Down‐regulation of human long non‐coding RNA LINC01187 is associated with nephropathies

Abstract Chronic kidney diseases affect a substantial percentage of the adult population worldwide. This observation emphasizes the need for novel insights into the molecular mechanisms that control the onset and progression of renal diseases. Recent advances in genomics have uncovered a previously unanticipated link between the non‐coding genome and human kidney diseases. Here we screened and analysed long non‐coding RNAs (lncRNAs) previously identified in mouse kidneys by genome‐wide transcriptomic analysis, for conservation in humans and differential expression in renal tissue from healthy and diseased individuals. Our data suggest that LINC01187 is strongly down‐regulated in human kidney tissues of patients with diabetic nephropathy and rapidly progressive glomerulonephritis, as well as in murine models of kidney diseases, including unilateral ureteral obstruction, nephrotoxic serum‐induced glomerulonephritis and ischemia/reperfusion. Interestingly, LINC01187 overexpression in human kidney cells in vitro inhibits cell death indicating an anti‐apoptotic function. Collectively, these data suggest a negative association of LINC01187 expression with renal diseases implying a potential protective role.


| INTRODUC TI ON
Kidney and common systemic diseases such as diabetes, hypertension and systemic lupus erythematosus lead to compromised functional integrity of the kidneys, culminating in chronic kidney disease (CKD). 1-6 CKD, defined as the decrease and gradual loss of filtering function, is one of the most prevalent health problems. [7][8][9][10] For most of its course, CKD is asymptomatic as the manifestation of the disease appears only when there is pronounced deterioration of the renal function. At this late stage, therapeutic intervention is of limited effectiveness. 11 Therefore, it is crucial to elucidate the molecular mechanisms underlying renal pathophysiology in order to develop new therapeutic insights and alleviate the symptoms of CKD.
Recent advances in genomic technologies [12][13][14][15] have begun to unveil the complex molecular networks involved in human renal pathologies. Interestingly, these approaches have linked differential expression of long non-coding RNAs (lncRNAs), a newly discovered class of regulatory RNAs, to the onset and progression of renal diseases. LncRNAs are defined as RNAs larger than 200 nucleotides long. 16,17 They are mostly transcribed by RNA polymerase II and can be modified by 5′ capping, 3′ polyadenylation and splicing. In many cases, their expression is regulated in a tissue-specific and spatio-temporal manner. [18][19][20] RNA sequence conservation of lncRNAs among species is observed at a much lower level as compared to protein-coding transcripts. 21,22 However, sequence conservation at the promoter sites is comparable to protein-coding genes, especially for functionally important lncRNAs. [23][24][25] Thus, lncRNAs constitute a critical new class of biomolecules for understanding cell physiology, organ homeostasis and pathological processes.
Herein, we report that LINC01187, a conserved lncRNA between mouse and human, is critically involved in nephropathies.
We found that LINC01187 is strongly reduced in human renal pathologies, including diabetic nephropathy (DN) and rapidly progressive glomerulonephritis (RPGN), as well as in murine models for kidney diseases including unilateral ureteral obstruction (UUO), nephrotoxic serum-induced glomerulonephritis (NTS-induced GN) and ischemia/reperfusion (I/R). Importantly, forced expression of LINC01187 in human kidney cells inhibits cell death suggesting an anti-apoptotic function. Together these data propose a negative association of LINC01187 expression with renal pathologies implying a protective role.  Table 2 in human kidney cells in vitro inhibits cell death indicating an anti-apoptotic function.

| Human samples
Collectively, these data suggest a negative association of LINC01187 expression with renal diseases implying a potential protective role.

K E Y W O R D S
apoptosis, Gm12121, in situ hybridization, LINC01187, long non-coding RNAs, renal diseases

| Unilateral Ureter Obstruction (UUO) mouse model
Adult male C57BL/6 mice at the age of 8 to 12 weeks, were operated in order to ligate the right ureter. During the surgery, the animals were randomly divided in three groups, sham operated (control), 2 days ligated and 8 days ligated, exactly as previously described. 14,26

| Glomerulonephritis model (nephrotoxic serum-induced, NTS)
The NTS experiments were performed on SV129 male mice aged 8-10 weeks according to a previously described protocol. 14 Mice received intravenous injections of totally 12 µl NTS/g body weight over 2 consecutive days (days 0 and 1) to induce crescentic GN, while control mice were injected with PBS. Periostin (Postn) expression was inhibited with a cocktail of two different ODNs specifically targeting Postn mRNA, designed using IDT (Integrated DNA Technologies) platform.
Scrambled (SCR) non-specific ODNs were used as control. The ODN sequences were modified with phosphorothioate to prevent their in vivo hydrolysis by nucleases (Sigma-Aldrich). For administration to mice, the ODNs were diluted in normal saline and placed in osmotic minipumps (Alzet, model 1002) which were subsequently implanted subcutaneously in mice, constantly releasing a dose of 0.25 µl per hour which corresponded to a release of 150 pmol/ODN/day. The implantation of mini-pumps was performed at day 3 after administration of NTS, taking one day for the pump to start functioning. Mice were euthanized 9 days after the first injection (n = 4 per group). Renal tissues were collected.

