Triptolide is a traditional Chinese medicine-derived inhibitor of polycystic kidney disease

Abstract

During kidney organogenesis, tubular epithelial cells proliferate until a functional tubule is formed as sensed by cilia bending in response to fluid flow. This flow-induced ciliary mechanosensation opens the calcium (Ca(2+)) channel polycystin-2 (PC2), resulting in a calcium flux-mediated cell cycle arrest. Loss or mutation of either PC2 or its regulatory protein polycystin-1 (PC1) results in autosomal dominant polycystic kidney disease (ADPKD), characterized by cyst formation and growth and often leading to renal failure and death. Here we show that triptolide, the active diterpene in the traditional Chinese medicine Lei Gong Teng, induces Ca(2+) release by a PC2-dependent mechanism. Furthermore, in a murine model of ADPKD, triptolide arrests cellular proliferation and attenuates overall cyst formation by restoring Ca(2+) signaling in these cells. We anticipate that small molecule induction of PC2-dependent calcium release is likely to be a valid therapeutic strategy for ADPKD.

Fig. 1.

Identification of PC2 as a protein associated with [³H]triptolide-binding activity. (A) Triptolide is a biologically active diterpene from the medicinal vine Tripterygium wilfordii Hook F. (B) [³H]triptolide-associated proteins were fractionated by FPLC over an increasing NaCl gradient. (Left) Fractions were tested for the presence of [³H]triptolide (cpm) and analyzed by SDS/PAGE. (Right) The Coomassie stained gel is shown; a main protein band at 110-kDa in fractions 43–51 (peak binding activity) is represented by the arrow. MW markers are on the left. (C) MALDI-MS analysis of 110-kDa band in B identified as PC2. (D) Western blot analysis with anti-PC2 antisera. Lane 1, positive control lysate; lane 2, purified sample before FPLC fractionation; lane 3, pooled fractions of [³H]triptolide positive samples; lane 4, pooled fractions of [³H]triptolide-negative samples.

Fig. 2.

PC2 expression is necessary for rapid cell death or growth inhibition induced by triptolide. (A) Murine kidney epithelial cells were plated at a starting density of 3 × 10⁵ (0 h) and then treated with 100 nM triptolide for 24 h or 25 nM for 48 h. Cell viability was determined by trypan blue dye exclusion, and results are represented graphically (n = 6, mean ± SE). (B) PC2 expression was assessed by indirect immunofluorescence (IF) detection and by Western blot analysis in each cell line (Pkd2^+/−, Pkd2^−/−, and Pkd2^REX). A normal growth curve is shown for a 72-h time course (n = 3, mean ± SE). (C) Caspase-3 activity assays from Pkd2^−/− and Pkd2^REX cells treated with 100 nM or 25 nM triptolide; 0 h represents untreated baseline levels of caspase-3 activity (n = 3, mean ± SE). Caspase-3 activity was also assessed in Pkd2^−/− and Pkd2^REX intact cells. Nuclear localized green fluorescence is indicative of active caspase-3. (D) Pkd1^−/− cells were treated with 100 nM triptolide or DMSO control over a 96-h time course. Representative fields were photographed under bright field microscopy (10×). Cell viability was assessed, and results are graphically represented (n = 3, mean ± SE). PC2 localization in untreated Pkd1^−/− cells was visualized by using confocal microscopy (Fluorescein isothiocyanate, 40×) and propidium iodide nuclear stain (red). (E) Pkd1^−/− cells (starting density of 20 × 10⁵) were cultured in the absence or presence of calcium-containing media ± 100 nM triptolide for 72 h. Brightfield images show cell density after 72 h of culture, and cell viability following triptolide treatment is graphically represented (n = 3, mean ± SE). (F) Western blot analysis of p21^CIP/WAF (Upper) and p53 (Lower) expression in Pkd1^−/− cells during a time course of 100-nM triptolide treatment.

Fig. 3.

Triptolide causes a PC2-dependent calcium release in murine kidney epithelial cells. Changes in intracellular calcium levels from Pkd1^+/− (A Left), Pkd1^−/− (B Left), Pkd2^−/− (C Left), and Pkd2^REX (D Left) cells during a typical experiment with 100 nM triptolide (solid bar). The average triptolide response for each genotype is represented on the right side of each panel. (E) Average maximum change in intracellular calcium levels elicited by triptolide in Pkd1^+/−, Pkd1^−/−, Pkd2^−/−, and Pkd2^REX cells in HBSS, Ca²⁺-free HBSS, or in the presence of 200 μM 2APB. Error bars are ± SEM. ∗, Pkd1^+/−, Pkd1^−/−, Pkd2^−/−, and Pkd2^REX cells were significantly different from each other, P < 0.01; one-way ANOVA, F = 123.5, df = 3, Tukey multiple comparison test, P < 0.5. #, Pkd1^−/− in HBSS significantly different from Ca²⁺ free and 2APB, P < 0.01; one-way ANOVA, F = 123.5, df = 2, Tukey multiple comparison test; one-way ANOVA, F = 160.2, df = 3, P < 0.01. ##, Pkd2^REX in HBSS significantly different from Ca²⁺ free and 2APB, P < 0.01; one-way ANOVA, F = 65.9, df = 2, Tukey multiple comparison test.

Fig. 4.

Triptolide reduces cystic burden in a Pkd1^−/− murine model of ADPKD. (A–C) Representative kidneys from Pkd1^−/− pups treated with DMSO during gestation (E10.5–birth). Large cysts are present throughout the medulla and cortex (4× magnification). (D–F) Representative kidneys from Pkd1^−/− pups treated with triptolide during gestation. (G) Pkd1^+/+ kidney from a pup treated with DMSO or (H) triptolide. (I) Pkd1^+/− kidney treated with DMSO or (J) triptolide. (K–M) IHC staining of Pkd1^−/− kidneys for active caspase-3 expression: (K) secondary antibody negative control, (L) DMSO treated, and (M) triptolide treated. (N) Percent of cyst burden in the kidney as determined by area in each of the Pkd1 genotypes (mean ± SE, Pkd1^−/−, n = 18; Pkd1^+/−, n = 10; and Pkd1^+/+, n = 10). (O) Cystic burden values comparing DMSO or triptolide treatment for each Pkd1^−/− animal are represented as open circles. The mean value (as shown in N) is represented as a horizontal line.