and degradation took place during the adaptation to low (0.4 g/kg) versus high (2.4 g/kg) protein intake. However, the 0.4 g/kg protein diet did not lower fractional muscle protein synthesis. These data suggest that muscle protein synthesis in some healthy young individuals may be unexpectedly maintained at very low protein intakes. However, a 0.45 g/kg/day intake is not sufficient to maintain protein synthesis and muscle mass in otherwise healthy elderly subjects [17, 18]. In CKD, early attempts in young or middle-aged subjects using protein intakes as low as 0.25 g/kg, even if supplemented with keto acid (KA), showed the occurrence of muscle wasting after a few months [19]. However, many studies [11, 20, 21] and a recent Cochrane meta-analysis [22] suggest that in young/middle-aged patients, LPDs (protein intake set at 0.55–0.6 g/kg) and supplemented very low-protein diets (sVLPDs; 0.38–0.4 g/kg), when properly prescribed, do not engender protein energy wasting.

CONDITIONS THAT CAN IMPAIR THE ADAPTATION TO LOW PROTEIN INTAKE IN CKD: METABOLIC ACIDOSIS, INSULIN RESISTANCE AND INFLAMMATION

The dietary protein requirements of clinically stable, non-nephrotic adult CKD patients appear to be similar to those of normal healthy subjects, therefore the recommended LPD provides 0.55–0.80 g protein/kg/day [20]. More than 50% of the proteins should be of high biological value. In non-dialysed patients with CKD, sVLPDs offer ~0.3–0.4 g protein/kg/day supplemented with a mixture of ~7–15 g KA or hydroxy acid analogues of five essential AAs plus tryptophan, histidine, threonine and lysine. By comparison, the recommended dietary protein intake in maintenance dialysis patients are greater, ~1.2–1.3 g/kg/day [20]. Inadequate nutritional intake in patients with advanced CKD is very common, with a large percentage of patients beginning dialysis therapy affected by some degree of wasting [22–24]. More often, patients undergo weight loss during the 3–6 months before their commencement of chronic dialysis therapy, suggesting that anorexia and/or catabolism are responsible for wasting.

Anorexia results from increases in anorexigenic hormones [23] and activation of pro-inflammatory cytokines [24]. However, mechanisms causing loss of muscle protein and fat are complex and not always associated with anorexia but are linked to abnormalities that stimulate protein degradation and/or decrease protein synthesis. Major recent advances in our knowledge of the pathophysiology of protein metabolism have allowed us to understand the mechanisms by which acidosis, insulin resistance (in CKD Stages 3-5) and inflammation (in CKD Stage 5d) affect anabolic intracellular signals that decrease protein synthesis and promote protein degradation.

Metabolic acidosis

Early studies examining net AA exchange across peripheral tissues in non-malnourished CKD Stage 4–5 patients matched with controls eating similar diets [25–27] failed to show an increase in net muscle breakdown. An inverse relationship has been observed between muscle net protein breakdown and blood bicarbonate, suggesting that acidosis is responsible for a more negative protein balance. Two studies have shown that correction of metabolic acidosis decreases protein degradation in the whole body [28, 29] and one study in muscle [30] in CKD Stages 5-5d. In addition, Adey et al. [31] were able to show lower synthetic rates of myosin heavy chain and mitochondrial proteins, muscle cytochrome c-oxidase activity and citrate synthase in patients at an early CKD stage, suggesting that decreased energy availability and/or mitochondrial damage are responsible for alterations in specific protein synthesis rates.

In summary, available data in vivo indicate that acidosis accelerates muscle protein degradation, but muscle protein metabolism is normal in non-acidaemic patients with CKD Stages 4–5. However, some unknown factor [including the reduced availability of insulin-like growth factor I (IGF-I) or some unidentified uraemic toxin] may reduce oxidative metabolism and impair the synthesis of specialized muscle protein, such as the myosin heavy chain.

