Nutritional interventions based on dietary restriction and nutrient reductions for the prevention of doxorubicin chemotherapy side effects

2.1 Caloric restriction

Mitra et al. [45] hypothesized that moderate caloric restriction protects against DOX cardiotoxicity. For this, male Sprague-Dawley rats (250–275 g) were maintained on 35%diet restriction for a period of 43 days exhibiting normal cardiac function and survival after receiving a single dose of DOX (12 mg/kg of body weight (bw), intraperitoneal (ip)). In contrast, those animals ad libitum maintained showed a 100%mortality after 7 and 13 days due to the higher cardiotoxicity of DOX and cardiac dysfunction. From the mechanistic standpoint it seems that a reduction of oxidative stress, an enhanced ATP production and JAK/STAT3 pathway activation would be responsible for the moderate diet restriction-associated resiliency against DOX cardiotoxicity. The improved cardiac antioxidant defense system would result in oxidative stress reduction and downregulation of uncoupling proteins (UCPs) 2 and 3. Peroxisome proliferators activated receptor-alpha (PPARα) induction in heart and increased levels of plasma adiponectin cardiac fatty acid oxidation and mitochondrial AMP-activated protein kinase (AMPK) α2 subunit (AMPKα2). These changes led to higher levels of cardiac ATP levels and ATP/ADP ratio, increased cardiac erythropoietin and decreased levels of suppressor of cytokine signaling 3 (SCS3), which upregulated the cardioprotective JAK/STAT3 pathway. Overall, this study strongly supported that moderate diet restriction protects against DOX cardiotoxicity. Although additional mechanisms could exist, such effect would be consequence of decreased cardiac oxidative stress and levels of triglycerides (TG), an enhancement of cardiac fatty acid oxidation and ATP production and the upregulation of the JAK/STAT3 pathway in heart In this sense, a toxicokinetic analysis was also carried out in the same study revealing an equal accumulation of DOX and doxorubicinol in hearts from animals maintained on both regimens [45]. Thus, cardiotoxicty prevention does not depend on this aspect, at least under these experimental conditions.

2.2 Lipid restriction

High-fat diets, particularly rich in saturated fat have shown to have an agonistic effect on DOX toxicity at different levels in rats [46]. At heart, electrocardiogram (ECG) and histopathology results indicated that a diet containing butter as main fat, cholesterol, cholic acid and propyl thiouracil resulted in a significant augmentation of the cardiotoxicity of a chronic treatment with DOX (1 mg/kg, ip, until 18 injections). In addition, animals showed atrophy of the lymphatic tissue after DOX treatment only when received the high fat diet. Interestingly, higher levels of DOX were found in high fat diet fed rats which could explain, at least partially, the mentioned differences (Table 1). Histopathological changes at liver and elevated levels of serum glutamate-piruvate transaminase (SGPT) were also observed. Taken these results into account, authors proposed that diminished excretion by the liver as consequence of damage derived form the high fat diet consumption may contribute to DOX toxicity [47]. This would be in consistency with clinical observations identifying liver disease as one of the most important DOX-associated cardiomyopathy risk factors [47]. Similarly, in male Sprague-Dawley rats, intake of a high-fat diet (40%Kcal) for 42 days before receiving an injection of DOX (8 mg/kg of bw, ip) was associated with higher cardiotoxicity-derived consequences such as cardiac dysfunction and lipid peroxidation, and an 80%increased mortality. Curiously, this occurred in absence of any significant renal or hepatotoxicity. Moreover, in this case, DOX toxicokinetics studies revealed no differences between diets in accumulation of DOX and its toxic metabolite doxorubicinol in rat hearts [48]. This would indicate that the possible contribution to DOX cardiotoxicity increase of alteration of liver, but also kidney, by obsesogenic diets is secondary. Different alternative mechanisms for the reported effects on heart have been revealed. These include decreased mitochondrial ATP generation and fatty-acid oxidation in heart, decreased plasma adiponectin levels, decreased cardiac levels of AMPKα2, cardiac PPARα downregulation, upregulation of UCP2 and UCP3 by increasing oxyradicals, and putative JAK/STAT3 pathway downregulation by cardiac erythropoietin decrease and SCS3 increase [48]. In addition, a subsequent study evaluated the possible attenuating effect of switching from a high-fat diet to a low-fat diet on DOX cardiotoxicity exacerbated by high-fat diets. For this, male rats were maintained on a Western diet (where fat content represents 42%kcal) or a low-fat diet (11%Kcal from fat) during 6 weeks, but one week before receiving DOX. A half of Western diet-fed rats were then switched to the low-fat diet or continue with same diet. When compared to animals fed a low-fat diet treated or not with DOX, those fed a Western diet exhibited more cardiac damages 4 weeks after initiate a 10-day DOX treatment (1 mg/kg of bw, ip, per day) as evidenced the decreased septal wall thickness, fractional shortening and ER Ca^2 +-ATPase expression as well as the increased dimensions of left ventricular cavity and expression of β-myosin heavy-chain in this group of animals, correlating with elevated lipid peroxidation markers. Interestingly, this exacerbated cardiotoxicity was not found in animals fed a Western diet but switched to a low fat diet one week prior to, during, and after the chemotherapy with DOX [49].

