MicroRNA-30 regulates left ventricular hypertrophy in chronic kidney disease

Left ventricular hypertrophy (LVH) is a primary feature of cardiovascular complications in patients with chronic kidney disease (CKD). miRNA-30 is an important posttranscriptional regulator of LVH, but it is unknown whether miRNA-30 participates in the process of CKD-induced LVH. In the present study, we found that CKD not only resulted in LVH but also suppressed miRNA-30 expression in the myocardium. Rescue of cardiomyocyte-specific miRNA-30 attenuated LVH in CKD rats without altering CKD progression. Importantly, in vivo and in vitro knockdown of miRNA-30 in cardiomyocytes led to cardiomyocyte hypertrophy by upregulating the calcineurin signaling directly. Furthermore, CKD-related detrimental factors, such as fibroblast growth factor-23, uremic toxin, angiotensin II, and transforming growth factor–β, suppressed cardiac miRNA-30 expression, while miRNA-30 supplementation blunted cardiomyocyte hypertrophy induced by such factors. These results uncover a potentially novel mechanism of CKD-induced LVH and provide a potential therapeutic target for CKD patients with LVH.

Introduction 1 Cardiovascular complications are now considered to be a major factor that 2 affects the prognosis of CKD patients (1,2). Increased knowledge about the 3 epidemiology of CKD-related cardiovascular disease gives us insight into the leading 4 cause of death for CKD patients (3-6). LVH is a common feature of cardiac changes 5 in CKD and contributes to more severe cardiovascular anomalies (7). LVH presents 6 even in very early CKD stages and occurs in up to 65% of predialysis patients (8).

7
Haemodynamic overload is a well-known inducer of LVH development in patients 8 with CKD (9), whereas other CKD-related detrimental factors, such as the renin- 9 angiotensin system (RAS), uraemic toxin, microinflammatory state, and phosphorus 10 metabolism disorder are all closely associated with LVH (10). Identification of 11 common mechanism in cardiac hypertrophy induced by such pathogenic factors can 12 provide an understanding of the CKD-induced LVH, leading to more reliable therapy 13 for CKD patients with cardiovascular complications. 14 MiRNAs are a class of short noncoding RNAs that regulate gene expression at 15 the posttranscriptional level (11), and they can be found in a large number of 16 biological processes, including cardiac hypertrophy (12). Numerous miRNAs with 17 important roles in cardiac hypertrophy have been identified, such as miRNA-1 (13-18 15), miRNA-9 (16), and miRNA- 133 (17-20). The miRNA-30 family (miR-30) is 19 highly expressed in the heart (21-23), and is closely related to cardiac remodelling 20 (24). It has been reported that miR-30 is downregulated in the heart from cardiac 21 hypertrophy or in heart failure models (25), and such an expression pattern can also 22 be found in diabetic cardiomyopathy (26). Mechanistically, miR-30 can regulate 1 autophagy (27-30), apoptosis (26,(30)(31)(32)(33)(34), and oxidative stress (30,32), which are all 2 related to cardiac hypertrophy. Hence, we hypothesized that disturbed expression of 3 myocardial miR-30 may act as a common mediator of LVH progression in CKD. 4 Here, we performed a series of experiments to confirm the role of miR-30 in the 5 development of CKD-induced LVH. We demonstrated that subtotal nephrectomy (SN) 6 is sufficient to induce LVH and miR-30 suppression in the heart. Cardiac miR-30 7 rescue inhibits the progression of LVH in these CKD models. Mechanistic 8 experiments further revealed that calcineurin signalling associates miR-30 with LVH. 9 In addition, we found that pro-hypertrophic stimuli are inducers of cardiac miR-30 10 suppression, while their pro-hypertrophic effects can be blocked by miR-30. These 11 results establish a model to illustrate the pivotal role of miR-30 in CKD-induced 12 LVH, and provide evidence for the potential therapeutic role of miR-30 in CKD-13 related cardiovascular complications.

15
Time-dependent effects of SN on the heart. To construct an appropriate CKD-induced 16 LVH model, we performed SN in male Sprague-Dawley (SD) rats (35). The body 17 weights of nephrectomized rats were reduced at 1 week after surgery, and delayed 18 body growth became increasingly obvious (Supplemental Figure 1A).  impairments were continuously observed via biochemical values of serum and kidney 20 tissue slice at the 1, 3 and 5 weeks after surgery ( Figure 1A). Renal insufficiencies 21 were apparent in SN rats, as indicated by increased serum creatinine and serum urea 22 nitrogen and a decreased creatinine clearance rate (Supplemental Figure 1, B to D). 1 The average systolic blood pressure and diastolic blood pressure showed a gradually 2 increasing trend in nephrectomized rats as well (Supplemental Figure 1, E and F).

