Increased expression of HOXA11-OS, Cyr61, Beclin-1, and LC3B in the kidney tissue of lupus mice and serum of lupus patients

The expression of HOXA11-OS, Cyr61, and the autophagy factors Beclin-1 and LC3B in the kidney tissue of lupus mice and serum of lupus patients was detected by qRT-PCR and western blotting. Compared with the control group, the expression of HOXA11-OS, Cyr61, Beclin-1, and LC3B was significantly enhanced (P < 0.05) (Fig. 1A).

Fig. 1

HYPERLINK "sps:id::fig1||locator::gr1||MediaObject::0" HOXA11-OS, Cyr61, Beclin-1, and LC3B are highly expressed in kidney tissues, serum, and cell lines of LN. A Expression of HOXA11-OS, Cyr61, Beclin-1, and LC3B in the kidney tissue of lupus mice and serum of lupus patients as detected by qRT-PCR and western blotting. *P < 0.05 vs. control group. Data are presented as means ± standard deviation. B Podocytes induced by different concentrations of serum IgG (250, 500, or 1000 μg/mL) from lupus patients for 3, 6, and 12 h. Podocyte activity was detected by CCK-8. *P < 0.05 vs. 0 μg/mL; #P < 0.05 vs. 3 h. C Expression of HOXA11-OS, Cyr61, Beclin-1, and LC3B as detected by qRT-PCR and western blotting after podocytes were induced by different concentrations of IgG from lupus patients sera for 6 h. *P < 0.05 vs. control group. D Expression of Cyr61, Beclin-1, and LC3B as detected by qRT-PCR and western blotting after podocytes were induced by different concentrations of IgG from normal human serum for 6 h. E Expression of HOXA11-OS mRNA in control, NC-IgG (1000 μg/mL, 6 h) and LN-IgG (1000 μg/mL, 6 h) groups was detected by qRT-PCR. *P < 0.01 vs. control group

Serum IgG from lupus patients induces the expression of HOXA11-OS, Cyr61, Beclin-1, and LC3B in podocytes in a concentration-dependent manner

To establish a cell model, podocytes were induced with different concentrations of serum IgG from lupus patients (LN-IgG 250, 500, or 1000 μg/mL) for 3, 6, and 12 h. The results showed that cell viability decreased gradually with increased induction time and dose, and the half inhibitory concentration appeared at induction using IgG 1000 μg/mL for 6 h (Fig. 1B). The results showed that LN-IgG could inhibit the activity of podocytes, and podocytes induced at 1000 μg/mL LN-IgG for 6 h were used in the follow-up experiments.

According to the effects of LN-IgG on podocyte viability under the different conditions, we induced podocytes with different concentrations of LN-IgG for 6 h, and then collected them for qRT-PCR and western blot analyses. The expression of HOXA11-OS, Cyr61, Beclin-1, and LC3B were significantly increased, and the increase was most obvious when podocytes were induced by IgG at 1000 μg/mL for 6 h (P < 0.05) (Fig. 1C).

Subsequently, the effects of podocyte induction with LN-IgG or serum IgG for 6 h on the expression of Cyr61, Beclin-1, and LC3B were compared using the same method and different normal serum IgG concentrations (NC-IgG 250, 500, or 1000 μg/mL). There were no significant differences in the expression of Cyr61, Beclin-1, and LC3B (P > 0.05) between LN-IgG- and NC-IgG-induced groups (Fig. 1D). According to the above experimental basis, we detected the expression of HOXA11-OS in cells of Control, NC-IgG (1000 μg/mL, 6 h) and LN-IgG (1000 μg/mL, 6 h) groups by qRT-PCR. The results showed that, compared with the Control group, there was no significant difference in HOXA11-OS mRNA expression in NC-IgG group (P > 0.05), while the mRNA expression level of HOXA11-OS in LN-IgG group was significantly increased (P < 0.01) (Fig. 1E). These results suggest that serum IgG of lupus patients may regulate the expression of HOXA11-OS, Cyr61 and autophagy factors in podocytes.