Insulin resistance

Insulin resistance is a common complication of CKD [32, 33]. Although not clinically remarkable, insulin resistance appears to be strongly implicated in the pathogenesis of hypertension, accelerated atherosclerosis and cardiovascular events [34]. A post-receptor defect in muscle responsiveness to insulin is the cause of insulin resistance with regard to glucose metabolism in CKD patients [35]. Besides its effects on glucose and lipid metabolism, insulin also influences a number of metabolic pathways related to protein metabolism, which lead to anabolic or anticatabolic effects [36]. In particular, even small increases in blood insulin levels, well within the physiological range, are associated with pronounced inhibition of muscle protein breakdown [37]. An emergent hypothesis is that resistance to the anabolic drive by insulin may contribute to the loss of strength and muscle mass that is observed with the progression of CKD [38]. In studies in animal models of CKD, Bailey et al. [39] identified a series of abnormal post-receptor signalling changes in the insulin–IGF-1 pathway in muscle, which include functional abnormalities in the insulin receptor substrate (IRS)→phosphotidylinositide 3-kinase (PI3K) cascade that decrease the phosphorylation of the downstream effector Akt. Low phosphorylated Akt activity has been shown to stimulate the expression of specific E3 ubiquitin-conjugating enzymes, atrogin-1/MAFbx and MuRF1, to accelerate muscle protein degradation. In addition, a decrease in muscle PI3K activity per se activates Bax, leading to the stimulation of caspase-3 activity and increased protein degradation [38, 39]. It is interesting that metabolic acidosis, a common complication observed in patients with CKD, also interferes with insulin-induced intracellular signalling by suppressing IRS/PI3K activity in muscle and thus increases protein degradation through an upregulation of the ubiquitin-requiring pathway [39]. Of note, the defective pAkt signal can be partially recovered by high-dose insulin, correcting acidosis and/or muscle overloading [40, 41], suggesting that physical exercise could partially restore abnormal insulin signalling in uraemia.

In humans, evidence that insulin resistance also extends to protein metabolism in CKD is suggested by cross-sectional studies in which protein turnover was compared with basal insulin levels or with ‘static’ indexes of insulin resistance [26, 42]. However, when AA/protein metabolism [43, 44] was evaluated during the euglycaemic hyperinsulinaemic clamp, a normal antiproteolytic response to insulin was observed, even in the presence of metabolic acidosis [45]. It is interesting that in these studies the effects of insulin were tested in the high (~60–100 µU/mL) insulin physiological range, which is needed to test insulin sensitivity regarding glucose. However, protein metabolism appears to be maximally sensitive at only modestly increased (up to 25–30 μU/mL) insulin levels [37]. Therefore the occurrence of defects in muscle insulin’s response in uraemia could have been obscured by the high insulin levels attained in previous studies. Recently we explored the effects of two different insulin levels on two selected endpoints of its action, such as glucose and protein metabolism in the muscle of patients with non-diabetic CKD Stages 4–5 [46]. Forearm glucose balance and protein turnover were measured basally and in response to insulin infused at different rates to increase local forearm plasma insulin concentration by ~20 and 50 μU/mL. In response to insulin, forearm glucose uptake was increased to a lesser extent (−40%) in patients with CKD than controls, thus indicating insulin resistance for glucose. However, in patients with CKD, net protein balance and protein degradation were sensitive to the high but not the low insulin infusion. Thus, from these results, the normal suppression of proteolysis by insulin at low insulin levels, such as during fasting, is impaired in CKD. In contrast, the sensitivity to insulin at higher insulin concentrations, as may occur with meals containing refined carbohydrates, is preserved.

At a more advanced CKD stage, insulin resistance appears to blunt protein synthesis, as shown recently by Deger et al. [47]. In haemodialysis patients during hyperinsulinaemic-euglycaemic-euaminoacidaemic (dual) clamp, they observed a lower increase in body protein synthesis and a lower suppression of body protein breakdown compared with controls.