3
Remnant kidney tissues from nephrectomized rats presented trends of tubular dilation 4 and interstitial fibrosis (Supplemental Figure 1G).

5
Sequential cardiac changes were analysed by echocardiography at the 1, 3 and 5 6 weeks after surgery ( Figure 1A). Echocardiographic analysis showed progressive 7 thickening of the left ventricular (LV) wall (Figure 1, C and D), and a relative 8 reduction of chamber diameter was observed at 3 weeks after surgery in SN rats 9 ( Figure 1E). Representative short-axis echocardiography and M-mode images are 10 shown in Figure 1B. Interestingly, nephrectomized rats exhibited a relatively higher 11 ejection fraction of LV, although statistical significance was only observed at 3 weeks 12 after surgery ( Figure 1F). Detailed measurements of short-axis echocardiography are 13 shown in Table 1. Moreover, there were significant increases in heart weight/tibial 14 length, LV weight/tibial length, and LV weight/heart weight ratios at 5 weeks after 15 surgery (Figure 1, G to I). Collectively, these results indicated that a CKD-induced 16 LVH model was successfully established in SD rats.

17
Cardiac miR-30 is downregulated in CKD. A previous study showed that miR-30a, 18 miR-30c, and miR-30d are the three most highly expressed types of miR-30 in mouse 19 (21) and rat (22) myocardial tissues; hence, we performed TaqMan quantitative 20 polymerase chain reaction (qPCR) to detect the alterations in these three members of 21 miR-30 in cardiac tissues from above-mentioned rat model. We found that myocardial 22 miR-30a, miR-30c, and miR-30d were downregulated, and the longer the kidney 1 injury lasted, the lower the cardiac miR-30 observed ( Figure 1J). In addition, we also 2 investigated the cardiac expression of miR-30e at 5 weeks after surgery, because it is 3 expressed at same levels as miR-30c in rat cardiac tissue (22). Consistently, we 4 observed reduced cardiac expression of miR-30e in CKD rats (Supplemental Figure   5 2). These results suggested a downregulated expression pattern of myocardial miR-30 6 in CKD.

5
Compared with those of the sham group, the body growth and renal function of 6 nephrectomized rats were significantly impaired. The body weights of 7 nephrectomized rats were significantly reduced, and AAV2/9 injection did not inhibit 8 this decrease (Supplemental Figure 5A). Moreover, the SN groups showed increased 9 serum creatinine and urea nitrogen with a decreased creatinine clearance rate, but 10 there was no difference between the SN groups that received AAV2/9-miR-30-

13
30SP mice developed significant cardiac hypertrophy at 24 weeks after birth, as 14 exhibited by increased LV wall thickness and relative wall thickness ( Figure 4, B and 15 C). Moreover, a decreased LV diameter was observed in 30SP mice ( Figure 4D).

16
However, LV ejection fraction did not change remarkably ( Figure 4E). Figure 4A   heart from 30SP mice demonstrated LV hypertrophy ( Figure 4A). Consistent with the 20 echocardiography data described above, the heart weight/tibial length, LV 21 weight/tibial length and LV weight/heart weight ratios of 24-week-old 30SP mice  Inhibition of miR-30 leads to calcineurin activation and pathological hypertrophy. We 6 collected LV tissues from 30SP mice to analyse the phenotype of these hypertrophic 7 hearts. Increased expression of hypertrophic genes and interstitial fibrosis are 8 common features of pathological cardiac hypertrophy (7, 43). We performed qPCR to 9 reveal the expression pattern of hypertrophic genes, and found that α-Mhc

18
In our previous study, we found that PPP3CA, PPP3CB and PPP3R1, all of 19 which are subunits of calcineurin, are direct targets of miR-30 (37). Hence, we 20 speculated that inhibition of miR-30 in cardiomyocytes would result in the activation 21 of the calcineurin pathway, which is considered as an critical mediator of cardiac 22 hypertrophy (38). Calcineurin activity assays showed a significantly enhanced 1 phosphatase activity of calcineurin in the myocardium of 30SP mice, which indicated 2 that calcineurin signalling might contribute to this hypertrophic phenotype ( Figure   3 5G). Correspondingly, cardiac PPP3CA was upregulated in 30SP mice ( Figure 5H).  Subsequently, we verified above mentioned results in vitro. We transfected a 10 miR-30 sponge plasmid into H9c2 cells, and found that NFATc3 was distinctly 11 localized in the nuclei of these miR-30 sponge-transfected cells ( Figure 6A).