Several studies have also focused on DOX neprhotoxicity, which have used a model of nephrosis induced by DOX to evaluate the contribution of diet-induced hypercholesterolemia on glomerulosclerosis by comparing rats fed standard or high-fat diets. Morphological alterations observed in rats maintained on the high-fat diet, which presented higher proteinuria and increases in the mesangial matrix and cells. However, more obvious pathological changes were found in those also treated with DOX (5 mg/kg of bw, iv). In mesangial areas these changes included lipid deposits and foam cell formation whereas, in some glomeruli, focal and segmental glomerulosclerosis were described [50]. Overall, these results would indicate an over-production of mesangial matrix components enhanced by high-fat diet that, in turn, would be responsible for aggravation of glomerulosclerosis. Following a similar experimental design, the same authors confirmed the effects of high-fat diets on protein excretion and kidney histology. In addition, a immunohistochemical assay showed that the production collagen IV, fibronectin and laminin, 3 major components of fibrosis, was increased in rats treated with DOX, especially in the rats maintained on the lipid-rich chow [51].

The effect of fat source was also compared in relation to DOX nephrotoxicity, although in combination with other drugs. Namely, the free radical scavenger, dimethylthiourea. After 7, 14, and 21 days, a clear reduction of proteinuria was found in rats receiving either corn oil or fish oil, both combined with dimethylthiourea in comparison with untreated control rats. However, those receiving fish oil displayed reduced urine protein excretion compared with corn oil or evening pimrose oil fed rats after 21 days. In contrast to corn oil, fish oil consumption was associated also with a reduction of cholesterolemia and serum TG levels. Renal glutathione content was increased in control rats receiving the fish oil-rich diet compared with those rich in corn oil alone [52].

Mechanistically, high-fat diets effect seems mediated, at least in part, by consequences of systemic alterations, as well as obesity associated to these diets. Actually, in most of studies hypercholesterolemia [50, 51], hiperlipidemia [47] or obesity [48] were confirmed. Likewise, a marked fatty infiltration of the liver was also observed [47]. This implies that diet that led to the same alterations have the same effect in DOX toxicity and also should be avoided. An old study comparing the effects of high-fat (15%Kcal) diets enriched with different fat types (beef tallow, fish oil, evening primrose oil, or a combination of the last two 75:25) on rats with DOX (3 mg/kg, iv)-induced nephrotic syndrome points in that sense. Diets containing fish oil (rich in -3 PUFA), evening primrose oil (rich in n-6 PUFA) and both fat decreased plasma levels of TG and total cholesterol (TC) respect than diet enriched with beef tallow. Overall, these diets rich in unsaturated fats would modulate eicosanoid synthesis and plasma lipids in a potentially antiatherogenic manner preventing glomerular sclerosis. In particular, the combination of fish and evening primrose oil also raised high density lipoprotein-cholesterol (HDL-C) levels compared with beef tallow. In addition, this prevented fish oil-induced suppression of aortic 6-keto prostaglandin F1α (6-keto-PGF1α). These findings suggest that, in this model of nephrotic syndrome, an adequate combination of n-6 and n-3 PUFA would result advantageous over either family of fatty acids [53].