| Ischemia/Reperfusion model (I/R)
Eight-to 10-week-old male mice (n = 4 per group) were anesthetized with intraperitoneal injection of ketamine (100 mg/kg)/xylazine (10 mg/kg) and subjected to right kidney nephrectomy in order to enhance the I/R aggression on the remaining kidney. Sham-operated animals were used as controls. The left renal artery was clamped for 30 min of warm ischemia at 37°C followed by 24 or 72 h of reperfusion. After reperfusion, the mice were euthanized, and renal tissues were collected for subsequent analyses.

| Chromatin immunoprecipitation analysis on UUO mouse kidney
Chromatin immunoprecipitation (ChIP) was performed in kidneys of sham-operated and UUO mice to evaluate the promoters of lncRNA gene as previously described. 14 The antibodies used were against RNA polymerase II (Millipore), Histone F I G U R E 1 Expression analysis of selected lncRNAs in human renal biopsies. (A) Human Genotype-Tissue Expression analysis (GTEx Portal) indicates that LINC01187 is predominantly expressed in kidney tissue (cortex and medulla). TPM: transcripts per million. (B-K) Expression profiles of 9 lncRNAs with real-time RT-qPCR analysis in renal biopsies from kidney patients (Pt.) and healthy controls (HC). The lncRNA genes that were examined in human samples have been found deregulated in the kidney of UUO mouse model (conserved lncRNA genes). (J,K) LINC01187 lncRNA is significantly down-regulated in individuals with renal disorders compared to HC. The lncRNA levels of each lncRNA were normalized to GAPDH and expressed as relative fold of expression to the sample with the minimal fold of expression. *p < 0.05, **p < 0.01, ns: p ≥ 0.05 (t-test 2;2) 3 (tri-methylated K4) (ab8580, Abcam), and a normal rabbit IgG (sc-2027) was used as a control. Precipitated DNAs were examined by real-time qPCR using the following oligonucleotide primers for the promoter region of Gm12121 lncRNA: Gm12121-Prom (forward, 5′-GGCCTATTGCTCAGAGAGGA-3′; reverse, 5′-GGCCACTTGTGTGTCTGTGT-3′). All experiments were performed in triplicate. Analysis was carried out via input percent method (ThermoFisher Scientific guidelines). Statistical analysis was performed by Student's t test.

| RNA extraction and real-time RT-qPCR analysis
Total RNA was isolated by using the TRI reagent solution (Sigma) followed by treatment with RQ1 DNase (Promega).

| Microarray analysis of human kidney biopsies
Human kidney biopsy specimens and Affymetrix microarray expression data were obtained within the framework of the European Renal cDNA Bank -Kröner-Fresenius Biopsy Bank. 27 Biopsies were obtained from patients after informed consent and with approval of the local ethics committees.

| RNA probe construction for in situ hybridization
Digoxigenin-labelled RNA probes, sense (control) and anti-sense to human LINC01187 lncRNA, were prepared to be used for in situ hybridization on renal tissue sections. For the preparation of the RNA DIG probe, a region of LINC01187 gene sequence was isolated via PCR using as a template renal cDNA from healthy individuals obtained from the Institute of Pathology, RWTH, Aachen, Germany. T3 and T7 promoter sequences were added to primers when necessary for the function of T3 and T7 polymerases during the in vitro transcription for the generation of the probes.
Anti-sense DIG-RNA probe was made by using T7 polymerase, and sense probe (control) was made by using T3 polymerase. The and RNA probes clean-up were carried out using the NucleoFast ® 96 PCR kit (MN) according to manufacturers' instructions.

| In situ hybridization with Digoxigenin-RNA probes on paraffin-embedded sections
The procedure followed has been previously described. 29 The slides were deparaffinized and rehydrated as following:

| Proliferation and Apoptosis detection assays
We evaluated the role of LINC01187 in cell proliferation and apoptosis in HEK293 cells overexpressing LINC01187. Therefore, we deter-

| Identification of conserved human lncRNAs with potential involvement in renal diseases
Using genome-wide transcriptomic analysis we have previously identified several mouse lncRNAs differentially expressed in the kidney of UUO model. 14 These data raised the question of whether our newly identified lncRNAs are conserved and functionally important in humans.
To tackle this question, we initially compared the gene sequences of mouse lncRNAs with human homologues and other vertebrate species ( Figure S1). We also compared the promoter sequences of these genes, considering the significance of sequence conservation at the promoter level equally important. This notion is a commonly emerging paradigm in lncRNA analysis, where lncRNA genes found in syntenic genomic regions between species exhibit low conservation in the RNA sequence, yet they retain high conservation in the promoter sequence. 23 Figure 1J). Two alternative transcripts for LINC01187 have been reported in NCBI RefSeq annotation and Ensembl/GENCODE annotation. Using exon-specific primers for real-time RT-qPCR assays, we found that healthy subjects express the Ensembl/GENCODE annotated LINC01187 transcript variant with four exons ( Figure   S3). The differential expression of LINC01187 was further experimentally validated in an independent cohort of kidney samples from patients with renal fibrosis ( Figure 1K). These data raise the intriguing hypothesis that this conserved lncRNA (Gm12121 in mouse or LINC01187 in human) may be involved in kidney diseases.