HOXA11-OS regulates the expression of Cyr61 and autophagy and podocyte markers in podocytes

We successfully constructed HOXA11-OS overexpression and knockdown stable cell lines after lentivirus transfection of podocytes and verified HOXA11-OS expression by qRT-PCR. The expression of HOXA11-OS was significantly increased in the ex-HOXA11-OS group (P < 0.001), while that of HOXA11-OS#1 and #3 was significantly decreased in the sh-HOXA11-OS group. HOXA11-OS#1 decreased most significantly, and the knockdown effect was the most pronounced (P < 0.001). Hence, HOXA11-OS#1 was selected for further experiments (Fig. 2A).

Fig. 2

Expression of Cyr61, autophagy factors, and podocyte markers in podocytes regulated by HOXA11-OS. A HOXA11-OS mRNA expression was detected by qRT-PCR after HOXA11-OS overexpression and knockdown lentiviruses were successfully transfected into podocytes. *P < 0.001 vs. ex-NC; *P < 0.001 vs. sh-NC. B After HOXA11-OS overexpression and knockdown lentiviruses were successfully transfected into podocytes, the expression of Cyr61, Beclin-1, LC3B, nephrin, and podocin was detected by qRT-PCR and western blotting. *P < 0.01 vs. ex-NC; #P < 0.01 vs. sh-NC. C After HOXA11-OS overexpression and knockdown lentiviruses were successfully transfected into podocytes, the localization and expression of Cyr61, LC3B, and nephrin proteins in each group were detected by IF. Relative average optical density; *P < 0.01 vs. ex-NC; #P < 0.01 vs. sh-NC. Scale bars represent 15 µm

qRT-PCR and western blot analyses were also used to detect the expression of Cyr61, Beclin-1, LC3B, and the podocyte markers nephrin and podocin. Compared with ex-NC, the expression of Cyr61, Beclin-1, and LC3B in the ex-HOXA11-OS group was significantly increased, while that of nephrin and podocin was significantly decreased (P < 0.01). Compared with sh-NC, the expression of Cyr61, Beclin-1, and LC3B in the sh-HOXA11-OS#1 group was significantly decreased, while that of nephrin and podocin was significantly increased (P < 0.01) (Fig. 2B).

In addition, the localization and expression of Cyr61, LC3B, and nephrin proteins were observed by IF. Cyr61, as a secretory protein, was expressed in the nucleus and cytoplasm, nephrin was expressed in the cell membrane, and LC3B was expressed in the cytoplasm. The expressions of Cyr61, LC3B, and nephrin obtained by IF were consistent with those obtained by western blotting (Fig. 2C). These results suggest that HOXA11-OS may affect podocyte function by regulating Cyr61 expression and autophagy.

miR-124-3p directly targets the 3' untranslated region of HOXA11-OS and Cyr61 in podocytes

Firstly, the subcellular localization of HOXA11-OS was detected by FISH. The expressions of 18S, U6, and HOXA11-OS were similar, and there was no significant difference in optical density (P > 0.05). HOXA11-OS was distributed in both the nucleus and cytoplasm but mainly in the cytoplasm (Fig. 3A). In addition, qRT-PCR results showed that miR-124-3p expression was low in both the kidney tissue of lupus mice and podocyte injury model when compared with the control group (P < 0.05) (Fig. 3B). In the podocyte injury model, overexpression of HOXA11-OS significantly decreased the expression of miR-124-3p, while HOXA11-OS knockdown significantly increased the expression of miR-124-3p (P < 0.001). Furthermore, the overexpression of miR-124-3p significantly decreased the expression of HOXA11-OS, while miR-124-3p knockdown significantly enhanced the expression of HOXA11-OS (P < 0.001) (Fig. 3C). These results indicate that the expression of HOXA11-OS is negatively correlated with that of miR-124-3p and that HOXA11-OS may regulate miR-124-3p as a ceRNA.

