Biochemical Journal

Research article

Heparanase enhances nerve-growth-factor-induced PC12 cell neuritogenesis via the p38 MAPK pathway

Hengxiang Cui , Chenghao Shao , Qin Liu , Wenjie Yu , Jianping Fang , Weishi Yu , Amjad Ali , Kan Ding

Abstract

Heparanase is involved in the cleavage of the HS (heparan sulfate) chain of HSPGs (HS proteoglycans) and hence participates in remodelling of the ECM (extracellular matrix) and BM (basement membrane). In the present study we have shown that NGF (nerve growth factor) promoted nuclear enrichment of EGR1 (early growth response 1), a transcription factor for heparanase, and markedly induced heparanase expression in rat adrenal pheochromocytoma (PC12) cells. K252a, an antagonist of the NGF receptor TrkA (tyrosine kinase receptor A), decreased heparanase protein expression induced by NGF in PC12 cells. Suramin, a heparanase inhibitor, decreased heparanase in PC12 cells and blocked NGF-induced PC12 neuritogenesis. Stable overexpression of heparanase activated p38 MAPK (mitogen-activated protein kinase) by phosphorylation and enhanced the neurite outgrowth induced by NGF, whereas knock down of heparanase impaired this process. However, overexpression of latent pro-heparanase with a Y156A mutation still led to enhanced NGF-induced neurite outgrowth and increased p38 MAPK phosphorylation. Inhibition of p38 MAPK by SB203580 suppressed the promotion of NGF-induced neuritogenesis by the wild-type and mutant heparanase. The impaired differentiation by knock down of heparanase could be restored by transfection of wild-type or mutant heparanase in PC12 cells. The results of the present study suggest that heparanase, at least in the non-enzymatic form, may promote NGF-induced neuritogenesis via the p38 MAPK pathway.

  • heparanase
  • neurite outgrowth
  • neuritogenesis
  • neuron differentiation
  • p38 mitogen-activated protein kinase (MAPK)

INTRODUCTION

HSPGs [HS (heparan sulfate) proteoglycans] are ubiquitously distributed on the cell surface, ECM (extracellular matrix) and BM (basement membrane) of a wide range of cells in vertebrate and invertebrate tissues [1,2]. HSPGs consists of a core protein to which several linear HS GAG (glycosaminoglycan) chains are covalently O-linked. HSPGs play key roles in the self-assembly and integrity of the multimolecular architecture of the BM and ECM [2,3]. Hence, they affect diverse biological processes, such as cell migration, embryonic morphogenesis, angiogenesis, metastasis, inflammation, neurite outgrowth and tissue repair. HS on HSPGs act as co-receptors for growth factors, such as FGF (fibroblast growth factor) and VEGF (vascular endothelial growth factor) to influence cell behaviour, including differentiation [4,5]. Theoretically, the HS structural motif modified by its biosynthesis and/or degradation enzyme may mediate cell differentiation [610]. Heparanase, which is considered to be a dominant endo-β-D-glucuronidase in mammalian tissues and has been identified in a variety of cell types and tissues, contributes to the cleavage of the HS chains on HSPGs and hence participates in degradation and remodelling of the ECM and BM [11,12]. Heparanase activity is commonly considered to be correlated with the metastatic ability of various tumour-derived cells [1315]. In addition to tumour-derived cells, heparanase is also expressed in normal cells, including leucocytes such as neutrophils, macrophages, endothelial cells and astrocytes [12,1618]. Heparanase expressed by these cells is thought to facilitate cellular migration and invasion, which is associated with autoimmunity, inflammation and angiogenesis. It is also believed to function as an adhesion molecule, which mediates cell adhesion to the ECM resulting in integrin-dependent cell spreading, tyrosine phosphorylation of paxillin and reorganization of the actin cytoskeleton [19,20], independent of its endoglucuronidase activity. A previous study demonstrated that both heparanase mRNA and protein are also expressed in neuronal cells and glial cells. Heparanase immunoreactivity was found not only in the cytoplasm, but also in the nucleus of the neurons [21]. These findings suggest that heparanase probably plays roles in the normal function of the CNS (central nervous system) [21], as mobile HS fragments generated as a result of the enzymatic heparanase activity may facilitate HS-binding growth factor signalling in the regulation of proliferation and differentiation [22].

Rat adrenal pheochromocytoma (PC12) cells have been extensively used as a model of catecholaminergic neurons in culture, as well as for the study of neuronal apoptotic death. Associated with the NGF (nerve growth factor)-induced morphological change is the growth of axon-like processes, called neuritogenesis, which is characterized by growth arrest, the elaboration of long branching neurites and electrical excitability [23]. The morphological hallmark of neuronal differentiation is neurite sprouting, elongation and subsequent maturation of neurites into axons and dendrites. Phenotypic changes associated with NGF-induced differentiation include the biosynthesis of neurotransmitters, the acquisition of electrical excitability, along with the growth of axon-like extensions during neuritogenesis [23]. Neuritogenesis is a complex phenomenon that involves multiple interactions between the growing neurite and the ECM. Although increased heparanase expression was observed during NGF-induced neuronal differentiation in PC12 cells, the function of heparanase in neuritogenesis is not clear [24]. As heparanase participates in remodelling of the ECM and BM, which are essential for neuronal differentiation, in the present study we investigated its role in neuritogenesis using the PC12 model.