Inflammation

Although wasting in CKD has for a long time been considered a non-immune disease, an emerging hypothesis is that innate immunity plays a role in its development and progression [48, 49], analogous with cancer [50] and cardiac [51] cachexia. Recently an upregulation of several genes associated with inflammation in muscle has been demonstrated to occur both in rodent models [52, 53] and humans with CKD [54–57]. The direct evidence of catabolic effects of inflammation on muscle derives from recent studies performed in maintenance haemodialysis patients. Net muscle protein breakdown in haemodialysis patients with inflammation is greater than in non-inflamed patients [55]; in addition, muscle Interleukin-6 (IL-6) release and protein breakdown are directly related [55]. Recently Deger et al. [58] used stable isotope kinetic studies to assess whole-body and skeletal muscle protein turnover in a large cohort of patients on haemodialysis. They were able to show an association between circulating C-reactive protein and muscle protein breakdown [58]. Moreover, Zhang et al. [53] recently demonstrated that high IL-6, by the activation of the Janus kinase–signal transducer and activator of transcription pathway, upregulates the gene expression of myostatin, a major negative regulator of muscle protein content and regeneration. Such a mechanism has been observed both in rodent models and in the muscle of CKD patients [59]. In addition, in CKD patients, pro-inflammatory cytokines cause a resistance to the action of insulin [40] and growth hormone (GH)/insulin-like growth factor-1 (IGF-1) [59–61]. Thomas et al. [40] demonstrated a marked muscle increase in the expression of signal regulatory protein α (SIRP-α), a transmembrane glycoprotein, which was shown to interact directly with IRS-1 phosphorylation and proteolysis [40]. IRS-1 phosphorylation was also reduced, suggesting that a low value of IRS-1 causes insulin resistance by interrupting insulin-stimulated intracellular signalling. The trigger of increased SIRP-α expression in muscle is activation of the nuclear factor κB (NF-κB) pathway, which indicates a novel mechanism linking inflammation, insulin resistance and protein degradation in uraemia [40].

Toll‐like receptors (TLRs) appear to control many of the inflammatory effects in skeletal muscle in CKD. TLRs are a family of receptors in the innate immune system that mediate signal transduction pathways through the activation of transcription factors that regulate the expression of pro-inflammatory cytokines in several cell types and tissues [62]. Skeletal muscle possesses both the afferent and efferent limbs of the innate immune system, including both early and late‐phase cytokines [63]. During sepsis, in skeletal muscle, TLRs monitor for the presence of endotoxin [63] and, upon activation, induce a local inflammatory response [64–66] culminating in the translocation of NF-κB to the nucleus and activation of inflammatory genes. In addition to microbial products, TLRs can also be activated by endogenous signals of tissue injury, including debris from apoptotic and necrotic cells, oligosaccharides, heat shock proteins and nucleic acid fragments [66, 67].Recent observations suggest that TLR4 is a link between uraemia and innate immunity in skeletal muscle. In CKD Stage 5, an upregulation of TLR4 in muscle is predicted by the estimated glomerular filtration rate (eGFR) and subjective global assessment, suggesting that toxins produced or retained during progressive decline in renal function mediate TLR4 activation [67, 68] (Figure 4). In this regard, muscle cell myotubes exposed to uraemic serum show an increase in TLR4 and tumour necrosis factor (TNF)‐α and a downregulation of pAkt. These effects are prevented by blockade of TLR4 [67]. These findings suggest therefore that muscle inflammation is, at least in part, due to an upregulation of TLR4 with activation of its downstream inflammatory pathway [67, 68].

FIGURE 4

Muscle TLR4 is inversely related to residual eGFR in patients with CKD (graphical analysis from Verzola et al. [67].

OTHER FACTORS THAT CAN ACCELERATE CATABOLISM IN PATIENTS WITH ADVANCED-STAGE CKD

At late CKD stages, other stressors may accelerate catabolism and increase protein requirements. Elevated angiotensin II levels [38] blunt insulin- and IGF-1-induced intracellular signalling to promote muscle catabolism. Other triggers of insulin resistance include uraemic toxins accumulation. P-cresyl sulphate, both in animal models and in patients with CKD, resulted in impaired insulin-stimulated intracellular signalling [69]. An excess of urea has also been implicated as a mechanism causing insulin resistance in cultured 3T3-L1 adipocytes [70].

In conclusion, both metabolic acidosis and insulin resistance are frequently found and promote net catabolism in CKD Stage 3–4 patients. On the one hand, acidosis needs correction before restricting protein intake, but on the other, the defects in the regulation of insulin by protein metabolism are potentially dangerous in conditions characterized by low insulinaemia, such as during fasting and/or low-energy intakes. At a more advanced stage and in haemodialysis patients, inflammation accelerates protein breakdown. Even a properly implemented LPD or sVLPD could aggravate protein energy wasting in patients who are inflamed or with chronic infections. Clinical evidence shows that in these groups of patients a higher dietary protein intake might be necessary.

