Notch3 deletion regulates HIV-1 gene expression and systemic inflammation to ameliorate chronic kidney disease

Abstract

Anti-retroviral therapy (ART) has decreased human immunodeficiency virus (HIV)-1-associated morbidity. However, despite ART, immune cells remain latently infected, leading to chronic inflammation and HIV-1-associated comorbidities. New strategies are needed to target viral proteins and inflammation. We found activation of Notch3 in renal cells of the HIV-1 transgenic mouse model (HIV-Tg26) and in patients with HIV-associated nephropathy. We hypothesized that targeting NOTCH3 activation constitutes an effective therapy for HIV-related chronic kidney disease. We generated HIV-Tg26 mice with Notch3 knocked out (Tg-N3KO). Compared to HIV-Tg26 mice at 3 months, Tg-N3KO mice showed a marked reduction in renal injury, skin lesions and mortality rate. They also showed reduced renal infiltrating cells and significantly reduced expression of HIV genes. Moreover, Notch3 activated the HIV long terminal repeat promoter, and induction of HIV-1 increased Notch3 activation, indicating a feedback mechanism. Further, bone marrow-derived macrophages from HIV-Tg26 mice showed activation of Notch3, indicating systemic effects. Consistent with that observation, systemic levels of TNF and MCP-1 were reduced in Tg-N3KO compared to HIV-Tg26 mice. Thus, Notch3 deletion/inhibition has a dual-therapeutic effect in HIV-related chronic kidney disease, which might extend to other HIV-related pathologies.

Keywords: HIV-1; Kidney; Notch3.

Fig. 1.

Activation of Notch3 in HIV-1-associated nephropathy. (A-D) Immunolabeling was performed for the presence of Notch3 in renal paraffin sections. Notch3 (green) labeling in kidney sections of 3-month-old wild-type (WT) (A,C) and HIV-Tg26 (Tg) (B,D) mice is shown. Arrows indicate glomerular and tubular interstitial cells highly positive for Notch3 expression. Asterisks indicate blood vessels in which Notch3 is normally expressed. (E,F) Quantification of glomerular (E) and tubular (F) Notch3 expression (intensity, green) as assessed using ImageJ. (G,H) Kidney biopsy from an unaffected individual (G) versus kidney biopsy from a patient with HIV-1-associated nephropathy (HIVAN) (H) (representative of n=3 in each group), showing nuclear NOTCH3 expression (arrows), indicating activation. Asterisks indicate blood vessels with bright Notch3 expression. Dashed line outlines glomerulus with no Notch3 expression. (I) Quantification of NOTCH3 expression (intensity, green) in patients with HIVAN and unaffected individuals (NHK), as assessed using ImageJ. Unpaired two-tailed Student's t-test was used; data are presented as percentage area positive for green labeling. All experiments were performed in triplicates. *P<0.05, **P<0.01, ****P<0.0001. Scale bars: 50 µm.

Fig. 2.

Notch3 deletion improves disease progression and lifespan in HIV-Tg26 mice. (A) Kaplan–Meier curve showing the 6 months mortality rate in HIV-Tg26 (Tg) mice (n=26) and HIV-Tg26 with Notch3 knocked out (Tg-N3KO) mice (n=28). (B) Phenotypic appearance of WT, WT with Notch3 knocked out (N3KO), Tg and Tg-N3KO mice; note the skin papillomata on the forehead of the Tg mouse, whereas the Tg-N3KO mouse appears normal. (C,D) Skin lesions/mouse were quantified (n=9 each), and area per lesion was measured and expressed as surface area per cm². Unpaired two-tailed Student's t-test was used for statistical analysis. (E) Urine was collected in metabolic cages overnight from WT, N3KO, Tg and Tg-N3KO mice at 3 months of age before euthanasia. Proteinuria was assessed in 2 µl urine by SDS-PAGE followed by Coomassie Blue staining of the gels. Bovine serum albumin (BSA) was used as a positive control. (F) Albumin and creatinine ratio in urine was measured using ELISA (n=4, WT; n=4, N3KO; n=10, Tg; n=10, Tg-N3KO) and expressed as µg albumin/mg creatinine. One-way ANOVA. (G) Renal function was also assessed in serum using a blood urea nitrogen (BUN) assay kit, and results are expressed as BUN mg/dl (n=4, WT; n=4, N3KO; n=6, Tg; n=7,Tg-N3KO). One-way ANOVA. *P<0.05, ***P<0.001.

Fig. 3.

