Vitamin D deficiency is not associated with increased oxidative stress in chronic kidney disease pre-dialysis patients

Abstract Introduction: The progressive decline in 25-hydroxyvitamin D [25(OH)D] in chronic kidney disease (CKD) limits the kidney ability of synthesizing the vitamin. Vitamin D deficiency as defined by KDIGO (25(OH)D <20 ng/mL) is prevalent in CKD patients and associated to oxidative stress (OS). We studied a possible association between vitamin D deficiency and OS in pre-dialysis patients. Methods: A cross-sectional study with 206 CKD patients was carried out. Laboratory tests for 25(OH)D, 1,25(OH)2D, inflammatory markers, and OS were added to routine tests including creatinine, albumin, calcium, phosphorus, alkaline phosphatase, iPTH, glucose, hemoglobin, uric acid, total cholesterol, LDL, HDL, and triglycerides. Results: Vitamin D deficiency was present in 55 CKD patients and normal vitamin D levels were seen in 149 patients. There was a significant association between vitamin D and estimated glomerular filtration rate (eGRF). Homocysteine levels were best predicted by eGRF, sex, and age; high sensitivity C-reactive protein (hsCRP) by staging and BMI; nitric oxide metabolites (NOx) were increased in late disease; leptin was influenced by BMI and higher in women than man; and adiponectin levels were higher in women. Conclusions: OS biomarkers were not correlated with vitamin D deficiency but increased NOx were seen in stages 4-5 CKD patients. Even though a relatively large number of CKD patients was included and a broad number of OS and inflammatory biomarkers were used in this studied we failed to find an association between vitamin D levels and eGRF. More studies are needed to evaluate the influence of vitamin D status in OS in pre-dialysis CKD patients.


IntroductIon
In recent years, vitamin D (vit D) is no longer considered to be only a hormone essential to bone metabolism, which deficiency promotes the onset of rickets in children and osteopenia, osteoporosis, and osteomalacia in adults 1 . Its deficiency has been shown to be associated with various clinical conditions, such as secondary hyperparathyroidism, diabetes mellitus, increased cardiovascular risk, and more rapid progression of chronic kidney disease (CKD) 2 .
In clinical practice, serum levels of 25(OH)D are used to define vit D deficiency 1 . In accordance to KDIGO 2 , vitamin D deficiency has been defined as 25(OH)D level <20 ng/mL. Supporting the KDIGO definition, Bouillon et al. 3 in a critical analysis of evidence-based medicine about optimal vitamin D status confirm the 20 ng/mL cutoff and concluded that serum levels of 25(OH)D above 20 ng/mL are sufficient to normalize calcium and bone homeostasis, as measured by surrogate endpoints such as 1,25(OH)2D, PTHi, calcium absorption, or bone mass.
The prevalence of vit D deficiency is very high in the general population and in CKD patients. In addition, CKD patients have reduced serum levels of the most active form of vit D, 1,25(OH)2D. These deficiencies can be associated with high cardiovascular mortality compared to the general population 4 . Obviously, the high cardiovascular mortality is multifactorial, including traditional cardiovascular (CV) risk factors, as well as uremia-related and non-traditional risk factors such as systemic inflammation and increase oxidative stress (OS), and altered mineral metabolism of bone. Recent evidence suggests that vit D deficiency may contribute to systemic inflammation and increased OS in otherwise healthy people and patients with congestive heart failure 5 .
In certain pathological conditions, including diabetes mellitus, the overproduction of reactive oxygen species (ROS) in the kidneys themselves has been implicated in renal inflammation, glomerular basement membrane derangement, and alteration in tubular function possibly contributing to CKD progression 6 .
Fibroblast growth factor 23 (FGF23) is a key regulator of phosphate and vitamin D metabolism, and it is a known predictor of metabolic bone disease in CKD. The actions of FGF23 include inhibition of 1,25OH2D synthesis and increase in urinary phosphorus. Therefore, excessive serum FGF23 concentrations may interfere with bone mineralization by decreasing the availability of calcium and phosphate substrates 7 . According to Richter et al. 8 , FGF23 and its co-receptor Klotho dysregulations were associated with endothelial dysfunction in humans, but in vitro experiments were performed to assess the effects of FGF23 in relation to its co-receptor Klotho on nitric oxide (NO) synthesis and ROS formation via activation of NADPH oxidase 2. In addition ROS degradation via superoxide dismutase 2 (SOD2) and catalase (CAT) is stimulated by FGF23. The authors conclude that FGF23 excess may primarily promote oxidative stress and thus endothelial dysfunction.
It is known that CKD patients have a significant increase of OS and reduced plasma concentrations of 25-dihydroxyvitamin D and 1,25-dihydroxyvitamin D to varying extent 9 . Based upon these studies, we investigated a possible association between deficiency of 25