Protein content of the diets had been proposed to modulate DOX nephotoxicity at kidney since it was reported that high-protein feeding leads to some changes in glomerular hemodynamics in both, animals and humans. Such changes would induce a progressive deterioration of renal function favoring loss of glomerular permselectivity properties and subsequent glomerulosclerosis. This is particularly relevant under conditions affecting renal mass [56]. Therefore, high-protein diets could facilitate kidney damage in patients under DOX treatment. To investigate this topic and possible mechanism explaining it, Benigni et al. [56] carried out an experiment in a rat model of nephrosis induced by DOX, which is characterized by heavy and persistent proteinuria as was above. After chemotherapy, they maintained rats on two different isocaloric diets containing 20%(standard diet) and 35%protein (high-protein diet), respectively, for long-term. Concerning renal function, glomerular filtration rate was increase in both normal and nephrotic animals receiving the high-protein diet. Therefore, hemodynamic changes that occur in normal animals given a high-protein diet also would take place when glomeruli are uniformly damaged by a DOX treatment. In addition, they also investigate vasodilatory prostaglandins generated at the renal level since they could be responsible for the adaptive hemodynamic changes. In nephrotic rats, the high-protein diet increased urinary excretion of 6-keto-PGFα, the stable breakdown product of prostacyclin. However, urinary excretion of prostaglandin E2 was no modified by the diets. In contrast, animals not treated with DOX showed maintained on the high-protein diet did not shown modified urinary excretion of 6-keto-PGF1α, but urinary excretion of prostaglandin E2 was reduced [56]. Anyway, it seems that global glomerular synthesis of the vasodilatory prostaglandins measured paralleled the urinary excretion pattern. Additionally, the effect of the cyclooxygenase inhibitor indomethacin on renal function and prostaglandines production was also tested. In animals fed the high-protein diet but not treated with DOX, this compound inhibited urinary excretion of vasodilatory prostaglandines but did not prevent hyperfiltration, whereas in those with nephrotic syndrome fed the same diet, both reduced urinary excretion of 6-keto-PGF1α and prostaglandin E2 urinary, and inhibited hyperfiltration were reported [56]. This last would indicate that in nephrotic animals, an enhanced renal synthesis of prostaglandines, particularly prostacycline, has a key role in the adaptative changes responsible for hyperfiltration, although in health animals, hyperfiltration appears to be independent of these compounds.

With this background, many other studies on dietary protein amount, but testing the effects of low-protein diets, have been conducted in rodent models of nephrotic syndrome induced by a DOX treatment. In an old study, the effects of feeding rats with a standard diet containing 20%protein or low-protein diet containing 6%protein starting 7 days before and just after injecting a single dose of DOX (5 mg/kg of bw, iv) were compared. In spite of animals fed the standard diet showed alterations of glomerular visceral epithelial cells with fusion of foot processes in optical and transmission electronic microscopy studies, glomerular epithelial cell abnormalities in those maintained on the low-protein diet. Functionally, the low-protein regimen led to an elevation in renal plasma flow compared to the standard diet, but it did not influence on glomerular filtration rate. Unlike animals fed the standard diet, these animals did not develop proteinuria. However, in animals maintained on a standard diet, renal function did not result affected by DOX treatment since no differences were found respect than control animals. Interestingly, the diet has no effect on kidney DOX distribution after drug injection [57]. In other study, xanthine oxidase and xanthine dehydrogenase levels were reduced in rats fed a low-protein (6%Kcal of casein) after receiving DOX treatment. Moreover, this group was only slightly proteinuric versus the notably increases of activity and massive proteinuria found in standard diet-fed rats. Moreover, pharmacological block with allopurinol and tungsten of the mentioned enzymes caused a reduction of proteinuria to one-third of the original levels, reinforcing the role proposed for these enzymes in proteinuria. In addition, a group of rats was shifted to a standard diet from a low-protein one showing a massive proteinuria and absence of increases in renal levels of xanthine oxidase and xanthine dehydrogenase that are in part responsible for the renal damage induced by DOX [58]. These results confirmed the effect of a low-protein diet on proteonuria after DOX treatment and add a possible mechanism under such effect in what low-protein diet would lead to a decrease in levels of xanthine metabolizing enzymes. Notwithstanding, the existence of additional mechanisms explaining the protective role of low-protein diets against the of DOX-induced proteinuria development has been evidenced. More recently, it was reported that after 3 days intaking a low-protein diet (8.5%Kcal), proteinuria decreased slightly, whereas bw did not change in either group. However, urinary urea nitrogen excretion was reduced by 37%and leucine oxidation decreased by 18%in nephrotic rats. After 12 days maintained on the same diet, proteinuria decreased by 45%in DOX-treated rats fed low-protein diets during 12 days, and leucine oxidation increased to values similar to those found in untreated rats. Then, a constant infusion of L-[1–14C] leucine was used to measure fasting whole body protein turnover, clarifying the effects of these diets. According to the results, it seems that protein conservation was stimulated by proteinuria even when dietary protein intake was restricted since rats maintained on low protein diets showed lower rates of whole-body protein synthesis and degradation than standard diet fed rats. In addition, the amino acid oxidation decrease seemed to depend on moderate proteinuria, since prolonged low protein diet feeding increased leucine oxidation rates until reach control group levels ameliorating nephrosis development. However, no differences were found for protein turnover rates between nephrotic and control rats receiving the experimental diet. As weight loss and rates did also not change, this suggests that moderate proteinuria does not increase protein catabolism [59]. In relation with the previous study, other study compared the effect of a low-protein (5%) diet maintained for different periods after a DOX injection. An 8-week intervention significantly ameliorated kidney lesions and remarkably decreased the synthesis of fibronectin [60]. Moreover, it decreases of the latent transforming growth factor beta (TGFβ) secretion and the expression of TGFβ1 gene. It has been reported that the increase in both, fibronectin and TGFβ protein synthesis or in expression of encoding for them would be involved in the progressive process of DOX-induced nephropathy [65]. In contrast, a 2-week low protein diet did not help to preserve kidneys from DOX-induced damage, although reduced cortical synthesis of fibronectin. Furthermore, there was no significant decrease in the latent TGFβ secretion in the DOX-treated rats. Moreover, mRNA levels of fibronectin or TGFβ1 were not affected at any stage. Thus, it is possible that a 2-week intervention has a quicker influence on fibronectin production at translation level than at transcriptional level. These differences between short and long-term protein restriction suggest that decreased fibronectin and TGFβ synthesis would prevent kidney damage [65]. More recently, the effect of a low-protein (6%Kcal) diet, but supplemented with keto acids, also has been tested in a similar model. In this case, the experimental treatment avoids proteinuria. Further, the level of urinary glycosaminoglycan was also assessed in this study finding values comparable to that of controls. Likewise, the reduction of glomerular glycosaminoglycan content observed in rats only treated with rats was partly eliminated by the supplemented low-protein diet [66].

