Biomechanical strain causes maladaptive gene regulation, contributing to Alport glomerular disease

Abstract

Patients with Alport's syndrome develop a number of pro-inflammatory cytokine and matrix metalloproteinase (MMP) abnormalities that contribute to progressive renal failure. Changes in the composition and structure of the glomerular basement membranes likely alter the biomechanics of cell adhesion and signaling in these patients. To test if enhanced strain on the capillary tuft due to these structural changes contributes to altered gene regulation, we subjected cultured podocytes to cyclic biomechanical strain. There was robust induction of interleukin (IL)-6, along with MMP-3, -9, -10, and -14, but not MMP-2 or -12 by increased strain. Neutralizing antibodies against IL-6 attenuated the strain-mediated induction of MMP-3 and -10. Alport mice treated with a general inhibitor of nitric oxide synthase (L-NAME) developed significant hypertension and increased IL-6 and MMP-3 and -10 in their glomeruli relative to those of normotensive Alport mice. These hypertensive Alport mice also had elevated proteinuria along with more advanced histological and ultrastructural glomerular basement membrane damage. We suggest that MMP and cytokine dysregulation may constitute a maladaptive response to biomechanical strain in the podocytes of Alport patients, thus contributing to glomerular disease initiation and progression.

Figure 1

Basal level of expression for MMP-3, MMP-10, and IL-6 are lower for cells cultured on placental laminin compared to collagen I. Glomerular podocytes were differentiated for two weeks and plated on plastic coated with placental laminin, collagen I and then laminin, or collagen I alone. After 48 hours RNA was isolated and analyzed for the indicated transcripts by real time PCR. Data points represent results from two independent experiments run in triplicate, with standard deviations. Based on these findings we carried out all mechanical stretch experiments on plates coated with placental laminin, since this is the predominant ligand for α3β1 integrin in the GBM. (45,46)

Figure 2

Mechanical strain induces MMP-3, 9, 10, and 14 mRNA expression in cultured podocytes. Glomerular podocytes were cultured on elastic membranes coated first with collagen I and then with placental laminin and subjected to 10% cyclic stretch using the FlexCell system. After 15 hours, mRNA was isolated and analyzed by RT-PCR for mRNAs encoding the indicated MMPs. Cyclophilin and GAPDH mRNAs were included as loading controls. Results shown are characteristic of several similar experiments.

Figure 3

Biomechanical strain induces IL-6, MMP-3, and MMP-10 in cultured podocytes, and IL-6 contributes to elevated MMP-3 and MMP-10 expression through an autocrine loop. Panel A Differentiated podocytes were plated onto elastic membranes pre-coated with collagen I and then placental laminin. Cultures were subjected to 10% cyclic mechanical strain for 15 hours and the RNA isolated and analyzed by real time PCR for the indicated transcripts. Panel B IL-6 neutralizing antibodies partially ameliorate the strain-mediated induction of MMP-3 and MMP-10. Glomerular podocytes were cultured on plates coated with collagen I and placental laminin in medium containing either neutralizing antibodies against IL-6, or an isotype matched control antibody. Cells were subjected to 10% biomechanical strain for 15 hours, and total RNA analyzed by real-time qRT-PCR using TaqMan probes specific for either MMP-3 or MMP-10. Panel C Recombinant IL-6 induces MMP-3 and MMP-10 mRNA expression levels in podocytes cultured on placental laminin. Differentiated podocytes were plated on plates pre-coated with placental laminin. Cells were harvested after 24 hours and analyzed for expression of mRNAs encoding MMP-3, MMP-10, and IL-6. A significant increase in MMP-3 and MMP-10 expression was observed at all three concentrations of recombinant IL-6 tested relative to untreated control cells. For all three experiments, amplifications were done in multiplex with TaqMan probes specific for GAPDH, which was used to normalize the data. Bars represent mean and standard deviations. GAPDH was run with all samples to control for loading. Data represent statistically analyzed results for at least 5 independent experiments run in triplicate. Asterisks denote significance (p<0.05).