Study prOtOcOl
A cross-sectional study was conducted at the CKD clinics of the State University of Londrina and Kidney Institute of Londrina, Brazil, from October 2010 to February 2011. The inclusion criteria were: age greater than 18 years, and CKD stages 3 to 5, according to National Kidney Foundation 10 ; all included patients signed the written informed consent for the study. The exclusion criteria were: age less than 18 years, patients that were renal transplant recipients, history of intestinal malabsorption disease, intestinal bypass or bowel resection surgery, presence of liver cirrhosis, infectious disease or active malignancy, unexplained loss of more than 5% of body weight within the last six months, use of steroids orally or via inhalation, use of anticonvulsants drugs, or vit D supplementation. This study was approved by the Ethics Committee of the University of Londrina (086-10 -07/08/2010), CAAE: 0083.0.268.000-10.
Blood samples (approximately 40 mL) were obtained by venipuncture into vacuum tubes (Vacutainer ® , Franklin Lakes, NJ USA) after 12 hours of fasting, and were centrifuged for 30 minutes at 3000 rpm (2100 g) at 20°C to obtain serum and plasma, and frozen at -80°C.
Vit D deficiency was defined as serum levels of 25(OH) vitamin D (calcidiol or 25(OH)D) <20 ng/ mL. Two hundred and sixteen patients met the inclusion criteria and 206 were included in the study. Structured interviews followed by clinical examinations were used to collect data that were relevant to the study objectives, such as gender, age, race / color (self-definition), smoking, presence of hypertension, diabetes, and medications used.
A digital weight scale (Filizola) and a measuring tape were used to measure weight (kg) and height (m), respectively. Body mass indexe (BMI: weight / height 2 ), were classified according to the World Health Organization 11 . Waist circumference was measured with a measuring tape at the umbilicus level with patients wearing light clothes, standing with arms at the sides and at the end of expiration, and results were approximated to the nearest centimeter.
The vit D levels were analyzed from serum samples, where the 25(OH)D was measured by chemiluminescence (QL) and 1.25(OH)2D by high efficiency liquid chromatography (HPLC).

antiOxidant defenSe and Oxidative StreSS markerS
TRAP was evaluated by luminescence in a technical adaptation described by Repetto et al. 12 . Lipid hydroperoxide dose (CL-LOOHs) was measured by an adapted technique described by Flecha et al. 13 . The estimate of nitric oxide metabolites was assessed by determination of serum concentration of NOx by an adaptation of the technique described by Navarro-Gonzalves et al. 14 . To quantify the AOPP, a technique described by Witko-Sarsat et al. 15 was used. To quantify interleukin 6 (IL-6), leptin, adiponectin, FGF-23, and isoprostane ELISA kits were used (eBioscience, SPI-Bio, SPI-Bio, Cayman Millipore and Eagle Biosciences, respectively). Table 1 shows the baseline socio-demographic and clinical data of the individuals in our study. There were no significant differences in sex ratio, age, ethnicity, BMI, waist circumference, and number of subjects with diabetes, hypertension, obesity, metabolic syndrome (MetS), cardio-vascular events, and use of angiotensin-converting enzyme inhibitors (ACEI), angiotensin receptor blockers (ARB), or statins between subjects with a lower versus normal vit D levels. There was a significant association between the vit D groups and CKD stages.

