Zinc As A Factor Affecting Serum Calcication Propensity in Patients With Type 2 Diabetes Mellitus

Background: Zinc inhibits vascular calcication in vivo and in vitro. Patients with type 2 diabetes mellitus show hypozincemia and are at an elevated risk of cardiovascular events. Recently, the in vitro test (T 50 test) was developed for the determination of serum calcication propensity. This cross-sectional study investigated the association between serum zinc and T 50 in type 2 diabetes mellitus patients and the effect of zinc on T 50 in vitro. Methods: The subjects were 132 type 2 diabetes mellitus patients with various kidney function. We measured T 50 levels by the established nephelometric method. Results: The median (interquartile range) levels of T 50 and serum zinc were 306 (269 to 332) min, and 80.0 (70.1 to 89.8) µg/dL, respectively. Serum zinc level was signicantly and positively correlated with T 50 (r s = 0.219, p = 0.012). This association remained signicant in multivariable-adjusted analysis, and was independent of known factors including phosphate, calcium, and magnesium. Renal function and glycemic control were not signicantly associated with T 50 . Finally, addition of physiological concentration of exogenous zinc chloride signicantly increased the serum T 50 in vitro. Conclusions: This is the rst report to investigate the association between serum calcication propensity and zinc levels in type 2 diabetes mellitus patients. Our data suggest that serum zinc is an independent factor that inhibits serum calcication propensity .


Background
Vascular calci cation is common and contributes to cardiovascular mortality in patients with type 2 diabetes mellitus [1,2], and those with chronic kidney disease [3,4] . The excess cardiovascular morbidity and mortality in those patients could be explained by redistribution and/or overload of calcium and phosphorus as well as imbalanced-calci cation regulators [3,5] in these conditions. The mechanism of vascular calci cation is supposed to be due to ectopic deposition of hydroxyapatite [6,7] induced by increased calcium-phoshprous product (Ca × P) in serum [8,9], dedifferentiation of vascular smooth muscle cells (VSMCs) into osteoblast-like cells [10,11] and accumulation of degenerative extracellular matrix [12]. It is hypothesized that diabetic condition and renal dysfunction share some common causal pathways leading to vascular calci cation.
In serum, precipitation of supersaturated calcium and phosphate is prevented by the formation of amorphous primary calciprotein particles (CPPs) [13,14]. Primary CPPs spontaneously convert into secondary CPPs, containing crystalline hydroxyapatite [13,14]. The in vitro test (T 50 -test) for the determination of serum calci cation propensity was developed [15]. This assay measures time required for primary CPPs to transform into secondary CPPs in the presence of supersaturating doses of calcium and phosphate, which increase turbidity of samples. T 50 can be measured by laser light scatter in turbid samples using nephelometry. Thus, a shorter T 50 means a higher calci cation propensity. Previous studies have shown that lower T 50 predicts vascular stiffness progression and all-cause mortality in patients with chronic kidney disease stage 3 and 4 [16], and all-cause mortality and cardiovascular composite endpoint in hemodialysis patients. [17]. A Lower T 50 was also shown to predict cardiovascular and all-cause mortality in renal transplant recipients [18,19].
T 50 is depended on the complex interplay of pro-calcifying (i.e. calcium and phosphate) and anticalcifying serum components (i.e. magnesium and albumin) [15]. Among them, a higher serum phosphate level was the factor most closely associated with lower T 50 [17,20]. Phosphate has been reported to induce calci cation of VSMCs in vitro [21]. Hyperphosphatemia is a risk factor for vascular calci cation and cardiovascular mortality, not only in patients with chronic kidney disease [22], but also in the general population [23]. Thus, suppression of phosphate-induced vascular calci cation is clinically important.
Zinc is an essential micronutrient that plays catalytic, structural, and regulatory roles [24]. Recently, zinc was found to inhibit phosphate-induced vascular calci cation, in vitro and in vivo [25]. In human, zinc level in blood was reported to be lower in patients with type 2 diabetes mellitus compared to non-diabetic subjects [26][27][28]. So far, however, the role of zinc in serum calci cation propensity is not established.
These previous studies raise the hypothesis that zinc could be one of the factors affecting with serum calci cation propensity. To test the hypothesis, we examined the association between serum zinc and T 50 levels and the effect of increasing zinc concentration on T 50 in vitro.

Ethics statement
This study followed the ethical guidelines for medical and health research involving human subjects by the Japanese Ministry of Health, Labour and Welfare, and the Declaration of Helsinki. This study was approved by the Ethics Committee of Osaka City University Graduate School of Medicine (approval No. 4100). Opt-out option for informed consent was performed as explained in instructions posted on the website of the institution.

