Introduction

Procalcitonin (ProCT) is the 116 aminoacid precursor of the hormone calcitonin. ProCT serum concentration is an important biomarker of the severity and prognosis of systemic infections [1]. Both experimental sepsis and exposure to lipopolysaccharide (LPS) resulted in a multifold elevation of ProCT arising from a diverse cellular origin, including the kidney [2–4]. In our previous studies with a porcine sepsis model, both early [5] and late [6] immuno-neutralization of ProCT significantly ameliorated systemic and renal complications, supporting a role for this prohormone as a biomediator of the systemic inflammatory response [1, 7–9].

The glomerular mesangial cell (MC) is a pericyte with a measurable role in renal function both in health and disease [10–14]. In vitro studies indicated that the creation of a proinflammatory milieu that is similar to that unleashed by the injection of bacterial endotoxin altered the biological properties of MCs [15–19] resulting in abnormalities of glomerular function and structure. Whether ProCT directly affects MCs has not been studied.

In the present study, we evaluated the functional, biological and ultrastructural effects of ProCT on MCs. We hypothesized that, similar to vasoactive peptides and mediators of inflammation, ProCT exerts a direct effect on MCs through activation of proinflammatory genes, alterations of the actin cytoskeleton and induction of apoptosis.

Materials and methods

Cell culture

Murine MCs (primary cultures) were grown in a 3:1 mixture of DMEM and F12 medium containing 6 mM glucose, 1 mM glutamine, 0.075 % NaHCO3, 100 U/mL penicillin/100 μg/mL streptomycin and 10–20 % fetal bovine serum (FBS). Upon reaching approximately 80 % confluence, cells were dispersed with trypsin/EDTA, and plated in cell culture dishes pre-coated with human fibronectin, at an adequate cell density for each experiment.

RNA extraction and real time RT-PCR

MCs were plated at a density of 20,000 cells/well into a six-well plate and grown with DMEM and 20 % FBS for 2 days. The medium was replaced by DMEM with 2 % FBS for 24 h and the cells were subsequently incubated with bioactive ProCT of human origin [9, 20]. Dose–response curves and time-course experiments were performed to determine the optimal ProCT dose and time of study. The cells were homogenized and total RNA was extracted using the RNeasy mini kit (Qiagen, Valencia, CA) following the manufacturer’s protocol. RNA samples (100 ng) were employed in duplicate to determine a relative quantification of target transcript using real time RT-PCR with the TaqMan® One-Step RT-PCR Master Mix Reagents (Applied Biosystems, Foster City, CA) and the ABI 7500 Sequence Detector System (Applied Biosystems). Normalization was calculated by the difference between target and GAPDH C T levels (∆C T). Primers and TaqMan® probes specific for TNF-α, IL-6 and iNOS genes were designed using Primer Express software (Applied Biosystems). Specificity of transcript amplification was verified using a nucleotide Basic Local Alignment Search Tool (BLAST, National Center for Biotechnology Information, NIH).

PCR array

MCs were grown in two T-25 flasks as previously described and incubated with or without ProCT (5 ng/mL) for 4 h. The cells were harvested and RNA was extracted and transcribed to cDNA with RT2 first strand kit (SABiosciences/Qiagen, Maryland). A signal transduction pathway finder and cytokines/chemokines (SABiosciences/Qiagen, Maryland) PCR arrays were performed to build a panel of genes related to several transduction and inflammatory pathways affected by ProCT.

Immunofluorescence analysis (F-actin and α-smooth muscle actin, or α-SMA)

To assess ProCT-induced morphological changes, a qualitative study was performed by exposing MCs (2,000 cells/well) to ProCT (5 ng/mL) for 4 h and fixed in 4 % paraformaldehyde. Fixed cells were blocked with 1 % BSA, and incubated with Alexa Fluor 488 phalloidin (Invitrogen, Carlsbad, CA) or with mouse monoclonal anti-α-SMA antibody (Sigma-Aldrich, St Louis, MO) followed by goat anti-mouse Alexa 488 IgG (Invitrogen, Carlsbas, CA). Slides were mounted using ProLong Gold with DAPI (Invitrogen, Carlsbad, CA) and images were captured using a Nikon NIS-80i fluorescent microscope coupled to a digital camera.

Apoptosis assay

MCs were plated (2,000 cells/well) into a dark 96-well plate and grown with DMEM (without phenol red) as previously described. Apoptosis and cell viability were measured in the same sample with ApoLive-Glo Multiplex Assay (Promega, Madison, WI) 4 h after incubation with ProCT (2.5 and 5 ng/mL). In this assay, a substrate was used to measure the activity of caspases three and seven, which play a central role in the execution-phase of cell apoptosis. Cell viability was measured by a fluorogenic substrate to a live-cell protease, which is restricted to intact viable cells. Staurosporin (10 μM) was used as a positive control of apoptosis.

Statistical analysis

Each quantitative experiment was performed independently at least twice, with 2–5 replicates. For RT-PCR time–course curve, a non-paired t test was used to compare ProCT with respective time-controls. For all other comparisons, one-way ANOVA was used, followed by Student–Newman–Keuls test. Results are mean ± SEM. Statistical significance was defined as p < 0.05.