dIscussIon
Our findings complement observations suggesting that 25(OH)D levels are positively associated with eGFR levels and alkaline phosphatase, and negatively with urea, creatinine, urinary protein/creatinine ratio, and total cholesterol. It is known that the majority of CKD patients have restricted protein and caloric intake, which contributed to the relatively low vit D levels seen in this population. Besides, many CKD patients have limited outdoor physical activities, with reduced exposure to sunlight, and greater loss of urinary vit D metabolites occurs in CKD patients with overt proteinuria, factors that also contribute to vit D deficiency. A Brazilian study investigated more than 1.800 CKD patients in various stages of the disease, with 25(OH)D deficiency (<22.0 pg/mL) in more than 60% of those with eGFR <30 mL/min/1.73m 2 16 .
Prevalence of 25(OH)D deficiency is common in CKD and it is implicated in PTH progressive increase that is seen with declining renal function and leads to secondary hyperparathyroidism, mineral bone disease, and increased cardiovascular risk 17 .
Albuminuria, increased urinary protein/ creatinine ratio, and low 25(OH)D level are typical findings in CKD patients 18 . According to Oh et al. 19 , LDL and total cholesterol levels were higher in CKD patients with vit D deficiency than the control group.
Our data did not corroborate the findings of Milovanova et al. 20 studies that showed lower FGF-23 in CKD 3b-4 stage patients.
Adipose tissue has been recognizing as an endocrine tissue releasing various factors involved in vascular functions, energy metabolism, or inflammation. Among them are interleukins such IL-6, tumor necrosis factor alpha (TNF-α), leptin, and adiponectin. In humans, hyperleptinemia has been shown to be associated with CKD progression 21 . Plasma levels of adiponectin are higher in CKD patients and correlate positively with CKD stage and albuminuria 22 .
It has been shown that C-reactive protein (CRP) concentrations seems to be associated with later stages of CKD and leptin, and adiponectin with BMI and sex. Numerous studies have reported an inverse association between renal impairment level and different mediators and biomarkers of inflammation including CRP, IL-6, TNF-α, and fibrinogen, pointing to a high prevalence of increased inflammation in CKD 23 .
The increased adiponectin, leptin, and IL-6 and dysregulation of 1.25(OH)2D are associated with increased FGF23 levels, and vit D deficiency is associated with low leptin. Notably, adiponectin levels are elevated by the degree of renal insufficiency 24 .
Rutkowski et al. 25 described that adiponectin primarily affected renal α-klotho expression and bone FGF23 release; high adiponectin levels suppress renal α-klotho secretion, reduce plasma FGF23 levels, and cause renal loss of calcium.
A 2017 study reported lower serum adiponectin levels in patients with type 2 diabetes mellitus (T2DM) and stage 1 CKD than in patients with T2DM without renal pathology, with progressively higher levels coinciding with the deterioration of kidney function 26 .
According to Agarwal et al. 27 , OS has an important role in CKD disease; the biologic markers of OS are found increased in cell cultures, animal models, and end-stage renal disease patients on hemodialysis.
Kidney cells contain abundant mitochondria and therefore their dysfunction have a fundamental role in the progression of renal failure. Many reports point out to a role of mitochondria in the increased OS in kidney disease 28,29 . Deposition of uric acid crystals in proximal tubule cells is also associated with increased OS and activation of the inflammatory system 30 .
According to Dounousi et al. 31 , levels of different markers including plasma F2-isoprostanes, advanced oxidation protein products, and malondialdehyde are increased in patients with varying degrees of renal function, including patients with end-stage renal failure.
The major finding of this study was that no association was found between vitamin D deficiency and increased state redox in CKD patients. The OS biomarkers (NOx) was not associated with vit D deficiency, but increased NOx levels were significantly higher in more advanced stages of CKD. Although many studies have demonstrated that vit D deficiency is associated with inflammation and OS in CKD patients, recent studies failed to show this association 32,33 .
Our data showed an increased NOx in stages 4 and 5 CKD patients. It is known that NO is synthesized in several cell types, being the vascular endothelium the major source of NO synthesis. The measurement of serum NOx levels estimates basal NO generation by endothelial cells. Therefore, it may be speculated that increased serum NOx in CKD people may be due to endothelial NOS (eNOS) inhibition and inducible NOS (iNOS) overexpression 34 . In most cell types, including adipocytes, iNOS is induced by inflammatory signals when hypoxia favors cell necrosis, which is a process that recruits macrophages and other phagocytic cells, and creates a milieu favoring the perpetuation of inflammation within the adipose tissue 35 . Moreover, when superoxide anion and NO are produced simultaneously in close proximity, a reaction leads to the formation of peroxynitrite (ONOO -), and, subsequently, to hydroxyl radicals, amplifying the oxidative process on biomolecules 36 .
Uncoupling of eNOS and defective production of NO results in impaired vasorelaxation of renal resistance arteries, diapedesis of polymorphonuclear leukocytes and monocytes, and local procoagulant and a proaggregant conditions. Local vasoconstriction is further exaggerated by the loss of endothelial integrity. These vascular events are followed by tubular mechanisms that include not only the induction of iNOS and increased production of reactive oxygen intermediates in the renal epithelial cells, but also generation of ONOOby the infiltrating polymorphonuclear cells and macrophages 37 .
Some of the deleterious effects induced by hyperuricemia are similar to those associated with increased OS such as reduced NO bioavailability and endothelial dysfunction, vascular hypertrophy, and inflammation 38 . However, our findings did not reveal differences between the groups with and without vit D deficiency. The inhibitory influences include increase in the endogenous NOS inhibitors such as asymmetric dimethylarginine (ADMA) and decrease in activity of the NOS enzymes for many reasons, including the reduction in protein abundance events that reduce inherent enzyme activity and reduced availability of essential cofactors. In addition, deposition of advanced glycosylated end-products occurs in advanced renal disease, which decreases the access of NO to its target tissue and contributes to NO deficiency 39,40 .
Our study has limitations, such as the small number of stage 5 CKD, limiting our interpretation about the relationship of vit D and OS in that group. In addition, this was a single-center cross-sectional study, not allowing conclusion on the causality between vit D deficiency and redox state. Even with these limitations, we believe that our results may be reliable, as we included a relatively large number of patients and used a broad panel of OS biomarkers.

conclusIon
Our findings suggested that vit D deficiency may not play a very important role in the increased OS state seen in pre-dialysis CKD patients, which is clearly multifactorial. Due to the cross-sectional study design, we were unable to exclude that vit D supplements could ameliorate OS in similar patients, question that could only be answered by an adequately powered clinical trial.