Introduction

Diabetic nephropathy is a leading cause of end-stage kidney disease throughout much of the world [1]. Hyperglycaemia, a central prerequisite of diabetic nephropathy, correlates with both the development [2] and progression of the disease [3]. Over the past two decades a number of glucose-dependent pathways have been implicated in the pathogenesis of diabetic kidney disease [4]; these include activation of protein kinase C (PKC) [5], with both experimental and clinical studies highlighting the importance of the beta isoform [6, 7]. The alpha isoform has also been implicated in the pathogenesis of diabetic nephropathy, potentially mediating changes that lead to albuminuria in the diabetic glomerulus [8, 9].

Binding of diacylglycerol (DAG) to the inactive forms of both the classical and novel PKC isoforms plays an important role in inducing activation of PKC [10]. Indeed traditionally, the increased activity of PKC in the diabetic state has been attributed to enhanced production of DAG as a consequence of increased flux through the glycolytic and pentose phosphate pathways [11]. However, in addition to allosteric modification, a number of in vitro studies have shown that high glucose levels not only change PKC activity, but also increase PKC protein [12, 13]. Increases in PKC-β protein levels have also been demonstrated in the in vivo context in rodent models of diabetic nephropathy [14], with more recent studies also showing increased PRKC-β (also known as PRKCB) expression [15]. In this study, we sought to determine the role of PRKC-β transcription in the subsequent amplified activation that has been demonstrated in several models of diabetic nephropathy.

While the glomerulus, and in particular the mesangium, has been the focus of intense investigation in diabetes, tubulointerstitial injury is also a major feature of diabetic nephropathy and an important predictor of renal dysfunction [16]. Accordingly, we sought to determine the effects of high glucose on PRKC-α (also known as PRKCA) and PRKC-β expression in tubular epithelial cells and to examine the magnitude of expression and site-specific localisation of the two isoforms in human diabetic nephropathy in relation to glycaemic control.

Methods

In vitro study

The methods for primary culture of human proximal tubular cells (PTC) are described elsewhere [17]. In brief, segments of macroscopically normal renal cortex were obtained under aseptic conditions from patients undergoing nephrectomy for small (<6 cm) tumours. Patients were accepted for inclusion into the study if there was no history of renal or systemic disease known to be associated with tubulointerstitial pathology. In addition, the presence of a histologically normal cortical tubulointerstitium at sites removed from the tumour was confirmed by independent pathological examination. Written informed consent was obtained from each patient prior to surgery and ethical approval for the study was obtained from the Royal North Shore Hospital Human Research Ethics Committee. The current study assayed the PRKC-α and PRKC-β transcription response to a high-glucose environment in PTCs from three separate patients.

PTCs were seeded at 15,000 per well in 96-well culture dishes in hormonally defined media. When 50% confluence had been reached, cells were quiesced for 24 h in basal medium consisting of DMEM and Hams F-12 (1:1, vol./vol.; ThermoTrace, Noble Park, VIC, Australia), supplemented with 5 μg/ml human transferrin (Sigma Chemicals, St Louis, MO, USA), penicillin (100 U/ml) and streptomycin (100 μg/ml; Gibco BRL, Gaithersburg, MD, USA). The cells were then cultured for 48 h in basal medium containing 5 mmol/l glucose and 30 mmol/l glucose, each done in triplicate.

To quantify PRKC-α and -β transcription, RNA was extracted and PRKC-α and -β mRNA measured with PCR as described below.

Biopsy study

Renal tissue was obtained from biopsy samples stored from a study in which histology and clinical data have been previously reported [18]. In brief, 49 patients with type 2 diabetes and evidence of proteinuria but no retinopathy underwent percutaneous needle biopsy of the kidney to assess renal histology. Of the entire cohort, 69% (34) of the patients had evidence of diabetic nephropathy in their biopsy, the remainder had either normal glomerular (nine) structure or glomerulonephritis (six). Of the diabetic nephropathy cohort, 25 archival biopsies contained sufficient biopsy material for molecular analysis. Relevant clinical data were also analysed.

