Introduction

Vascular calcification (VC) is one of the most common disorders in aging, but especially in patients with chronic kidney disease (CKD). The increase of reactive oxygen species (ROS) in response to overloads in phosphorus (P) which occurs during VC contributes to the phenotypic dedifferentiation of vascular smooth muscle cells (VSMC) from a properly muscular (contractile) phenotype to an osteoblastic phenotype ^1,2,3,4. In particular, hydrogen peroxide is considered one of the most common ROS, capable of inducing the expression of the osteoblastic transcription factor Cbfa1/RUNX2 and inducing calcification ^5,6.

MicroRNAs (miRs) are small, non-coding single stranded RNAs (~22 nucleotides) that mediate the posttranscriptional silencing of genes by binding through complementarity of bases to the 3 'UTR region of the target mRNAs. The miRs are involved in crucial biological processes, including cell proliferation, differentiation and tissue development ^7,8. Recent studies have shown that several miRs are important regulators of VSMC differentiation into cells resembling osteoblasts and, therefore, vascular calcification ^9,10,11,12.

The miR-377 appears as an important aging regulator by regulating, among others, superoxide dismutase 2 enzyme (SOD-2) ^13,14. SOD is the body's main antioxidant, which catalyzes the conversion of superoxide radicals to hydrogen peroxide. SOD-2 corresponds to the mitochondrial form of the enzyme. Several articles have considered the role of oxidative stress as an inducer of VC, so in the present study we propose to analyze the possible role of miR-377 as regulator of aortic mineralization in an in vivo model.

Material and methods

Statistic analysis

The SPSS 17.0 program was used for the statistical analysis of the results. In the case of variables with normal distribution, the comparison of the treatment groups was carried out using the analysis of variance (ANOVA) with the Bonferroni test. In the case of variables with non-normal distribution, the Kruskal-Wallis test was used.

Results

Nephrectomy decreased creatinine clearance, an effect that did not worsen in animals fed the high P diet (Table 1). There was a discrete but not significant increase in serum P in the nephrectomized animals, which became more noticeable in those who received the high P diet. However, small increases in serum P were associated with more significant increases in serum levels of FGF23. Serum FGF23 levels increased twice with nephrectomy, while serum P increased by less than 50%. Similarly, in animals that received the diet high in P (CRF+HP), FGF23 levels increased twice as much as those in the normal diet (CRF+NP), with an increase in serum P less than 25%.

Table 1 Biochemical and bone metabolism markers in the different treatment groups. BMD values in the three tibial segments analyzed in the different treatment groups 

ap<0.05 relative to the Sham group; bp<0.05 relative to the CRF+NP group.

Nephrectomy had no effect on BMD changes in tibia, although CRF+HP in BMD in proximal tibia decreased significantly with respect to the CRF+NP group and the Sham group (Table 1).

In the CRF+NP group, Ca content in the aorta was increased 3-fold with respect to the Sham group. The increase with respect to the Sham group was increased up to 17 times in the CRF+HP group. The miR-377 expression decreased by 65% with nephrectomy, with no additional effect observed with the high-P diet. In fact, creatinine clearance was the biochemical parameter that was most strongly associated with the expression of this miR (R=0.83, p=0.001). In contrast, serum P was not associated with miR-377 expression (r=-0.302; p=0.34).

Protein expression of SOD-2 was clearly increased with nephrectomy (more than 3-fold), an effect that became more marked in nephrectomized animals with the high-P diet (more than 6-fold).

Figure 1 Ca content in aortas of 7/8 nephrectomy rats fed a diet with normal P content (0.6%) (CRD+NP) and high P content (0.9%) (CRD+HP), sacrificed at 20 weeks. Data represent the mean ± standard deviation. ap<0.05 relative to the Sham group; bp<0.05 relative to the CRD+NP group 

Figure 2 Relative levels of miR-377 in aortas of 7/8 nephrectomy rats fed a diet with normal content in P (0.6%) (CRF+NP) and high P content (0.9%) (CRD+HP), sacrificed at 20 weeks. Data represent the mean ± standard deviation. ap<0.05 relative to the Sham group 

Figure 3 Protein expression of SOD-2 in the aortas of 7/8 nephrectomy rats fed a diet with normal content in P (0.6%) (CRF+NP) and high P content (0.9%) (CRD+HP), sacrificed at 20 weeks. A) Image of a Western blot representing one aorta of each group; B) graphical representation of the protein expression of SOD-2 in the different groups as mean ± standard deviation. ap 0.05 compared to the Sham group; bp<0.05 compared to the CRD+NP 

Discussion

In this study, a decrease in aortic expression of miR-377 with nephrectomy was observed in parallel with the increase in SOD-2 protein expression as a possible compensatory mechanism to eliminate superoxide free radicals produced as a consequence of oxidative stress generated by nephrectomy.

