(A) All-trans retinoic acid (atRA) inhibits HIV-1–induced podocyte proliferation. After differentiation at 37°C for 10 d, Mock or HIV-1–infected podocytes were plated on collagen-coated six-well plates at 20,000 cells/well with or without atRA (10 μM) for 7 d. Bars represent mean cell number ± SD of six samples. (B) atRA inhibits HIV-induced loss of contact inhibition. Mock- or HIV-1–infected podocytes were plated on collagen-coated six-well plates at 80% confluence. Cells were cultured further with or without atRA (10 μM). Photographs were taken on day 7. C. Dosage response of retinoids on podocyte proliferation. After differentiation, control podocytes (◆) and HIV-infected podocytes (■) were incubated in collagen-coated dishes for 3 d with or without atRA at concentrations of 0.1, 0.5, 1, or 10 μM. Cell number was counted. Means ± SD of three independent experiments are shown. *P < 0.001 verus cells without atRA treatment. (D) Effects of atRA on mRNA expression levels of WT-1, synaptopodin, cyclin A, cyclin E, and glyceraldehyde-3-phosphate dehydrogenase (G3PDH). Normal podocytes (Mock) or HIV-1–infected podocytes (HIV) were cultured for 3 d with or without atRA (10 μM). Total RNA (10 μg/lane) was analyzed by Northern blotting. This is a representative blot of three independent experiments. (E) Dosage response of retinoids on synaptopodin expression. HIV-infected podocytes were treated with various concentrations (0.1 to 10 μM) of atRA or 9-cis-RA for 3 d. Total RNA (10 μg/lane) was analyzed by Northern blotting for synaptopodin and GAPDH. This is a representative blot of three independent experiments. (F) Effects of atRA on podocin expression. HIV-infected podocytes were treated with atRA (1 μM) for 3 d. Podocin expression was determined by Western blot using anti-podocin antibody. This is a representative blot of three independent experiments. (G) Effects of atRA on nephrin expression. HIV-infected podocytes were treated with atRA (1 μM) for 3 d. Nephrin/tubulin ratios were determined by real-time PCR. The fold of increased as compared with control podocytes are expressed. *P < 0.05 versus control (CL); **P < 0.05 versus HIV-infected podocytes; n = 3. Magnification, ×200.

(A) HIV reduces expression of retinoic acid receptor α (RARα) and retinoid X receptor α (RXRα). Northern blotting of RARα, RARβ, RARγ, RXRα, and RXRβ in control podocytes (Mock) and HIV-1–infected podocytes (HIV) that were treated with or without atRA (10 μM). HIV reduces the expression of both RARα and RXRα. Twenty micrograms per lane of total RNA was loaded. The representative blots of three independent experiments are shown. (B) Effects of RARα antagonist on atRA-induced cAMP production. After differentiation at 37°C, podocytes were cultured on collagen-coated six-well dishes and stimulated with atRA (1 μM) for 2, 5, and 10 min with or without preincubation with 10 μM of RARα antagonist (Ro-41-5253). DMSO served as a control. All cells were pretreated with rolipram. Intracellular cAMP level was determined using ELISA. A representative experiment from three independent assays is shown. (C) Effect of RXR agonist on intracellular cAMP production. Podocytes were cultured on collagen-coated six-well dishes and stimulated with RXR agonist (Ro-25-7186; 0.1 and 1 μM) for 2, 5, and 10 min after preincubation with rolipram. Intracellular cAMP level was determined using ELISA. A representative experiment from three independent assays is shown. (D) Effects of RARα and RXR agonists and RARα antagonist on HIV-induced podocyte proliferation. After differentiation, control podocytes and HIV-infected podocytes were incubated in collagen-coated dishes for 3 d with DMSO (control), atRA (1 μM), RARα agonists (Am580; 1 μM), RXR agonist (Ro-25-7386; 1 μM), 9-cis-RA (1 μM), or both atRA and RARα antagonist (Ro-41-5253). Cell number was counted. Mean ± SD of three independent experiments is shown. *P < 0.001. (E) Effects of RARα and RXR agonists and RARα antagonist on synaptopodin expression in HIV-infected podocytes. HIV-infected podocytes were treated with atRA (1 μM), 4-hydroxyphenylretinamide (poor activator of RAR; 1 μM), Am580 (RARα agonist; 1 μM), Ro-23-4272 (RARα agonist; 1 μM), or Ro-25-7386 (RXR agonist; 1 μM) for 3 d. Additional cells were preincubated with Ro-41-5253 (RARα antagonist; 10 μM) and then treated with atRA (1 μM). Total RNA (10 μg/lane) was analyzed by Northern blotting for synaptopodin and GAPDH. This is a representative blot of three independent experiments.