RNA extraction and cDNA synthesis

RNA was extracted from paraffin-embedded blocks of study tissue by cutting a 6 μm section and placing it into 200 μl RNA lysis buffer. Next, 200 μl water-saturated phenol and chloroform was added to the RNA sample, which was mixed well and centrifuged for 2 min at 4°C (16,000 g). Following removal of the supernatant fractions, the phenol-chloroform extraction was repeated and followed by a further chloroform extraction. Samples were then precipitated with 180 μl isopropanol, 18 μl 3 mol/l sodium acetate (pH 5.2) and 1 μl carrier glycogen (20 mg/ml) at 20°C. After centrifuging for 15 min at 4°C (16,000 g), the supernatant fraction was discarded and the RNA pellet was washed with 1 ml cold 70% diethyl pyrocarbonate-treated ethanol, centrifuged for 10 min at 4°C (16,000 g) and resuspended in 10 μl RNase-free diethyl pyrocarbonate water.

Of RNA extracted from tissue sections, 1 μl was reverse-transcribed in a final volume of 20 μl using M-MLV reverse transcriptase (Gibco-BRL, Carlsbad, CA, USA) in the manufacturer’s buffer containing 1 mmol/l dNTPs, 40 U RNase inhibitor (Amersham Pharmacia Biotech, Freiburg, Germany), 300 ng random hexamers (Pharmacia) and 7 μl RNA. The reactions took place at 42°C for 60 min, followed by 95°C for 5 min and 4°C for 5 min.

Real-time PCR

PRKC-α and PRKC-β expression was measured and quantified using the a sequence detector (GeneAmp 5700; PE Biosystems, Foster City, CA, USA) according to the manufacturer’s instructions. Primers were obtained from Sigma-Aldrich. Sequences of primer and probe are as follows: (1) PRKC-α; commercial PCR probe and primers (Gene Expression Assay; Applied Biosystems, Foster City, CA, USA), assay ID Hs00176973_m1; (2) PRKC-β forward primer CTTCAAGCAGCCCACCTTCT, reverse primer TCCCCGAAGCCCCAGATG, probe AGCCACTGCACCGAC. The fluorogenic probe (Applied Biosystems) includes a fluorescence reporter (6-carboxyfluorescein) at the 5′-end and a fluorescent quencher (6-carboxytetramethylrhodamine [TAMRA]) at the 3′-end. A commercial, pre-developed 18S control kit labelled with VIC, a fluorescent reporter dye on the 5′-end and the quencher (TAMRA) on the 3′-end (PE Biosystems) was used as the endogenous control. Similar amplification efficiencies of PRKC-α, PRKC-β and 18S were validated, permitting semi-quantitative measurement of gene expression.

The 25 μl PCR mixture contained 12.5 μl Taqman Universal PCR Master Mix (Applied Biosystems), 500 nmol/l primers (forward and reverse), 100 nmol/l Taqman probe and 1 μl of cDNA template. PCR was performed at 50°C for 2 min, 95°C for 10 min and then run for 50 cycles at 95°C for 15 s and 60°C for 1 min. Comparative cycle threshold (C t) calculations were all relative to the control group. 18S C t values were subtracted from PRKC C t values to give a ΔC t value. ΔΔC t values were achieved by subtracting the average control ΔC t value, for each animal group. Expression of the genes encoding PKC isoforms relative to control was derived using the equation \(2^{{ - \Delta \Delta \operatorname{C} _{t} }} \).

PKC-β localisation and activation

PKC-β1 and -β2 peptides and their phosphorylated isoforms (phospho-PKC-β1 and -β2) were localised by immunostaining using polyclonal rabbit anti-PKC-β1 and -β2 antibodies (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and polyclonal rabbit anti-phospho-PKC-β1 and -β2 antibodies (Biosource International, Camarillo, CA, USA).

Tissue sections (4 μm thick) were prepared from paraffin-embedded tissue of study biopsies and immunostained as previously described [19] according to the manufacturer’s instructions. Sections incubated with 1:10 normal goat serum, instead of the primary antiserum served as the negative control. Sequential sections were obtained to determine co-localisation of PKC-β1, PKC-β2 and phospho-PKC-β1 and-β2.

Western blot analysis

PKC-β activation was measured by western blot analysis of phosphorylated PKC-β peptide normalised to actin (Labvision, Fremont, CA, USA) using a method already described [12].

Results

In vitro study

After 48 h, exposure of human proximal tubular epithelial cells exposed to 30 mmol/l glucose led to a 65% increase in PRKC-β mRNA compared with cells grown in a 5 mmol/l glucose environment (Fig. 1a). Western blot measuring the amount of phosphorylated PKC-β demonstrated a discernible increase in phospho-PKC-β in PTCs exposed to a high-glucose environment for 48 h compared with those exposed to low glucose (Fig. 1b).