Fig. 3

miR-124-3p directly binds to the 3’ untranslated regions of HOXA11-OS and Cyr61 in podocytes. A The subcellular localization of HOXA11-OS was detected by FISH assay. The expression of 18S, U6, and HOXA11-OS were similar, and there was no statistical difference in optical density. B The expression of miR-124-3p in the kidney tissue of lupus mice and podocyte injury model was detected by qRT-PCR. *P < 0.01 vs. control group. C The expression of HOXA11-OS in miR-124-3p mimic or inhibition cell lines and the expression of miR-124-3p in HOXA11-OS overexpression or knockdown cell lines were detected by qRT-PCR. *P < 0.001 vs. miR-NC/ex-NC/sh-NC. D The binding sites of miR-124-3p on HOXA11-OS and Cyr61 in HOXA11-OS-wild-type (WT) and Cyr61-WT, and the mutation sites in HOXA11-OS-mutant (MUT) and Cyr61-MUT are indicated in red. Bi-luciferase reporter genes were detected 48 h after transfection of HOXA11-OS-WT, Cyr61-WT, HOXA11-OS-MUT, or Cyr61-MUT and miR-124-3p mimic or miR-NC. **P < 0.01 vs. HOXA11-OS-WT + miR-NC/Cyr61-WT + miR-NC

Data in the Bibiserv database (https://bibiserv.cebitec.uni-bielefeld.de/rnahy Brid? ID = rnahybrid_view_submission) and Targetscan database (http://www.target-scan.org) predicted miR-124-3p targeted binding sites in both HOXA11-OS and Cyr61. The double luciferase reporter gene assay showed that the miR-124-3p mimic significantly reduced luciferase activity in cells transfected with the HOXA11-OS or Cyr61 wild type reporter gene but had almost no inhibitory effect on cells transfected with HOXA11-OS or Cyr61 mutant reporter gene (Fig. 3D). In conclusion, miR-124-3p can bind directly to the predicted binding sites of HOXA11-OS and Cyr61.

HOXA11-OS mediates Cyr61 regulation of autophagy factors and cell damage by targeting miR-124-3p

The experiments described above showed that the expression of miR-124-3p was opposite to that of HOXA11-OS and Cyr61, both containing direct targeting binding sites of miR-124-3p. This suggested that HOXA11-OS could mediate Cyr61 to regulate autophagy factor expression in podocytes by sponging miR-124-3p, thereby affecting podocyte function. To validate this hypothesis, a miR-124-3p-overexpressing lentivirus was transfected into the HOXA11-OS stable cell line. Cyr61 expression, levels of autophagy factors, and podocyte injury were detected by qRT-PCR and western blotting. The overexpression of Cyr61 and abnormally elevated levels of the autophagy factors Beclin-1 and LC3B induced by HOXA11-OS overexpression were downregulated, the podocyte marker proteins nephrin and podocin were upregulated, and podocyte injury was alleviated by miR-124-3p overexpression (P < 0.05) (Fig. 4A). After transfection of miR-124-3p knockdown lentivirus into the constructed knockdown HOXA11-OS stable cell line, the low expression of Cyr61 and the decreased levels of Beclin-1 and LC3B due to HOXA11-OS knockdown were reversed to some extent, while nephrin and podocin were downregulated and podocyte injury was aggravated (P < 0.05) (Fig. 4B). In addition, we observed the localization and expression of Cyr61, LC3B, and nephrin proteins by IF. The localization of Cyr61, LC3B, and nephrin proteins was the same as described above, and the protein expression trend was consistent with that detected by western blotting (Fig. 4C, D).

Fig. 4

HOXA11-OS regulates autophagy factors and cell damage by targeting miR-124-3p mediating Cyr61. A The expression of Cyr61, Beclin-1, LC3B, nephrin, and podocin was detected by qRT-PCR and western blotting after transfection of miR-124-3p mimic into the ex-HOXA11-OS cell line. B The expression of Cyr61, Beclin-1, LC3B, nephrin, and podocin was detected by qRT-PCR and western blotting after transfection of miR-124-3p inhibition into the sh-HOXA11-OS#1 cell line. C, D The localization and expression of Cyr61, LC3B, and nephrin were detected by IF. *P < 0.05 vs. miR-NC; #P < 0.05 vs. ex-HOXA11-OS + miR-NC/sh-HOXA11-OS#1 + miR-NC. Data are presented as means ± standard deviation. Scale bars represent 15 µm

Downregulation of Cyr61 can reverse the abnormal autophagy and podocyte injury induced by HOXA11-OS