A study also evaluated the consequences for bone of protein amount in the diet comparing the effect of high-protein (30%Kcal from protein) and low-protein (7%Kcal from protein) isocaloric diets 5 weeks after inducing nephrotic syndrome by weekly injections of DOX (2 mg/kg of bw, ip) for 6 weeks. As expected, urinary protein and phosphate excretion were higher in rats fed a high protein diet. Concerning bone, femur weight, femur mass index, femur calcium contents and bone mineral density were lower in the high protein diet fed animals. In addition, higher parathyroid hormone (PTH) and lower osteocalcin (OC) levels were found in serum. Despite no differences in bone mineral content were found between the two groups, these findings suggest a loss of bone with a high-protein diet, probably because loss of minerals derived from the intake of a high-protein combined with nephrotic syndrome affecting bone metabolism, which have negative consequences for bone health [61]. This points the importance of dietary protein content after chronic damage to kidney by DOX for maintain the health in other systems different from urinary system.

On the other hand, long-term low-protein diets before chemotherapy could have deleterious consequences on DOX toxicity susceptibility. A study in rabbits evaluating low-protein (5%Kcal), isocaloric diet effect on DOX pharmacokinetics points in this way. For this, animals were received the experimental treatments for 8–12 weeks before DOX (5 mg/kg, iv) was given. When compared with the values obtained for standard chow bits (15%Kcal), the low-protein diet decreased DOX clearance, prolonged the terminal elimination half-life, and increased the area under the curve concentration-time extrapolated to infinity in plasma. Moreover, in rabbits maintained on the low-protein diet, an extension of the terminal elimination half-life of doxorubicinol was reported, although without alteration of DOX distribution volume [54]. Overall, the study suggests that a low-protein diet feeding previous to the treatment lead to a reduction in the capability to remove DOX and possibly metabolites such as doxorubicinol. Other study in rats treated with a single dose of DOX (15 mg/kg of bw, ip) would confirm it [55]. In this, protein malnutrition due to a low-protein diet consumption significantly altered the DOX pharmacokinetic leading to a significant decrease in its elimination, which would prolong heart exposure time to this drug. This was accompanied by histopathological alterations and serum creatine kinase values confirming cardiotoxicity. Epirubicin was also used with the same aim rending similar results, which suggests that this role of protein malnutrition in DOX toxicity enhancement could be extent to more anthracyclines [55]. In any case, results from both studies are particularly important in the clinical context because the occurrence of malnutrition with protein deficiency is common in cancer patients receiving anthracycline treatment. In relation to anthracycline pharmacokinetics, it is important to note if a similar phenomenon occurs in malnourished cancer patients, this alteration can contribute to their increased risk of cardiac damages associated with DOX therapy. Actually, protein malnutrition is considered to be a risk factor for cardiac alterations derived from DOX use. The impairment of anthracycline metabolism by protein deficiency reported in the animal models presented here could explain, at least in part, the enhanced DOX toxicity under malnutrition conditions. According to this, dose adjustment would be necessary also in nutritionally-deprived patients.