Notch3 deletion ameliorates kidney injury in HIV-Tg26 mice. Kidney sections from 3-month-old WT, N3KO, Tg and Tg-N3KO mice were stained with periodic acid–Schiff (PAS) to determine kidney injury. (D-F) PAS staining indicated tubulointerstitial injury (D), glomerular tubular injury (E) and inflammation (F). Note the severe kidney injury in Tg kidneys (n=9) compared to Tg-N3KO kidneys (n=12). Arrows in D indicate protein casts; arrows in E indicate comparisons between normal-looking (N3KO), Tg26 and Tg-N3KO glomeruli, showing the extent of sclerosis. (A-C) Quantitation of percentage tubulointerstitial injury (A), percentage glomerular injury (B) and percentage infiltration (C). Unpaired two-tailed Student's t-test was used for statistical analysis. ns, not significant; *P<0.05. Scale bar: 100 µm.

Fig. 4.

Notch3 targets HIV-1 activity. (A) Heatmap showing differential expression of genes obtained from mRNA sequencing of kidneys from 3-month-old WT, N3KO, Tg and Tg-N3KO mice. Note the asterisks for nef and env. (B,C) Quantitative PCR (qPCR) validating nef (B) and env (C) (n=5-7), followed by unpaired two-tailed Student's t-test for statistical analysis. (D) qPCR to compare Tg-N3KO and Tg-N4KO in repressing nef (n=3 per condition). One-way ANOVA followed by Tukey's test was used for statistical analysis. (E) HIV-long terminal repeat (LTR) promoter luciferase construct and Notch3 intracellular domain (N3IC) or PCDNA3.1 [empty vector (EV)] were transiently transfected into podocytes, followed by promoter reporter luciferase assays. Data were averaged from eight assays in three independent experiments. Vector-transfected control in each experiment was set to one, and relative fold change in LTR promoter activity was calculated. (F) Podocytes were transfected with HIV-LTR construct (PNL4) or EV. After 24 h, lysates were subjected to western blot analysis of N3IC and β-actin. Data were normalized to β-actin via ImageJ and presented as fold change (n=4). Unpaired two-tailed Student's t-test was used for statistical analysis. *P<0.05, ****P<0.0001.

Fig. 5.

Notch3 targets immune cell infiltration. (A-C) Volcano plots comparing gene expression in kidneys from Tg versus WT (A), Tg-N3KO versus WT (B), and Tg versus TG-N3KO (C) mice. (D) qPCR for Ccl2 expression. One-way ANOVA followed by Tukey's test was used for statistical analysis. (E) Immunohistochemistry (IHC) for MMP10 and NFκB (p65) in renal sections of WT, N3KO, Tg and Tg-N3KO mice. Arrows show glomerular and tubular areas with high expression. Scale bar: 100 µm. (F) Paraffin sections from kidneys of WT, N3KO, Tg and Tg-N3KO mice, labeled for immune cell markers CD68, FOXP3 and CD8. Arrow indicates CD68⁺ macrophages found in clumps in Tg kidney. Scale bar: 50 µm. (G) Quantification of CD68⁺, FOXP3⁺ and CD8⁺ HIVAN cells from Tg and Tg-N3KO mice, assessed in Tg (n=9) and Tg-N3KO (n=9) kidneys unaware of experimental group. Each dot represents the percentage of positive cells from the entire kidney where inflammatory invasions were prominent. Unpaired two-tailed Student's t-test was used for statistical analysis. (H) Serial sections from a kidney biopsy from a patient with HIVAN immunostained for the presence of CD68 and NOTCH3. Note the presence of CD68⁺ cells in and around glomerulus; the same areas were also labeled brightly for NOTCH3 (arrows). Scale bar: 100 µm. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.

Fig. 6.

Notch3 deletion targets systemic inflammation in HIV-Tg26 mice. (A) Schematic showing macrophage isolation from bone marrow in mice. (B) Labeling of differentiated bone marrow cells expressing the macrophage marker F4/F80 in fixed cells. Scale bar: 50 µm. (C) Lysates obtained from differentiated macrophages of WT and Tg mice were subjected to western blot analysis for N3IC, Dll4 and jagged 1 (J1). CD68 was used as a loading control. Ponceau S provides an additional loading control. (D-F) Quantification of N3IC (D), Dll4 (E) and jagged 1 (F) in bone marrow cells obtained from three mice in each group (n=3). Unpaired two-tailed Student's t-test was used for statistical analysis. (G,H) ELISA was performed for the presence of TNFα (TNF; G) and MCP-1 (H) in serum obtained from 3-month-old WT, N3KO, Tg and Tg-N3KO mice (n=6-12), both males and females. Data are expressed in pg/ml. One-way ANOVA followed by Tukey's test was used for statistical analysis. ns, not significant; *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.