In the present study, the expression of Cyr61 was downregulated by knocking down the lentivirus-transfected podocytes (Fig. 5A). The qRT-PCR and western blotting results revealed that the expression levels of HOXA11-OS, Cyr61, Beclin-1, and LC3B were significantly decreased, while nephrin and podocin were significantly increased in stable podocyte lines with downregulated Cyr61 (Fig. 5B). These results suggest that Cyr61 downregulation may reduce podocyte injury by downregulating the high expression of HOXA11-OS and reducing the abnormally elevated podocyte autophagy. Subsequently, the Cyr61 knockdown lentivirus was transfected into the constructed HOXA11-OS overexpressed stable cell line, and the expression level of the target gene was detected by qRT-PCR, western blot, and IF analyses. The results showed that the abnormal increase of autophagy caused by HOXA11-OS overexpression could be reversed to some extent, and the expression levels of Beclin-1 and LC3B decreased significantly, while that of nephrin and podocin increased significantly (P < 0.05); podocyte injury was alleviated (Fig. 5C, D). The recovery validation test showed that Cyr61 downregulation could alleviate the abnormally elevated autophagy of podocytes induced by the overexpression of HOXA11-OS and reduce podocytes damage.

Fig. 5

The downregulation of Cyr61 gene can reverse the abnormal autophagy and podocyte injury induced by HOXA11-OS. A The Cyr61 gene was downregulated by knockdown lentivirus transfected into podocytes, and its expression was detected by qRT-PCR and western blotting. B The expression of HOXA11-OS, Beclin-1, LC3B, nephrin, and podocin was detected by qRT-PCR and western blotting after Cyr61 downregulation. *P < 0.05 vs. sh-NC. C The expression of Beclin-1, LC3B, nephrin, and podocin was detected by qRT-PCR and western blotting after sh-Cyr61#1 was transfected into the ex-HOXA11-OS cell line. D The localization and expression of LC3B and nephrin were detected by IF. *P < 0.05 vs. ex-NC; #P < 0.05 vs. ex-HOXA11-OS. Data are presented as means ± standard deviation. Scale bars represent 15 µm

miR-124-3p inhibition leads to the deterioration of lupus mice with relieved injury after HOXA11-OS knockdown

To further verify the molecular mechanism of HOXA11-OS in the occurrence and development of LN in vivo, sh-HOXA11-OS and miR-124-3p inhibition AAVs were injected into mouse kidneys, and changes in the renal function and autoantibody levels were detected. The levels of 24 h urinary protein, BUN, Scr, as well as anti-dsDNA, ANA, and anti-Sm antibodies in the Lupus + sh-HOXA11-OS group were significantly lower than those in the Lupus group (P < 0.05), while those in the Lupus + miR-124-3p inhibition group were significantly higher (P < 0.05). After treatment with both sh-HOXA11-OS and miR-124-3p inhibition AAV, the levels of 24 h urinary protein, BUN, Scr, as well as the levels of anti-dsDNA, ANA, and anti-Sm antibodies in mice were significantly higher than those in the Lupus + sh-HOXA11-OS group (P < 0.05) (Fig. 6A).

Fig. 6

miR-124-3p inhibition leads to the deterioration of lupus mice but podocyte injury was alleviated after HOXA11-OS knockdown. Levels of 24 h urinary protein, BUN, Scr, and autoantibody in mice in each group. &P > 0.05, *P < 0.05 vs. sh-NC; #P < 0.05 vs. sh-HOXA11-OS; &*P < 0.05 vs. Lupus + sh-NC; ##P < 0.05 vs. Lupus + sh-HOXA11-OS. B,C hematoxylin–eosin and periodate-Schiff staining and histological score of kidney tissues of different groups: a sh-NC; b sh-HOXA11-OS; c miR-124-3p inhibition; d sh-HOXA11-OS + miR-124-3p inhibition; e Lupus + sh-NC; f Lupus + sh-HOXA11-OS; g Lupus + miR-124-3p inhibition; and h Lupus + sh-HOXA11-OS + miR-124-3p inhibition. Scale bars represent 20 µm. &P > 0.05, *P < 0.05 vs. sh-NC; #P < 0.05 vs. sh-HOXA11-OS; &*P < 0.05 vs. Lupus + sh-NC; ##P < 0.05 vs. Lupus + sh-HOXA11-OS