Study design and participants
This study comprised of two parts. The rst part was a cross-sectional study using clinical data derived from our previous study including 143 patients with type 2 diabetes mellitus [29]. The inclusion and exclusion criteria for the clinical study were described as previously [29]. For this analysis, we excluded 11 participants because data of T 50 and zinc were not available. Finally, 132 patients were included in the present study. The second part was an in vitro study in which the effect of increasing zinc concentration on T 50 assay was examined.
Determination of calci cation propensity As previously reported [15], calci cation propensity was evaluated by overloading of calcium and phosphate into sera ex vivo. Spontaneous transformation of primary to secondary CPPs can increase turbidity in each serum in the presence of supersaturated solution with time. The light scattering intensity accompanied by progressive turbidness was measured by time-resolved nephelometry. T 50 was determined one-half maximal transition time, that is, a half-time of transformation from primary to secondary CPPs. According to the original method, we prepared three stock solutions as follows: (1) NaCl solution: 140mM NaCl, (2) Calcium solution: 40 mM CaCl 2 +100 mM HEPES+140 mM NaCl pH-adjusted with 10 M NaOH to 7.40 at 37°C, and (3) Phosphate solute on: 19.44 mM Na 2 HPO 4 +4.56 mM NaH 2 PO 4 +100 mM HEPES+140 mM NaCl pH-adjusted with 10 M NaOH to 7.40 at 37°C. In the 96-well plates, 20 µl of NaCl stock solution and 80 µl of serum were mixed in each well and then shacked for 1 minute. Subsequently, 50 µl of phosphate stock solution and 50 µl of calcium stock solutions were added and shacked for 1 minute, automatically, in the pre-warmed thermo-constant room at 34.5°C. The nal concentrations of calcium and phosphate in each sample were 10 mM and 6 mM, respectively. T 50 was determined in duplicate over 600 minutes per one measurement using a nephelometer (Nephelostar Plus R , BMG Labtech, Ortenberg, Germany).
All serum samples were measured in a blinded manner. Serum samples from healthy volunteers and dialysis patients were also measured as quality control in serum calci cation assay. The coe cients of variation (CV) of inter-and intra-assay were 4.4 % and 4.5 % in healthy control serum, 3.2 % and 4.5 % in hemodialysis control serum.

Blood and urine sampling and Measurements
Serum zinc levels were measured by a commercial laboratory (SRL Co., Ltd., Tokyo, Japan). Renal function was assessed by estimated glomerular ltration (eGFR) using a formula for the Japanese [30]. In this study serum calcium denotes calcium level adjusted for serum albumin according to Payne et al [31]. Urinary albumin to creatinine ratio was calculated as an index of albuminuria. Other measurements were obtained using routine laboratory methods at Osaka City University Hospital.

Other clinical information
We collected information on age, sex, height, weight, duration of diabetes, current medications, past history of cardiovascular disease (coronary artery disease, peripheral artery disease, aortic disease, and congestive heart failure requiring hospitalization), smoking habit, and laboratory data by asking the participants and/or by reviewing their medical records.
The diagnosis of type 2 diabetes mellitus was based on medical record and the criteria for diabetes mellitus as de ned in the Report of the Expert Committee on the Diagnosis and Classi cation of Diabetes Mellitus [32].

Statistics
In clinical study, we summarized continuous variables as medians (interquartile ranges, IQRs) and categorical variables as numbers and percentages. Correlations were analyzed according to the nonparametric Spearman's rank correlation test. Independent associations between the variables and T 50 were assessed by multiple regression analysis. In in vitro experiments in which ZnCl 2 was added, T 50 was expressed mean (SD) of the triplicate determinations, and comparison was made by one-way analysis of variance followed by Tukey's test. These statistical analyses were performed using GraphPad Prism version 6.0 (GraphPad Software, San Diego, CA, USA) or JMP software version 10 (SAS Institute, Inc., Cary, NC, USA). P-values < 0.05 by two-sided tests were considered statistically signi cant.

Results
Clinical characteristics of the type 2 diabetes patients  Correlations between serum calci cation propensity and clinical factors Table 2

Discussion
In the present study, we examined the association between serum zinc levels and serum calci cation propensity in patients with type 2 diabetes mellitus. Serum zinc level was signi cantly and positively correlated with T 50 in the present study. The positive correlation between serum zinc level and T 50 was  Table 3.Factors associated with serum calcification propensity (T 50 ) in 132 type 2 diabetes patients Data are the standard regression coefficients (b -value) and levels of significance (p-value) (bolded if p < 0.05).