Fig. 1
figure 1

a Effects of high glucose environment on PRKC-α (hatched bars) and PRKC-β (black bars) expression of cultured human PTC, demonstrating a 50% increase in PRKC-β in high glucose medium, but no discernible change with PRKC-α. Values are mean ± SEM; *p < 0.05. b Western blot analysis and (c) quantification of phosphorylated PKC-β in protein lysates from cultured human proximal tubules exposed to high (30 mmol/l) and low (5 mmol/l) glucose

Full size image

Human biopsy study

Clinical data

Of the 34 biopsies from the previously reported study, 25 contained sufficient tissue to extract suitable quantities of RNA. This subset of patients displayed the characteristic features of diabetic nephropathy with proteinuria, hypertension in all but one, declining glomerular filtration rate, glomerulosclerosis and tubulointerstitial fibrosis (Table 1).

Table 1 Clinical data collected at the time of biopsy
Full size table

PRKC-β and PRKC-α expression

Analysis of PRKC-β expression in the diabetic biopsies demonstrated a near tenfold increase in specific mRNA when compared with control tissue (Fig. 2). A close correlation (r = 0.63) was noted (Fig. 3) between PRKC-β expression and HbA1c at the time of biopsy, but not with any of the other clinical or structural indices documented in the study. As only one patient was classified as not hypertensive, the role of hypertension in the gene transcription of PRKC-β could not be determined. There was no detectable increase in PRKC-α gene expression in the biopsy samples.

Fig. 2
figure 2

RT-PCR measurement of PRKC-α (hatched bars) and PRKC-β (black bars) expression in archival human kidney tissue, demonstrating a tenfold increase in PRKC-β in diabetic nephropathy samples and no significant change in PRKC-α mRNA. Mean ± SEM of six (control) and 33 (diabetic nephropathy) samples; *p < 0.001

Full size image
Fig. 3
figure 3

Graphical representation of the relationship between PRKC-β renal gene expression and HbA1c at the time of biopsy, demonstrating a close correlation (r = 0.63, p < 0.05)

Full size image

PKC-β localisation

Immunolocalisation of PKC-β1 and PKC-β2 peptide revealed a predominant tubulo-epithelial pattern of increased abundance in the diabetic biopsies, with only sparse glomerular staining (Fig. 4). In serial sections, co-localisation of PKC-β1 and PKC-β2 was noted. Furthermore, immunostaining for phosphorylated PKC-β1 and PKC-β2 as a measure of PKC-β tissue activation confirmed co-localisation with PKC-β1 and -β2 (Fig. 4).

Fig. 4
figure 4

Representative photomicrograph demonstrating co-localisation of PKC-β1, PKC-β2, and phospho-PKC-β1 and -β2 in matched serial sections of human diabetic nephropathy (DN). Staining of both peptide and phosphorylated moiety (brown) was increased compared with control and was seen predominantly in tubulo-epithelial cells Magnification: ×160 (left), ×320 (centre), ×160 (right)

Full size image

Discussion

Intensive investigation over several decades has implicated excessive activity of four glucose-dependent pathways in the pathogenesis of the long-term complications of diabetes: the polyol pathway, advanced glycation end-product formation, hexosamine flux and activation of PKC [4]. The pathogenetic role of PKCs and the β-isoform, in particular, in diabetic complications has traditionally been attributed to high intracellular glucose inducing de novo synthesis of DAG, leading to enzyme activation by allosteric modification [4, 5]. However, observations by Kapor-Drezgic et al. [12] several years ago showed that in addition to activating the enzyme, high glucose also induced an increase in PKC-β protein (and also α and ɛ) after exposing mesangial cells to high glucose for 48 h, but not after 12 h of exposure. Consistent with this, Kelly et al. [14] reported increased PKC-β protein in the kidneys of diabetic rats and that this increase was attenuated by inhibiting enzyme activity with the PKC-β inhibitor, ruboxistaurin. While not readily explainable at the time, more recent studies have shown that PKC-β regulates its own expression in vivo, requiring the presence of the active form of the enzyme for modulation of its transcription [20]. In the present study, using archival biopsy tissue from a previously reported study, we documented an increase in PRKC-β mRNA, but not PRKC-α mRNA, in kidney biopsies from diabetic patients, showing a close correlation with glycaemic exposure. Moreover, when renal tubular epithelial cells, the major source of PKC-β protein in our tissues with advanced diabetic nephropathy, were exposed to 30 μmmol/l glucose for 48 h, PRKC-β mRNA increased, with evidence of phosphorylation following analysis of cell culture peptide fraction. The present study demonstrates that in addition to the already identified role of high glucose in the allosteric modification and activation of PKC isoforms, modulation of PRKC-β expression by high glucose may further increase the amount of peptide production that is then available for activation, thus compounding the contribution of high glucose to the pathogenesis of diabetic nephropathy and possibly of other long-term complications of diabetes.