EFFECTS OF LPD ON WHOLE-BODY AND SKELETAL MUSCLE PROTEIN METABOLISM IN PATIENTS WITH CKD STAGES 3–5

Most studies that have evaluated the effects of LPD have examined patients with non-diabetic CKD Stages 3–5. N balance studies have shown that these patients can maintain neutral or slightly positive N balance with protein intakes as low as 0.55–0.6 g/kg (see Hanafusa et al. [11] for a review). Indirect data suggest that the mechanisms by which protein turnover adapts to a low-protein intake can be impaired in the presence of metabolic acidosis. Several years ago, Williams et al. [71] evaluated the effects of LPDs on the urinary excretion of 3-methylhistidine (as an index of muscle catabolism) in adults with CKD before and after treatment of acidosis. Alkali treatment decreased both 3-methylhistidine and N urinary excretion, suggesting that acidosis impairs N utilization and promotes muscle catabolism. Other studies indicate that when acidosis is corrected, the adaptation of protein metabolism to LPD is not impaired. Goodship et al. [72] observed that in non-acidotic patients with CKD Stages 4–5, the adaptive response to a standard low-protein diet (0.6 g/kg/day) included both a normal decline in the rates of whole-body leucine oxidation and protein degradation. Bernhard et al. [73] studied the effects of LPDs (1.1 g/kg versus 0.71 g/kg) on ¹³C-leucine kinetics in 12 patients with CKD Stages 4–5. At the end of the 3-month low-protein period, protein degradation decreased by ~8% and leucine oxidation by 18%; nutritional status was unmodified. Taken together, these studies suggest that in CKD Stages 4–5, under sufficient energy intake, a low-protein diet containing 0.6–0.7 g protein/kg causes adaptive changes in body protein metabolism and is nutritionally safe.

However, the aforementioned studies have evaluated the whole-body protein turnover response, which is the result of the AAs derived from all organs and tissues. As a matter of fact, muscle is estimated to contribute 35–40% to whole-body protein turnover [74] in humans. Recently we addressed this issue by measuring muscle protein turnover in two cohorts of CKD patients given two levels of dietary protein: the more often used LPD providing 0.55 g protein/kg/day or a VLPD providing 0.45 g protein/kg/day, supplemented (0.1 g/kg) with essential AAs (EAAs) and KAs [75]. Compliance with the diets was excellent during the studies and there were no notable changes in nutritional biomarkers, such as serum albumin, body mass index or anthropometric measurements during either study. The first study was designed to evaluate changes in muscle protein metabolism induced by LPD and was a randomized, parallel design in which CKD Stage 4–5 subjects were randomly divided into two groups that received either a 0.55 g/kg LPD (six subjects) or a 1.1 g/kg/day usual protein diet (five subjects). The results showed that compared with the usual diet, a gradual decrease of protein to 0.55 g/kg body weight/day led to a 17% decrease in muscle protein degradation at the end of the study, while muscle protein synthesis was not affected substantially. The first new observation made from this study is that in patients with CKD Stages 4–5, skeletal muscle responds to a 0.55 g/kg LPD through a change in muscle protein dynamics, which is characterized by a less negative protein balance due to the combined effects of reduced protein degradation, unchanged protein synthesis and overall increased efficiency of protein metabolism (Figure 5). In prospective, these changes serve to decrease muscle N losses and may thus result in N and muscle mass preservation in response to decreased N intake. In accordance with some previous observations [73], whole-body protein turnover was not affected by a 0.55 g/kg LPD. Accordingly, extramuscle tissues and visceral proteins appear to be less sensitive to a decrease in protein intake compared with muscle tissue [76].

FIGURE 5

Adaptation to an LPD in patients with CKD. Under diets containing 1.1 g/kg protein, 66% of AA phenylalanine is recycled into protein synthesis. During an LPD (0.55 g/kg), AA recycling increases to 76%, indicating greater efficiency of protein turnover (Reprinted from Garibotto et al. [75]. © May 2018, with permission from Elsevier).

What are the molecular mechanisms underlying the response of muscle protein metabolism to decreased protein intake? A few studies are available on this issue. In soleus muscles from growing rats, an LPD decreases muscle protein degradation, an effect that is associated with reductions in the messenger RNA levels of atrogin-1 and MuRF-1, proteasome and caspase-3 activity [77]. Besides an effect due to reduced AA availability from ingested protein, it is suggested that a decrease in muscle protein degradation pathways follows an increase in insulin sensitivity, with an upregulation of insulin receptor and phosphorylation of Akt [78, 79] (Figure 6). In contrast, high-protein intake inhibits other actions of insulin at multiple levels, including the insulin-stimulated tyrosine phosphorylation of IRS-1 and IRS-2 and insulin-stimulated phosphatidylinositol 3-kinase [80].