EXPERIMENTAL

General materials

NGF, Lipofectamine™ 2000 and 1×DMEM (Dulbecco's modified Eagle's medium) were from Invitrogen. Suramin sodium salt, the TrkA antagonist K252a and MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide] were from Sigma–Aldrich. The selective p38 MAPK (mitogen-activated protein kinase) inhibitor (SB203580) was purchased from Calbiochem. The polyclonal antibody (catalogue number ab59787) against heparanase was from Abcam. The anti-(p44/42 MAPK) [ERK1/2 (extracellular-signal-regulated kinase 1/2)] antibody (catalogue number 9102), anti-(phospho-p44/42 MAPK) (ERK1/2) (Thr202/Tyr204) antibody (catalogue number 3477), anti-SAPK (stress-activated protein kinase)/JNK (c-Jun N-terminal kinase) antibody (catalogue number 9252), anti-(phospho-SAPK/JNK) (Thr183/Tyr185) antibody (catalogue number 4668), and monoclonal antibodies against phospho-p38 MAPK (Thr180/Tyr182) (catalogue number 4511) and EGR1 (early growth response 1) (catalogue number 4153), and a polyclonal antibody against p38α MAPK (catalogue number 9212) were purchased from Cell Signaling Technology. The anti-(histone H1) (catalogue number sc-8030) antibody was from Santa Cruz Biotechnology. BrdU (bromodeoxyuridine), rabbit anti-actin antibody, anti-BrdU monoclonal antibody and FITC-conjugated goat anti-mouse antibody were from Sigma–Aldrich. The pMD-18T easy-clone vector was from TaKaRa Biotechnology. IRDye® 800CW-conjugated goat (polyclonal) anti-rabbit IgG (catalogue number 926-32211) and IRDye® 800CW-conjugated goat (polyclonal) anti-mouse IgG (catalogue number 926-32210) were from LI-COR Biosciences. The M-MLV (Moloney murine leukaemia virus) reverse transcriptase, endonucleases and T4 ligase were from TaKaRa Biotechnology.

Cell culture and stable transfections

PC12 cells were obtained from the American Type Cell Culture collection. Native PC12 cells were maintained in DMEM (Invitrogen) supplemented with 10% heat-inactivated horse serum (Invitrogen) and 5% FBS (fetal bovine serum, Invitrogen) at 37°C in a humidified atmosphere containing 5% CO2. Lipofectamine™ 2000 (Invitrogen) was used for transfections according to the manufacturer's recommendations. The stably transfected PC12 cells were screened under neomycin selection. Briefly, after the vectors were transfected into PC12 cells using the Lipofectamine™ 2000 reagent, the cells were diluted at a ratio of 1:20 before being selected with 800 μg/ml neomycin (G-418). After 4 weeks under selection, independent colonies each grown from one cell were obtained by limiting dilution. Western blot analysis was used to confirm the stable expression of heparanase in the cells using a heparanase-specific antibody.

BrdU incorporation and MTT assay

Cells were incubated in the presence of 100 μM BrdU (Sigma–Aldrich) for 1 h after the cells were treated with or without suramin for 24 h or 72 h. The cells were then fixed with 4% paraformaldehyde and treated with 2 M HCl containing 1% Triton X-100. The cells were stained with an anti-BrdU monoclonal antibody (Sigma–Aldrich) and then stained with an FITC-conjugated goat anti-mouse antibody and DAPI (4′,6-diamidino-2-phenylindole). The green immunofluorescence, indicating BrdU staining, the total number of DAPI-positive cells, as well as BrdU-labelled cells, was counted in five to ten different fields of each well. In the MTT assay, 1×104 PC12 cells were treated with or without suramin for 72 h before being washed three times with PBS. MTT was then added to the cells at a final concentration of 0.5 mg/ml and incubated for 4 h. Finally, 100 μl of DMSO was added followed by an absorbance measurement at 570 nm using a Universal Microplate Reader (Bio-Tek).