To date, several hundred miRs have been identified in the human genome by proposing that at least 50% of the genes coding for proteins are regulated by miRs ^15,16.

The increase in the protein expression of SOD-2, as a possible compensatory effect to curb oxidative stress and cellular apoptosis in the mitochondria ^4,5,17, has been observed in our group’s other in vitro studies exposing VSMC to calcifying media. In fact, in a recent article we observed using two-dimensional gel-free proteomics techniques that, in the presence of a calcifying stimulus with high doses of calcitriol, the expression of this protein increases 3 fold with respect to the cells cultured with a control medium ¹⁸. Another in vitro study in VSMC subjected to a calcification stimulus by excess Ca and P shows an increase in the protein expression of SOD-2 by Western blot ⁶.

However, unlike what we have reported in our work, most studies in the literature show SOD-2 decreases in the presence of increased oxidative stress. In fact, in a recent study in rats with CRF, it has been found that the administration of a high P diet supplemented with calcitriol induces a decrease in the protein expression of SOD-2 at the aortic level ¹⁹. Other studies in other tissue models (renal podocytes and mesangial cells) have found that, in the case of stimuli inducing oxidative stress, there is a decrease in SOD-2 with increases in miR-377 (20). These increases have also been implicated with the increase of cellular senescence ^14,21.

SOD-2 is also referred to as mitochondrial superoxide dismutase or manganese superoxide dismutase (MnSOD). It is responsible for the reduction of ROS toxic to mitochondria.

The fact that in our case an increase in SOD-2 is observed, in the cell’s attempt to defend itself against oxidative stress, accompanied by a decrease in miR-377 expression, may be attributed to different experimental conditions from the rest of the studies, as well as the time frame of renal damage, which in our case is always very long term. What does concur with all published reports is the opposite effect between miR-377 and SOD-2 ^14,20,21. That is, increases in SOD-2 lead to decreases in miR-377 expression and vice versa, so the possible usefulness of miR-377 as a biomarker of oxidative stress and vascular damage may be inferred.

Despite using a high P diet, serum P levels were very similar in the nephrectomized animal groups with normal or high P diet. One possible explanation is that 7/8 nephrectomy has not been aggressive enough so a lower grade of nephrectomy may be 5/6 or less. This would indicate maintaining a residual renal function that would avoid aggravating the renal damage caused by P ²². However, FGF23 notably increases with nephrectomy and also in animals that received the high-P diet. FGF23 is known to be the first biochemical parameter to rise after CRF ²³. FGF23 begins to increase in the plasma of patients with CRD from very early situations, and continues to increase steadily as glomerular filtration declines ²⁴. During the early stages of CRD, an overload of P would stimulate the synthesis of FGF23 at the osteocyte level, and would act on the remaining nephrons by increasing the fractional excretion of phosphate to maintain normal phosphatemia ²⁵. In fact, in our case, increases of less than 25% in serum P were accompanied by much higher FGF23 variations as a mechanism to compensate for the increase in P in order to curb their excess.

The fact that kidney dysfunction obtained by nephrectomy was lower than initially predicted is reflected in the BMD results in the tibia. There were no differences in BMD between Sham animals and nephrectomized animals, despite the time course of CRD (20 weeks) ²⁶. In the nephrectomized animals that received the high P diet, there was a decrease in BMD in proximal tibia compared to the nephrectomized animals with normal diet in P and in the Sham group. However, other studies in our group, with the same follow-up time, show a more negative effect on cortical bone (BMD in distal tibia) than on the trabecular (BMD in proximal tibia) when the diet administered is rich in P ²⁷. This effect is accompanied by severe secondary hyperparathyroidism that mainly affects the cortical bone, a result that we did not observe in the present study.

Despite the lower degree of renal dysfunction observed, we detected a relationship between the VC increase and the decrease in BMD, as previously described by our group ²⁸. The group of animals with CRF and high P diet showed a clear increase in calcium content in the aorta, which was accompanied by the higher losses of BMD in the proximal tibia.

From the results of this study, nephrectomy, independent of serum P levels, is capable of modifying miR-377 expression. The high P content in the diet induced increased protein expression of SOD-2, probably as a protective mechanism to avoid further vascular damage due to oxidative stress. These results alert us to the desirability of maintaining serum levels of P when renal deterioration is accentuated thus avoiding progressive vascular damage. The use of miR-377 as a biomarker of vascular damage due to oxidative stress requires mechanistic studies, but opens a possible preventive and therapeutic route to the development and progression of vascular lesions in CRD.