FIGURE 6

Why low-protein intake causes a decrease in protein degradation? Studies in animals suggest that a decrease in muscle protein degradation pathways follows an increase in insulin sensitivity, with an upregulation of insulin receptor and phosphorylation of Akt (in orange ).

EFFECTS OF AN sVLPD ON SKELETAL MUSCLE PROTEIN SYNTHESIS AND DEGRADATION AND EFFICIENCY OF PROTEIN TURNOVER IN PATIENTS WITH CKD Stages 4–5

A common concern regarding the use of sVLPDs in patients with advanced-stage CKD is their potential adverse effect on nutritional status. A further concern is compliance with sVLPDs. Also, the optimal restricted dietary protein intake to exert the most beneficial effect is unknown. However, many of these concerns diminished after the publication of studies showing that sVLPDs followed by motivated and compliant patients were effective and do not have harmful effects on their nutritional condition [81], although one study indicated a significant decrease in bone mass over 2 years [82].

Two early studies indicate that protein metabolism and N balance adapt successfully in patients who are compliant with a VLPD supplemented with essential AAs and KAs. Masud et al. [83] administered a VLPD (0.28 g/protein/kg) supplemented with KAs or AAs in non-acidotic CKD patients. When patients were studied 3 weeks after the start of the sVLPD regimen, it was found that the body N balance was neutral and that neutrality was achieved by a marked reduction in whole-body AA oxidation and post-prandial inhibition of protein degradation. In a second study, Maroni et al. [84] evaluated the long-term adaptive response of six CKD patients to sVLPDs. At the end of the study, rates of N balance did not differ from basal values, as well as protein synthesis, degradation and AA oxidation, indicating that N balance was achieved by long-term suppression of AA oxidation and protein degradation.

To assess the effects of an sVLPD on muscle, we measured muscle protein turnover in six CKD Stage 4–5 patients given a VLPD providing 0.45 g protein/kg/day supplemented (0.1 g/kg) with EAAs and KAs [75]. The study was a three-period, two-treatment crossover trial in which patients served as their own controls. After the run-in period (during which patients were instructed to maintain a stable diet), the study consisted of three 6-week consecutive periods: the baseline period (LPD, 0.55 g/kg), the treatment period (VLPD, 0.45 g/kg + AA/KA 0.1 g/kg) and the washout period (LPD, 0.55 g/kg). In keeping with previous studies [83, 84], we observed that the sVLPD was not associated with changes in body weight and composition; however, in terms of muscle protein kinetics, this diet was associated with a lower net muscle catabolism compared with the 0.55 g/kg LPD. The already low muscle protein degradation rates attained during an LPD were not further reduced by the supplemented VLPD, suggesting that adaptation to a more restricted protein intake cannot be attained by further restraining protein degradation. However, the decrease in net protein catabolism by the supplemented VLPD was likely an effect of increased efficiency of protein turnover and preservation of muscle protein synthesis as compared with the LPD (Figure 7). Of note, the supplemented VLPD provided an intake similar to a 0.55 g protein/kg/day diet, but the quality of the protein and KA/EAA mix was higher. These effects are in keeping with studies showing that higher protein quality intake better stimulates muscle protein anabolism and protein synthesis. Leucine, in particular, has been identified as the key factor stimulating the post-prandial increase in muscle protein synthesis [85]. Recently, autophagy and apoptosis have been shown to be reduced when an LPD is supplemented with leucine and KAs, as compared with an LPD, in rodents with CKD [86, 87]. Several lines of evidence also show that in healthy subjects, proteins rich in essential AAs, such as animal proteins, have a greater nutritional value than plant-source proteins to sustain skeletal muscle mass [88–90]. Also, long-term vegetarianism resulted in reduced skeletal muscle mass in older women compared with consumption of an omnivorous diet [90].

FIGURE 7

The adaptation to an sVLPD in patients with CKD. Under diets containing 0.55 g/kg protein, 72% of AA phenylalanine is recycled into protein synthesis. In a VLPD (0.45 g/kg) supplemented with AA/KA (0.1 g/kg), the entity of AA recycling increases to 86%, indicating greater efficiency of protein turnover (Reprinted from Garibotto et al. [75]. © May 2018, with permission from Elsevier).

One constant effect of LPD/VLPD + KA treatment is the reduction of serum urea [91, 92]. Initially it was thought that an important quantity of urea was hydrolysed during CKD and the amino group released from urea could aminate KAs and form new AAs. However, current evidence indicates that the reduction in net urea generation with VLPD + KAs is due to decreased net muscle protein degradation.