(A and B) Rp-cAMP (Rp) blocks antiproliferative effect of atRA on podocytes. Differentiated podocytes were pretreated with Rp-cAMP (100 μM) for 30 min and then were stimulated with or without atRA (10 μM) for another 3 d. Cells were lysed for MTS assay using the Promega kit (A) or for counting cell number (B). Mean ± SD of four independent experiments is shown. *P < 0.001. (C) A synergic effect is observed between atRA and rolipram. Podocytes were pretreated with rolipram (10 μM) for 30 min and then stimulated with atRA (10 μM) for another 3 d. The cell number was counted. Mean ± SD of three independent experiments is shown. *P < 0.01. (D) atRA induces synaptopodin expression in a protein kinase A (PKA)-dependent pathway. Under the same culture conditions as above, cells were used for Northern blot analysis for synaptopodin and GAPDH. A representative blot of three independent experiments is shown.

atRA improves kidney disease in HIV-1–transgenic mice. (A) Mean urine protein/urine creatinine (Upro/Ucr) ratio measured at 20-d intervals between 40 and 120 d of age. HIV-Tg26 mice had a significant increase in Upro/Ucr ratio at day 60 when compared with wild type (WT) mice. atRA treatment decreased Upro/Ucr ratio in HIV-Tg26 to WT levels. (B) The number of glomeruli with segmental sclerosis or collapsed tufts was increased in HIV-Tg26 mice at day 120 when compared with WT mice. atRA treatment significantly decreased the number of glomeruli with segmental sclerosis and completely abrogated tuft collapse. (C) Mean average of Ki67-positive nuclei was increased in HIV-Tg26 mice at day 120 when compared with WT mice. atRA treatment decreased the average number of Ki67-positive nuclei to WT levels.

(A and B) atRA stimulates intracellular cAMP production. After differentiation at 37°C, control podocytes (A) and HIV-infected podocytes (B) were cultured on collagen-coated six-well dishes and were stimulated with atRA (10 μM) for 2, 5, 10, 20, and 30 min with or without preincubation with rolipram (100 μM). Intracellular cAMP levels were determined using ELISA. The representative experiment of three independent assays is shown. (C and D) Dosage response of atRA and 9-cis-RA on cAMP production. Control podocytes were cultured on collagen-coated six-well dishes and were stimulated with various concentration of atRA or 9-cis-RA for 2, 5, and 10 min after preincubation with rolipram (100 μM). The representative experiment of three independent assays is shown.

(A) Forskolin (FK) and cAMP inhibit HIV-induced podocyte proliferation. After differentiation, control podocytes and HIV-infected podocytes were incubated in collagen-coated dishes with or without FK (30 μM) and cAMP (100 μM) for 3 d. Cell number was counted. Mean ± SD of five independent experiments is shown. *P < 0.001 versus HIV-infected cells. (B) FK and cAMP restore synaptopodin expression. The HIV-infected cells that were treated with FK and cAMP were subjected to Northern blot analysis for synaptopodin and GAPDH. A representative blot of three independent experiments is shown.

atRA inhibits HIV-induced mitogen-activated protein kinase (MAPK) 1 and 2 and Stat3 phosphorylation. (A) HIV-infected podocytes were incubated with atRA or 9-cis-RA at 0.1 to 10 μM overnight in serum-free medium. Control podocytes served as a control (CL). Cells were lysed for Western blot of phospho and total MAPK1,2. (B) Control and HIV-infected podocytes were incubated with FK (30 μM), cAMP (100 μM), or vehicle overnight in serum-free medium. Cells then were lysed for Western blot of phospho and total MAPK1,2. (C) Cells were preincubated with Rp-cAMP (100 μM) for 30 min and then stimulated with or without atRA (10 μM) overnight in serum-free medium, and Western blots were performed for phospho and total MAPK1,2. A representative blot from a minimum of three independent experiments is shown for all experiments. (D) HIV-infected podocytes were treated with atRA (1 μM) overnight in serum-free medium. Control podocytes served as a control (CL). Cells were lysed for Western blot of phospho and total Stat3 and Nef. A representative blot from a minimum of three independent experiments is shown for all experiments.