| LncRNA Gm12121 is down-regulated in animal models of renal diseases
In agreement with the human data, Gm12121, the mouse homologue of LINC01187, is dramatically down-regulated in the kidney of UUO

| LINC01187 is down-regulated in diabetic nephropathy and rapidly progressive glomerulonephritis in humans
To further examine the correlation between the gene expression of this conserved lncRNA and human kidney diseases, we assessed the RNA expression of LINC01187 in the glomerular and tubulointerstitial compartments from patients with DN, focal segmental glomerulosclerosis (FSGS), minimal change disease (MCD) and RPGN in comparison with living donor (LD). We found that LINC01187 is significantly reduced in two specific renal diseases, DN and RPGN, but not in MCD or FSGS. The down-regulation in each disease is observed in both tubular and glomerular compartments (Figure 4).
Using in situ hybridization with a DIG-labelled, anti-sense riboprobe specific for LINC01187 ( Figure 5A), we tested whether downregulation of this lncRNA can be also detected with a morphological assay in the same pathological conditions (DN and RPGN) compared  Table 2). In agreement with the results from gene expression analysis, a strong reduction of LINC01187 is observed in DN and RPGN renal biopsies, while this reduction is not noticeable in biopsies from lupus nephritis (LN) patients ( Figure 5B-E). As an internal, negative control, we have also used a sense riboprobe not able to be hybridized with LINC01187 RNA. The LINC01187-specific signal was localized around glomeruli and the tubules, suggesting a predominant interstitial distribution.
In addition, we correlated the LINC01187 expression levels from in situ hybridization (shown in Figure 5b) with the corresponding eGFR rates (shown in Table 2). This correlation analysis indicates that expression levels of LINC01187 are positively correlated with eGFR rates (Figure S5), suggesting that its expression is related to an adequate kidney function. We have previously identified several lncRNAs associated with renal dysfunction in mouse disease models using genome-wide approaches. 14 Here, to further address the significance of these findings in the context of human kidney diseases, we first screened these murine lncRNAs, which are differentially expressed in the  at the RNA and chromatin organization levels. Interestingly, we documented that LINC01187 is significantly down-regulated in human renal pathologies, including DN and RPGN, as well as additional murine models of renal pathologies, including NTS-induced glomerulonephritis and I/R. Moreover, we were able to show its in situ expression pattern around glomeruli and tubules in healthy conditions compared to other kidney diseases, suggesting association of LINC01187 reduction with specific renal pathologies. In this respect, it will be interesting in future studies to assign its expression to one or more specific renal cell types.

| LINC01187 overexpression in human kidney cells protects from apoptosis
Another critical question arising from our data is about the underlying molecular mechanism that determines in which renal pathologies the expression of LINC01187 is reduced. We propose that  44 These observations may indicate the involvement of LCP2 in the proliferation and survival properties of renal cells. Thus, it would be extremely interesting to investigate in future studies the potential co-regulatory mechanisms and inter-play between these coding genes and LINC01187 non-coding gene. Besides, studies that will utilize knockout mice technology may further uncover the protective function of LINC01187 in renal tissue.
Our results from in situ hybridization studies suggested that a specific cell type of the renal interstitium may be responsible for the expression of LINC01187 in normal tissue. Our attempts to identify this cell type by culturing existing renal interstitial cell lines were not successful. This is not surprising, since several distinct cell types may be harboured in the renal interstitium, as recent studies suggest. 45 In conclusion, these data provide evidence that LINC01187 downregulation is involved in renal pathogenesis, whereas its expression may imply a protective role. The correlation between LINC01187 expression pattern and renal diseases provides novel insight into the mechanisms by which lncRNAs mediate renal pathogenesis. Finally, this is the first study, to our knowledge, that suggests a protective role for the newly discovered LINC01187 lncRNA in any tissue or organ.

ACK N OWLED G M ENTS
We would like to thank all participating centres of the European Renal cDNA Bank -Kröner Fresenius Biopsy bank (ERCB-KFB) and their patients for their cooperation.

CO N FLI C T S O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding authors, Aristidis Charonis and Panagiotis K Politis, upon request. Published datasets were also used in this study (GSE104954, GSE104948).