EFFECTS OF AN LPD AND AN sVLPD ON THE EFFICIENCY OF MUSCLE PROTEIN TURNOVER

The magnitude of daily protein turnover needs large reutilization of AA released by protein breakdown for protein synthesis. According to the models studied [75], as much as 60–65% of AAs deriving from proteolysis is recycled into protein synthesis in CKD under a 1.1 g/kg protein diet, a figure similar to what is observed in the normal condition [37]. In our studies, this percentage increased by 12% when patients ate a 0.55 g/kg LPD and increased further with an sVLPD. Therefore, during the adaptation to an LPD, protein turnover is characterized by more efficient AA recycling. In conclusion, in moderately severe CKD both LPD and sVLPD significantly affect forearm protein metabolism. Overall, the results of these studies indicate that skeletal muscle protein metabolism adaptation to restrained protein intake can be obtained via the combined responses of reduced protein degradation and increased efficiency of AA recycling.

AGEING

People >65 years of age are expected to soon become the majority of those who will need renal replacement therapy in many developed countries. Compared with younger adults, older adults on average are less hungry and thirsty, consume smaller meals and eat slower, eat less between meals and more commonly have lower energy intakes [93]. Weight loss in older adults is highly predictive of increased morbidity and mortality [94, 95] and it is believed to follow microinflammation and excess cytokine elaboration (‘inflammageing’). Therefore, microinflammation occurring with ageing could overlap with the upregulation of specific catabolic pathways in uraemia to promote wasting. In addition, in elderly subjects, a decreased sensitivity of insulin action is linked to frailty and cognitive impairment [96]. The traditional recommendation for protein intake of 0.8 g/kg/day for adults of all ages [97] is currently debated for healthy elderly subjects, based on increasing evidence that healthy older people might need higher amounts of protein for optimal preservation of lean body mass. Daily amounts of 1.0–1.2 g/kg body weight have been suggested for the elderly by expert groups [98,] and a recent guideline [99]. In a recent study, consumption of a diet providing twice the recommended daily allowance for protein compared with the current guidelines was found to have beneficial effects on lean body mass and leg power in elderly men [100]. Therefore a major question is whether restricting protein intake in the elderly with CKD may cause or aggravate sarcopenia. In elderly CKD subjects, a consensus paper suggests a minimum protein intake of 0.8 g/kg with an eGFR <30 mL/min [98]. However, some indirect observations suggest that individual protein requirements for may be less in CKD, in the absence of catabolic risk factors. As an example, a study on muscle protein metabolism including non-acidaemic, non-inflamed elderly CKD patients showed that these patients are minimally catabolic [55]. Also, available clinical studies suggest that elderly subjects with CKD can successfully adapt to a 0.6 g/kg LPD [21] and to sVLPD. As an example, the randomized controlled trial Diet or Dialysis in Elderly (DODE), which included 112 elderly patients >70 years of age with CKD [101], observed very modest nutritional changes following a VLPD + KAs high-calorie diet, suggesting that the proposed treatment is nutritionally safe. However, a definite response to this question can be offered only by studies specifically addressing the issue of optimal intake of protein based on physiological endpoints such as lean body mass, strength and physical function in elderly CKD patients.

CONCLUSIONS

The past two decades have witnessed great progress in our understanding of the mechanisms underlying wasting in CKD and of the response of muscle and body metabolism to protein-restricted diets. Recent observations show that, in the short term, patients with CKD Stages 4–5 can efficiently adapt their muscle protein metabolism to an LPD containing 0.55–0.6 g protein/kg or to an AA/KA-supplemented VLPD by decreasing muscle protein degradation and increasing the efficiency of muscle protein turnover. A possible limitation of these studies is that patients were selected for good compliance with protein-restricted diets. Therefore the results of these studies cannot be extended to patients who are not adherent to dietary treatment. In addition, whether these adaptive mechanisms can be maintained over a longer period, particularly if the patients undergo any complications, needs to be examined by further studies. Recent long-term randomized clinical trials on AA/KA-supplemented VLPD in patients with CKD have shown a very good safety profile, suggesting that observations shown by short-term studies on muscle protein turnover can be extrapolated to the long term.

FUNDING

This study was supported by grants from the University of Genoa (Finanziamento di Ateneo).

CONFLICT OF INTEREST STATEMENT

None declared.