Results

Figure 1 shows mRNA transcript levels of IL-6, iNOS and TNF-α after incubation of MCs with increasing concentrations of ProCT (panels b, d, f) and a time-course study (panels a, c, and e). IL-6 mRNA was significantly and progressively increased after 2, 4 and 6 h of ProCT exposure (Fig. 1a). ProCT induced an increase of 40-fold in IL-6 mRNA at 2.5 ng/mL and approximately 350-fold at 5 ng/mL (Fig. 1b). ProCT also induced a 5- and 10-fold increase in iNOS mRNA at 2.5 ng/mL (Fig. 1d) and after 4 and 6 h (Fig. 1c) while TNF-α increased only after a 6 h incubation with 5 ng/mL of ProCT (Fig. 1e, f). Thus, MCs exposed to ProCT produced soluble mediators of inflammation, a phenomenon consistent with a direct cytotoxic effect.

Fig. 1
figure 1

mRNA transcript levels of IL-6, iNOS and TNF-α in cultured mesangial cells (MC) exposed to ProCT (5 ng/ml) for 2, 4 and 6 h (a, c, e) or to several ProCT concentrations (b, d, f) for 4 h. Each graph represents one of two experiments performed in quintuplicate. *p < 0.01 and **p < 0.001 vs control

Full size image

Tables 1 and 2 show, respectively, a summary of the genes up- or downregulated by ProCT (5 ng/mL, for 4 h) in 18 different signal transduction pathways and in eight cytokine/chemokine pathways (Supplemental Tables S1 and S2). Among a total of 84 genes analyzed in each PCR array, 26 genes involved in several signal transduction pathways were significantly upregulated (Table 1, Supplemental Table S1). The Wnt, p53, stress and Ca2+/PKC pathways were more significantly affected by ProCT, with most of the genes increasing 5- to 24-fold compared to control (p < 0.05). Several elements involved in cell defense, such as PPAR, BIRC5, En1, and Ptch1 were downregulated by ProCT. Genes associated with cytokines, chemokines, and co-stimulatory molecules were mostly up-regulated by ProCT (Table 2). Some anti-inflammatory cytokines such as IL-10, IL-11, IL-12b and IL-24 were also upregulated, suggesting an attempt by the MC to counteract the cytotoxic effect of ProCT.

Table 1 Pathway analysis of gene expression changes in MC exposed to ProCT (5 ng/ml) for 4 h
Full size table
Table 2 Chemokines/Cytokines expression changes in MC exposed to ProCT (5 ng/ml) for 4 h
Full size table

MCs exposed to ProCT for 4 and 6 h lost their confluence and showed signs of cytoplasmic and nuclear shrinkage (Fig. 2). MCs at rest displayed abundant parallel actin microfilaments running throughout the cytoplasm (Fig. 3a, c). ProCT induced cytoskeletal alterations and contraction of MCs (Fig. 3d) as compared to control (Fig. 3c). Following 4 h of exposure to ProCT, MCs displayed changes in cell shape, and no longer exhibited a discernible array of F-actin filaments but instead showed a retracted cytoplasm with less intense staining that was consistent with the loss of the typical actin distribution (Fig. 3b). More dramatic changes were visualized in α-SMA stained cells characterized by coalescing, nuclear fusion, and a loss of the typical cytoplasmic extensions (Fig. 3d).

Fig. 2
figure 2

Mesangial cells in 2 % FBS DMEM (a) maintained their spindle shape and confluence throughout the experiment. After 4 h incubation with ProCT (5 ng/mL) the cells lost confluence and displayed changes in shape (b). After 6 h incubation with ProCT, cells showed signs of cytoplasmic and nuclear shrinkage (c). Magnification ×10

Full size image
Fig. 3
figure 3

Immunofluorescence staining of MC grown in chambers on glass slides and exposed or not to ProCT (5 ng/ml) for 4 h. The F-actin filaments of the cytoskeleton were stained with Alexa 488 Phalloidin (a, b) and the alpha smooth muscle actin with mouse monoclonal anti-α-SMA antibody (c, d). Both staining are shown in green; nuclei are stained in blue with DAPI

Full size image

As demonstrated in Fig. 4, ProCT (5 ng/mL) decreased MC viability by 36 % compared with control cells (black bars, p < 0.001), mainly due to the induction of apoptosis (114 % increase vs control, hatched bars, p < 0.001). At 2.5 ng/mL, ProCT decreased cell viability by 16 % and did not induce apoptosis. Staurosporin (10 μM) was used as a positive control for apoptosis. The combined measurement of apoptosis and cell viability used in this study provides an accurate distinction of apoptosis from other mechanisms of cell death. The ratio of caspase activity to viable cells is useful for determining the extent of caspase activation and for normalizing cell number.