12
Moreover, we found that the size of the miR-30 sponge-transfected cells was 13 increased, and this effect was inhibited by FK506, which is a specific inhibitor of 14 calcineurin (45) ( Figure 6, B and C). In support of this finding, the increased 15 expression levels of Anp were inhibited by FK506 ( Figure 6D), although there were 16 no significant changes in Bnp expression ( Figure 6E). Both Ppp3ca and Nfatc3 are 17 predicted to be miR-30 targets (37); therefore, we performed luciferase reporter 18 assays to confirm that Ppp3ca mRNA and Nfatc3 mRNA are direct targets of miR-30 19 in cardiomyocytes. For each sequence, we produced a construct containing the 20 luciferase coding region, followed by either the wild-type (WT) 3′-untanslated region 21 (UTR) or a mutant 3′-UTR ( Figure 6F). When cultured H9c2 cells were cotransfected 22 with the miR-30 sponge plasmid, the WT reporter exhibited higher luciferase activity 1 than the mutant reporter ( Figure 6G).

22
As pro-hypertrophic factors, FGF-23 and Ang-II are two major contributors to 1 LVH progression in CKD; therefore, we speculated that these factors also result in 2 cardiac miR-30 suppression under CKD conditions. To test this hypothesis, we treated 3 SN rats with the pan-FGFR inhibitor PD173074 or angiotensin receptor blocker 4 (ARB) once daily for 5 weeks ( Figure 8A). PD173074 and ARB both had no effects     In a previous report, FGF-23-induced cardiac hypertrophy was mediated by 3 calcineurin pathway (46). Thus, we speculated that miR-30 could suppress FGF-23-4 induced hypertrophy by inactivating calcineurin signalling. Indeed, miR-30 blunted 5 the nuclear translocation of NFATc3 (Supplemental Figure 11C) in FGF-23-treated 6 mice, although it did not affect the expression of PPP3CA (Supplemental Figure 11B). Although studies have revealed that miR-30 is expressed in cardiac fibroblasts 5 and is involved in cardiac remodelling (25, 58), our study, and others, demonstrated 6 that miR-30 can act in cardiomyocytes to regulate cardiac remodelling. We provided 7 several points to support this. First, hypertrophic stimuli suppressed miR-30 8 expression in cardiomyocytes in vitro (Figure 7), and this an effect was attenuated by 9 miR-30 mimics transfection ( Figure 9). Second, we chose a cardiomyocyte-specific 10 cTNT promoter to control the expression of miR-30 in AAV2/9-miR-30-Zsgreen 11 (Supplemental Figure 3); therefore, AAV2/9-miR-30-Zsgreen-injected CKD rats 12 showed reduced cardiac hypertrophy, suggesting that cardiomyocyte expressed miR-  Notably, miR-30 knockdown was only induced in cardiomyocytes, whereas these 20 mice not only demonstrated LVH, but also significant cardiac fibrosis ( Figure 4A).   Collectively, miR-30 exerts its function by simultaneously regulating distinctive 7 pathways, which means it is suited to protect cardiomyocyte from concurrent multiple 8 detrimental factors.

9
Of note, miRNAs can not only bind to the 3'-UTR of mRNA to inhibit mRNA 10 translation or even cut mRNA but also bind to the 5'-UTR of mRNA to activate 11 translation (69). We further speculated that miR-30 may be able to inhibit the  (Figure 11). We believe these data not only uncover a novel mechanism of 4 CKD-induced LVH but also hint at a potential therapeutic target for CKD patients 5 with cardiac hypertrophy.    MiR-30 expressing AAV2/9 preparation and jugular vein injection. AAV2/9-miR-30-1 Zsgreen and AAV2/9-Zsgreen were generated by HanBio, Inc. Briefly, FGF-23-2 treated mice and SN rats were placed in the supine position and were anaesthetized 3 with isoflurane. Then, an incision was made in the subcutaneous tissue of the neck 4 and the subclavian vein was located. Subsequently, we punctured the jugular vein via 5 an insulin syringe and slowly injected 10 11 vg AAV2/9-miR-30-Zsgreen or 10 11 vg 6 AAV2/9-Zsgreen. After injection, the wounds were closed with sutures.  Blood that derived from FGF-23-treated mice was also centrifuged at 4,500 g 8 and 4°C for 10 minutes and was used for circulating FGF-23 assays (Kainos 9 Laboratories, Inc.).  Primers. See Supplemental Table 1 for primer sequences. Statistical Analysis. Data were presented as mean ± SD or median with 5 interquartile range. Normal distribution of data was analysed by Shapiro-Wilk test.

6
Differences between 2 groups were analysed using Two-tailed Student's t test or 7 Mann-Whitney U test. One-way ANOVA test or Kruskal-Wallis test were used for 8 comparisons between multiple groups, followed by Tukey's multiple comparisons test  The author declares no conflicts of interest.   ± SD. n = 6 rats per group.  ± SD. n = 3 to 6 rats per group.