Figure 4

Biomechanical strain leads to disruption of the cytoskeleton in wild type podocytes. A–D: Differentiated wild type podocytes were plated onto Flexcell plates with placental laminin. Cells were immunostained with antibodies specific for synaptopodin (A), α-actinin-4 (B) or Phalloidin (D). Dual channel immunostaining shows co-localization of the synaptopodin and α-actinin-4 (C). E–H: Shows immunostaining of podocytes on laminin following 15 hours of 10% cyclic biomechanical stretching on the Flexcell system. Intracellular filaments are disrupted, and anchoring filaments no longer stain positive for synaptopodin, suggesting mechanical strain results in disruption of both the actin cytoskeleton (Phalloidin staining in H) and destabilization of the synaptopodin/α-actinin-4 protein interactions (E–G), which are known actin bundling proteins involved in actin cytoskeletal dynamics. (14)

Figure 5

Treatment of mice with salt elevates blood pressure and increases proteinuria in C57Bl/6 X-linked Alport mice. Mice were given either water or water containing L-NAME salts at a dose of 50 µg/g body weight based on known rate of water consumption. Blood pressure measurements and urine collections were carried out once per week from 6 to 9 weeks of age. Panel I Salt treated mice showed consistently elevated blood pressure relative to non-salt treated mice (Panel IA). The urinary albumin bands were scanned and values normalized to urinary creatinine. There was a general trend for increased blood pressure with age in both groups. With a few exceptions, salt treated animals showed higher levels of proteinuria across all timepoints (Panel IB). An example of proteinuria measurements is provided in panel IC, showing a Coomassie blue stained gel for weekly collections of salt and non-salt treated Alport mice along with systolic blood pressure measurements. Panel II There is a direct correlation between systolic blood pressure and proteinuria in C57 Bl/6 X-linked Alport mice. Blood pressure measures were plotted relative to normalized albuminuria measurements (arbitrary units) for 8 Alport mice (5 salt treated and 3 no-salt). The data in panel A shows a direct correlation exists between albuminuria and systolic blood pressure for each timepoint. When all the data are combined, a highly significant correlation between blood pressure and albuminuria is observed

Figure 6

Hypertension induced by inclusion of L-NAME salt in the drinking water accelerates glomerular disease progression in X-linked Alport mice. Mice were treated with salt from 6 to 10 weeks of age. Glomeruli were immunostained for either fibronectin (FN) (Panels A–C) or the α2 chain of laminin (LNα2) (panels C-F). Laminin α2 deposits in the GBM are denoted by arrowheads. Glomeruli from 10 week old salt or non-salt treated (6 to 10 weeks) animals were examined by TEM (panels G–I). Salt treated animals with high systolic blood pressure consistently showed more extensive GBM rarification and podocyte foot process effacement compared to non-salt treated animals with low systolic blood pressure. No differences were observed in salt treated wild type mice compared to wild type mice given normal drinking water (data not shown).

Figure 7

Treatment with L-NAME salt increases proteinuria and expression of mRNAs encoding MMP-3, -MMP-10, and IL-6 in 129 Sv autosomal Alport mice. Panel I Wild type and Alport mice were treated (or not) with L-NAME salt for one week between 4 and 5 weeks of age. Urine was collected, normalized to urinary creatinine by dilution, fractionated on polyacrylamide gels, and stained with Coomassie blue. Note that in some instances, microalbunimuria is observed in salt-treated wild type mice. Alport mice showed an increase in proteinuria with salt treatment (N=3, a typical result is shown). Systolic blood pressure measures are shown in parentheses. Panel II Hypertension results in elevated expression of IL-6, MMP-3, and MMP-10 mRNAs in glomeruli from 129 Sv Alport mice. Alport mice or wild type littermates were given either water or L-NAME salt in water from 4 to 5 weeks of age (two animals per group). At five weeks of age, glomeruli were isolated and RNA analyzed in triplicate by real time RT-PCR for MMP-3, MMP-10, MMP-12, and IL-6 mRNA normalized to GAPDH mRNA which was run in multiplex. mRNAs for MMP-3, MMP-10, and IL-6 were significantly elevated in salt hypertensive Alport (denoted by asterisks) mice relative to “normotensive” Alport mice. MMP-12 was not significantly elevated in salt versus non-salt treated Alport mice.