2.4 Sodium restriction

The effects of diets with different amount of sodium on kidney and urinary system alterations induced by DOX have been tested in different rat models where was expected to modulate progression in proteinuric kidney disease. In a focal glomerular sclerosis rat model induced by 2 injections of DOX (2.5 mg/kg of bw), feeding on a high-salt diet (8%NaCl) for the following 8 weeks led to lower urinary protein excretion and a progressive increase in the blood urea nitrogen and serum levels of creatinine respect than those values found rats fed a standard diet (1%NaCl). In association with circulating markers, both and glomerular hypertrophy and extensive glomerulosclerosis preceded by increased glomerular diameter were reported in the experimental group. However, there were no statistically significant differences in the values of blood pressure [62]. The early increase in glomerular diameter suggested that the glomerular sclerosis alterations aggravated by the sodium-rich diet may be related to glomerular hypertrophy. The effect of other diet with an elevated content of sodium, but in this case by adding sodium bicarbonate (10%NaHCO₃) was also evaluated and a lower urinary protein excretion was also observed. Lastly, it was reported that daily administration of the Na⁺/H⁺ exchanger inhibitor amiloride to rats taking the high-sodium diet prevented the appearance of the mentioned pathologial lesions [62]. It is possible that stimulation of the Na⁺/H⁺ exchanger was associated with glomerular hypertrophy and subsequently wit the reported high-salt diet effects on glomerular sclerosis. In other study in rats with nephrotic syndrome (among whose causes are glomerular sclerosis) induced by DOX administration (7.5 mg/kg, iv), and then, the animals were fed a sodium-deficient diet. Strict dietary salt restriction for 5 weeks retarded renal inflammation and fibrosis progression in this model. A sodium-deficient diet improved renal tubulointerstitial histopathology associated with amelioration of levels of osteopontin, collagen III, and fibronectin that was increased by DOX. Similarly in a subsequent experiment DOX-induced azotemia was improved by sodium restriction However, proteinuria or blood pressure result unaffected. Dietary salt restriction also slowed DOX-induced progression of renal interstitial fibrosis. In particular, it reduced the DOX-induced upregulation of osteopontin and α-smooth muscle actin (α-SMA) expression and infiltration of monocytes and/or macrophages. DOX-induced expression of inhibitor of κB α (IκB-α), tumor necrosis factor-α (TNF-α), and the components of the NADPH oxidase system, gp91-phox, p47-phox, and p67-phox was also prevented, probably as consequence of ROS-mediated NF-κB activation [63].

On the other hand, since DOX binds to endothelial nitric oxide synthase (eNOS) the reductase domain yielding O₂ in vitro and NO modulate urinary sodium excretion, DOX treatment could cause salt-sensitive hypertension. Taken this idea into account, Hirai et al. proposed that eNOS expression would be upregulated by a high-sodium diet in the kidney of DOX-treated Sprague-Dawley rats. In this context, oxidative stress also would be enhanced. These changes would result in a facilitation of hypertension. After 4 weeks of DOX treatment, only the group receiving the high-sodium diet showed an increased systolic blood pressure. Medullar and cortical expressions of eNOS protein in kidneys were significantly more elevated with high-sodium diets than in standard diets, regardless of DOX treatment. Urinary excretion of 8-hydroxy-2-deoxyguanosine (8-OHdG) was increased particularly in the group treated with DOX and the high-sodium diet. Likewise, more prominently 8-OHdG-stained tubules were observed indicating promotion of oxidative stress. However, this group showed lower urinary excretion of nitric oxides respect than low-sodium diet-fed rats treated with DOX, but also than rats not receiving chemotherapy [64].