In the present study, performed using biopsies from patients with advanced diabetic nephropathy, PKC-β was localised predominantly to tubular epithelial cells. These findings are consistent with the recent findings of Toyoda et al. [15], who reported an increase in glomerular PRKC-β mRNA in early diabetic nephropathy, but not in patients with advanced disease. In our study of patients with relatively advanced disease, we also did not find increased glomerular PKC-β expression by immunohistochemistry. However, we did note abundant PKC-β protein in tubular epithelial cells, consistent with the tenfold increase in PRKC-β mRNA noted by quantitative PCR analysis. Together these findings suggest that while production of PKC-β is predominantly glomerular in early diabetic nephropathy, advanced disease is characterised by predominantly tubular production.

Several observational and interventional studies have documented the close relationship between average glycaemic control over several years and both the development and progression of diabetic nephropathy [2, 3]. In the present study we found a significant correlation between PRKC-β mRNA and HbA1c at the time of biopsy. However, no relationship between PRKC-β mRNA and a range of other clinical or histopathological indices of injury was noted. While these findings raise the possibility that PKC-β levels may be a marker of glucose exposure rather than a pathogenetic factor in diabetic nephropathy, this seems unlikely given the ability of the PKC-β inhibitor, ruboxistaurin, to attenuate the disordered structure and function, not only in experimental diabetic nephropathy [7], but also in the human context. We speculate that the lack of correlation between PRKC-β mRNA and renal structure or function may be related to the chronology of disease. While HbA1c reflects average glycaemia over the preceding 8 to 12 weeks and alterations in gene expression typically occur over a few hours or days, changes in kidney structure and function take years to develop [21]. There are other potential mechanisms that may promote PRKC-β gene transcription. Our own preliminary studies have demonstrated a smaller but no less significant increase in gene expression of PRKC-β in biopsies from progressive IgAN (IgA nephropathy) [22].

In combination with clinical data, formalin-fixed, paraffin-embedded tissue provides an important resource for determining structure–function relationships in a range of human diseases. Formalin fixation, however, causes cross-links between RNA and protein, as well as the addition of mono-methylol groups to RNA bases and methylene bridging of adenine groups and fragmentation [23]. These changes in RNA structure have, until recently, limited not only the amount of RNA that can be extracted from archival tissue, but also its quantitative analysis. However, technological advances now allow for successful RNA extraction from formalin-fixed tissues by elevating the buffer temperature and by prolonged proteinase K digestion prior to acid-phenol-chloroform extraction and carrier precipitation [24]. Although such extraction methods provide good yields, RNA from formalin-fixed tissue is still fragmented. This previously insurmountable problem is now easily overcome using real-time PCR which allows for high-fidelity amplification of RNA fragment lengths of 50 to 150 bp. Furthermore, since the RNA degradation that occurs in formalin-fixed tissue is universal, housekeeping RNAs will be equally affected and remain suitable for use in quantitative analyses [24].

Although molecular analyses of archival tissues are now commonplace in the oncology literature, their use in nephrological studies has been less extensive, despite the frequency of kidney biopsies. The present report indicates that archival, formalin-fixed, paraffin-embedded human kidney tissue provides a valuable resource.

In summary, the present study demonstrates that PRKC-β mRNA is upregulated by high glucose in tubular epithelial cells and in advanced diabetic nephropathy. The mechanisms underlying this upregulation in PRKC-β expression and the temporal changes in its distribution in the nephron are uncertain, but underscore the need not only for further exploration of the regulatory elements of PRKC-β transcription, but also for prospective, longitudinal analyses of both gene and protein expression in human kidney tissue.