In uence of zinc on serum calci cation propensity
To examine whether zinc directly increases T 50 , zinc was added in the serum calci cation propensity assay. Addition of exogenous ZnCl 2 signi cantly modi ed the T 50 level in pooled serum from healthy subjects (0 µM, 347 ± 0.8 min; 10 µM, 357 ± 5.6; and 20 µM min, 379.5 ± 4.2 min; p < 0.001, Fig. 2A), and pooled serum from dialysis patients (0 µM, 156 ± 2.3 min; 10 µM, 163 ± 0.5 min; and 20 µM, 170 ± 0.8 min, p < 0.001, Fig. 2B), respectively. also shown in the previous study including healthy subjects and patients with chronic kidney disease [25], indicating this correlation is common in various populations. However, so far, whether serum zinc level could be the independent factor associated with serum T 50 were not examined. We showed that serum zinc level was positively associated with T 50 independent of calcium, phosphate, and magnesium. These novel ndings suggest that zinc has an important role in suppressing calci cation propensity in serum.
We also con rmed that addition of zinc increases T 50 , in vitro assay. The mechanisms underlying zinc inhibits serum calci cation propensity were still unclear. Even in the polyethylene glycol hydrogels, not in serum, zinc inhibits transformation from amorphous calcium phosphate (ACP) into hydroxyapatite [33].
In additive -free composite, ACP transformed into brushite within minutes. In contrast, in the presence of zinc, zinc-doped ACP was very stable and did not show any signs of crystallization for up to 20 days. In ACP, zinc iron readily substitutes calcium [34], suppressing crystallization by decreasing solubility [35]. It is thus likely that zinc suppresses the transformation from amorphous primary CPPs into secondary CPPs, containing crystalline hydroxyapatite, in serum.
In a recent study by Voelkl et al, addition of exogenous ZnCl 2 (15 µM) did not improve T 50 in sera from healthy controls and patients on hemodialysis, although serum zinc level was signi cantly correlated with T 50 in those subjects [25]. The discrepancy between the studies by Voelkl et al by us may be explained by difference in ZnCl 2 concentration. In the present study, we demonstrated that 10 µM ZnCl 2 (= 60.5 µg/dL) did not signi cantly modify T 50 in serum from hemodialysis patients, which was consistent with the study by Voelkl et al [25]. In contrast, ZnCl 2 concentration of 20 µM (= 131 µg/dL), the upper limit of reference range, could signi cantly increase T 50 in those subjects. The crystallization inhibition has been reported to be dependent on the zinc concentration in polyethylene glycol hydrogels [33]. Thus, a certain zinc concentration may be required to increase serum calci cation propensity.
In the present study, magnesium was also signi cantly and positively associated with T 50 . Magnesium is one of the known anti-calcifying factors, which improves T 50 in ex vivo [15]. In vitro, magnesium has been reported to prevent phosphate-induced calci cation in human aortic VSMC [36]. Similarly, zinc could increase zinc nger protein TNF-α-induced protein 3 (TNFAIP3) expression, which subsequently inhibits NF-kB activation and osteo-/chondrogenic reprograming, resulting suppression of phosphate-induced VSMC calci cation [25]. The ndings of zinc on T 50 in the present study, and the above in vitro effects of zinc on phosphate-induced calci cation in VSMC were similar to those of magnesium. In addition, recently, a randomized control trial has shown that magnesium supplementation increased T 50 in patients with chronic kidney disease stage 3-4 [37]. Thus, supplementation of zinc, as well as magnesium, might be a potential therapeutic option to attenuate serum calci cation propensity and the progression of vasculature calci cation. Clearly, however, randomized clinical trials are needed, before such a treatment is recommended.
Albumin is also the anti-calcifying factors associated with T 50 in ex vivo [15]. When zinc and albumin are included simultaneously in multiple regression analysis, the signi cant associations of both factors with T 50 turned to be non-signi cant (data not shown). Serum zinc acts as an extracellular zinc buffer that controls zinc concentration in blood, since approximately 75-80% of zinc is bound to albumin, accounting for as much as 98% of the exchangeable fraction of zinc in blood [38,39]. Serum albumin was signi cantly and positively correlated with serum zinc levels in the study, thus the confounding effect might explain the results. Measurement of free-zinc irons will be needed to address this issue.
The present study has several limitations. First, the number of subjects examined was relatively small. Second, we cannot be sure whether the ndings of this study are applicable to non-diabetic patients. And third, due to the cross-sectional design, we can demonstrate only association, not causality. To con rm the potential bene ts of zinc supplementation, further interventional studies are required.

Conclusions
In summary, this is the rst report to investigate the association between serum calci cation propensity and zinc levels in patients with type 2 diabetes mellitus. Serum zinc was found as an independent factor associated positively with T 50 , and zinc has an in vitro effect on overall propensity of calci cation in serum.

Declarations
Ethics approval and consent to participate