Fig. 4
figure 4

Apoptosis (dashed bars) and cell viability (black bars) of cultured MCs exposed or not to ProCT (2.5 and 5 ng/ml) for 4 h. Staurosporin was used as a positive control for apoptosis. *p < 0.001 vs control (ANOVA, Newman Keuls post test)

Full size image

Discussion

The principal finding of this study was a direct toxic effect of ProCT on MCs. Our previous studies had shown that blocking ProCT during the course of experimental sepsis in pigs mitigated the abnormal systemic hemodynamic parameters and renal function [5, 6]. The results obtained in the present study contribute to an elucidation of the mechanisms whereby neutralization of ProCT may be beneficial. MCs, recognized as playing an active role in both inflammatory and non-inflammatory renal injury [12], responded to ProCT by increasing synthesis of IL-6, iNOS, and TNF-α, enhanced expression of inflammatory genes and pathways, disruption of the actin microfilament network, and apoptosis. The murine MCs used in this study have been previously well characterized, and develop physiological and pathological characteristics that permit the study of human kidney diseases [21–24]. In this in vitro model, there is no appreciable conversion of ProCT into calcitonin. Previous studies using purified human ProCT in hamsters demonstrated a successful heterologous response, thus suggesting that the effects observed in murine MCs exposed to ProCT resulted from a specific hormone-receptor interaction. This hypothesis is supported by the fact that the homology between the human and mouse ProCT amino acid sequence is fairly high (77 %) [25]. Future studies to elucidate the interaction of calcitonin peptides with receptors in MCs and in other renal cells will provide an attractive interventional perspective for septic AKI.

Levels of iNOS were very low in normal MCs but increased dramatically after exposure to cytokine or LPS [26, 27]. An increased and prolonged production of NO by iNOS was associated with cytotoxicity via formation of iron-nitrosyl complexes and inactivation of iron-containing enzymes and through reaction with superoxide to generate toxic hydroxyl radicals [28]. Our studies show that iNOS is upregulated by ProCT in MCs, supporting the hypothesis that several renal alterations observed during sepsis and attributable to LPS, could be a direct consequence of ProCT effects.

Although TNF-α [29, 30] and IL-1β [29, 31] are usually the proinflammatory mediators of LPS in sepsis, in our study TNF-α was upregulated only after 6 h of ProCT incubation, a time-point when most of the MCs had reached advanced stages of apoptosis. These findings suggest that the cellular mechanisms triggered by ProCT may also involve distinct inflammatory pathways independently of LPS.

PCR arrays analysis revealed that several genes of the Wnt pathway were upregulated in MCs exposed to ProCT. Aberrant regulation of Wnt/β-catenin signaling pathway has been implicated in many types of chronic kidney diseases as well as in nephrotoxic acute kidney injury (AKI) [32–37]. The non-canonical Wnt signaling pathway has two intracellular signaling cascades that consist of the Wnt/Ca2+ pathway and the WnT/PCP pathway. The Wnt/Ca2+ pathway triggers cellular processes that involve activation of PLC, increase of (Ca2+)i, decreased cGMP (cyclic guanosine monophosphate) levels, and activation of CamKII (Ca2+—calmodulin-dependent protein kinase-II) or Caln (calcineurin) and PKC (protein kinase C) [32, 33]. These signals can then stimulate transcription factors like CREB (cAMP response element-binding protein-1). The Wnt/PCP pathway activates Rho/Rac small GTPase and JNK (Jun N-terminal Kinase) to assist in regulation of cytoskeletal organization and gene expression. Our findings that genes of PLC, Ca2+/PKC, CREBS and LDL pathways were also upregulated in MCs suggest that activation of the Wnt pathway may underlie ProCT toxicity. The upregulation of a vast number of genes associated with proinflammatory chemokines suggests that ProCT participates in the inflammatory response not only by stimulating local production of cytokines but also by triggering a chemotactic response that further augments its toxic effects.

MCs offer structural support to the glomerular capillary through sub-membranous microfilament bundles that are continuous with microtendinous projections tethered to the glomerular basement membrane. These microfilaments are ideally distributed to withstand tensional stress originating in the glomerular microcirculation [38]. The substantial cytoskeleton changes observed after incubation with ProCT suggest that MCs may be incapable of rendering tonic resistance to stretch, a phenomenon associated with the expression of certain inflammatory mediators such as ICAM-1 [39] and this may also contribute to the early decline in the glomerular filtration rate of experimental sepsis that occurs despite the systemic hypotension [40, 41].

IL-6 has recently emerged as an important marker and/or mediator of AKI. In humans with sepsis, IL-6 levels strongly correlated with the incidence of AKI [42]. In a model of porcine sepsis, renal expression of ProCT positively correlated with IL-6 [4]. Therefore, the powerful stimulation of IL-6 production by MCs induced by ProCT is of potential relevance to septic AKI.

In summary, this study demonstrates direct cytotoxic effects of ProCT on MCs that are independent of bacterial endotoxemia and hemodynamic perturbations. These findings reflect the motive force of ProCT as a toxic mediator in sepsis-related AKI. The fact that mouse mesangial cells have been validated as an in vitro model for human kidney disease [21–24] and that our previous work in pigs [5, 6] demonstrated beneficial effects of immuno-neutralization of ProCT on the course of experimental sepsis makes our findings potentially relevant to humans.