To further evaluate renal injury, Austin histological scores were used to assess global glomerular abnormalities and interstitial inflammation by examining HE- and PAS-stained renal tissue sections. The results showed contraction and deformation of glomeruli at different degrees, glomerular capillary stenosis and occlusion, mesangial cell and endothelial cell proliferation, and inflammatorcell infiltration in the renal interstitium in the Lupus group compared with the control group. After injection of sh-HOXA11-OS AAV, pathological lesions in the kidney tissue of mice in the Lupus + sh-HOXA11-OS group were significantly improved compared with those of mice in the Lupus group; the morphology of glomeruli was relatively normal, there was no obvious stenosis and occlusion of glomerular capillaries, the proliferation of mesangial cells and endothelial cells was reduced, and the number of infiltrated inflammatory cells was also decreased significantly. After injection of miR-124-3p inhibition AAV, the pathological lesions in the renal tissue of mice in the Lupus + sh-HOXA11-OS + miR-124-3p inhibition group were further aggravated compared to those of mice in the Lupus + sh-HOXA11-OS group. These results indicate that HOXA11-OS is involved in renal pathological processes and knocking down HOXA11-OS can alleviate renal injury in lupus mice; however, knocking down miR-124-3p may aggravate renal function injury in lupus mice by reversing the effect of HOXA11-OS knockdown (Fig. 6B, C).

Effect of HOXA11-OS on the expression of autophagy factors and podocyte markers in lupus mouse renal tissues by targeting miR-124-3p regulating Cyr61

The knockdown level of HOXA11-OS in mouse kidney tissues was detected by qRT-PCR after in situ injection of AAV (Fig. 7A). Our results confirmed that HOXA11-OS is highly expressed in LN, and the expression of HOXA11-OS can be downregulated by injecting AAV in vivo. The qRT-PCR and western blot analyses revealed that the expression of Cyr61, Beclin-1, and LC3B in lupus mice was significantly decreased (P < 0.01), while that of nephrin and podocin was significantly increased (P < 0.05) after HOXA11-OS knockdown compared with the control group, suggesting that renal inflammatory lesions were alleviated. The expression of Cyr61, Beclin-1, and LC3B was significantly increased (P < 0.01), while that of nephrin and podocin was significantly decreased (P < 0.05) after miR-124-3p expression was downregulated, suggesting that renal inflammatory lesions were aggravated. Compared with the Lupus + sh-HOXA11-OS group, the expression of Cyr61, Beclin-1, and LC3B in the kidney tissue of mice in the control group was significantly increased (P < 0.05), while nephrin and podocin expression was significantly decreased (P < 0.01) after knockdown HOXA11-OS and knockdown miR-124-3p AAVs were injected simultaneously, reversing the effect of HOXA11-OS knockdown and aggravating kidney damage (Fig. 7B, C). We further analyzed the protein expression of Cyr61, LC3B, and nephrin in mouse kidney tissues by IF, and found that results were consistent with those of western blotting (Fig. 7D). Our experiments further verified that HOXA11-OS can positively regulate the expression of Cyr61 and autophagy by targeting miR-124-3p, thereby affecting the renal function of lupus mice.

Fig. 7

Effect of HOXA11-OS on the expression of autophagy factors and podocyte markers in kidney tissues by targeting miR-124-3p regulating Cyr61. A The expression of HOXA11-OS mRNA was detected by qRT-PCR after AAV was injected into mouse kidney tissues. B The expression of Cyr61, Beclin-1, LC3B, nephrin, and podocin in kidney tissues of mice was detected by qRT-PCR and western blotting. C The expression of Cyr61, LC3B, and nephrin in kidney tissues of mice was detected by IF. &P > 0.05, *P < 0.05 vs. sh-NC; #P < 0.05 vs. sh-HOXA11-OS; &*P < 0.05 vs. Lupus + sh-NC; ##P < 0.05 vs. Lupus + sh-HOXA11-OS. Data are presented as means ± standard deviation. Scale bars represent 15 µm