Generation of heparanase and mutant heparanase expression constructs

The human HPSE (GenBank® accession number NM_006665) gene encoding heparanase was cloned using a PCR-based method. Briefly, total RNA from HEK (human embryonic kidney)-293 cells was reverse-transcribed into first strand cDNA with M-MLV reverse transcriptase. Subsequently, the CDS fragment of HPSE was amplified by overlapping PCR with PrimeSTAR HS DNA Polymerase (TaKaRa). The following primers were used: forward primer 5′-TATAGAATTCACCATGGTCCTGCGCTCG AAGC-3′ (A1), and reverse primer 5′-CTCCTGGTAGGGCCATTCCAACCGTAA-3′ (A2); and forward primer 5′-GAGCCCTCGTTCCTGTCCG-3′ (B1), and reverse primer 5′-CCTCTAGACTGATGCAAGCAGCAACTTTGGCATT-3′ (B2). Underlining indicates the specific sequences recognized by restriction endonucleases which were used for DNA construction. First, PCR was performed using primers A1 and A2 to obtain fragment A, and then performed with primers B1 and B2 to produce fragment B. A second PCR with primers A1 and B2 was then used to obtain the full-length clone using fragments A and B as a template after gel-purifying extraction (the molar ratio of A and B was 1:1). After digesting with the endonucleases EcoRI and XbaI, the amplified product of 1652 bp was inserted into pCDNA3.1/myc-HisB (Invitrogen) with T4 ligase, forming pCDNA3.1/myc-HisB-hHPSE which was the final heparanase expression plasmid with Myc and His tags fused to the C-terminus. This construct was sequenced by Sangon Biological Engineering Technology & Services (Shanghai, China). After transfection of PC12 cells with this construct, three stable cell lines were obtained after selection by neomycin and designated 9406, 9415 and 9416.

Since substitution of the conserved tyrosine residue at position 156 with alanine has been shown to abolish normal processing and activation of pro-heparanase [25], a Y156A mutant heparanase was constructed by site-directed mutagenesis. First, the full-length wild-type HPSE fragment from pCDNA3.1/myc-HisB-hHPSE was inserted into pLVX-IRES-ZsGreen1 at the EcoRI and XbaI sites to obtain the pLVX-IRES-ZsGreen1-HPSE vector (ZS-HPSE). In ZS-HPSE, there are two BamHI digestion sites. One is at nt 389 of the full-length wild-type HPSE (position ‘A’ in the start codon ‘ATG’ was counted as the first base). The other is at nt 8 after the XbaI site within the backbone vector pLVX-IRES-ZsGreen1. Secondly, the mutant HPSE fragment, corresponding to nt 390–1629, also containing two BamHI digestion sites, was substituted into ZS-HPSE at the two BamHI sites. The mutant HPSE fragment was amplified by two-step PCRs. The first-step PCR was conducted using the primer pairs HPSEMUT-STEP1-F, 5′-gagaagttacggtTGGAATGGCCCTACCAGGAGCAATTGCTACTCCGAGAACACgagCA-3′, and BamH1-HPSE-R, 5′-GGATCCTCAGATGCAAGCAGCAACTTTG. gag is the mutated site designed in the primer. The second-step PCR, using the PCR product after gel-purifying extraction of the first-step PCR as a template, was performed with the primers BamH1-389-STEP2-F, 5′-CAAATATGGATCCATCCCTCCTGATGTGGAGgagaagttacggt-3′ and BamH1-HPSE-R (primer sequence described above). The bold lowercase sequences in the two sense PCR primers are the overlapping bases. Underlining indicates the specific sequences recognized by restriction endonucleases which were used for DNA construction. The PCR product of the second step contains the mutant HPSE fragment (from 390 to 1629) and was subcloned into the pMD-18T vector before being sequenced and digested with the endonuclease BamHI. Subsequently, the BamHI-digested fragment obtained after gel-purifying extraction was substituted into ZS-HPSE at the two BamHI sites, obtaining pLVX-IRES-ZsGreen1-M-HPSE (ZS-M-HPSE).

Knock down of the heparanase gene by shRNA (short hairpin RNA)

The heparanase shRNA expression vector pGCsi-U6/Neo/GFP (GFP is green fluorescent protein) was designed and constructed by the Genechem Company. The target sequence is 5′-ACCTCCATAATGTCACCAA-3′, with the loop designed as TTCAAGAGA. The shRNA vector was transfected into PC12 cells using Lipofectamine™ 2000 according to the manufacturer's protocol. The empty vector served as a negative control. After transfection of PC12 cells with the shRNA vector and selection with neomycin, three stable cell lines, Sh-2, Sh-7 and Sh-12, were obtained and confirmed by RT (reverse transcription)–PCR and Western blot analysis (Figures 2A and 2B).

Lentivirus-based mutant heparanase overexpression

Virus carrying mutant heparanase was produced by transfecting HEK-293T [HEK-293 cells expressing the large T-antigen of SV40 (simian virus 40)] cells with pLVX-IRES-ZsGreen1 (ZS), pLVX-IRES-ZsGreen1-HPSE (ZS-HPSE) or pLVX-IRES-ZsGreen1-M-HPSE (ZS-M-HPSE) with viral packaging vectors (psPAX2, pMD2G) using a standard calcium phosphate transfection method. Viruses were harvested from the supernatant at 48 h post-transfection and concentrated by ultracentrifugation at 35000 rev./min before infecting 2×105 PC12 cells with 1×107 TCID50. The cells were used for neuritogenesis assays or Western blot analysis after they were cultured in complete medium without virus for 24 h.

RT–PCR

Total RNA was isolated using TRIzol® reagent (Invitrogen). RT was performed with random hexamers as primers. The final reaction volume was 20 μl, with 1 μg of total RNA from each cell sample. The reaction mixture contained 0.5 mM of each dNTP, 2 units/μl RNase inhibitor, 3 mM MgCl2, 10 units/μl M-MLV (AMV) reverse transcriptase and 150 ng of random hexamers, following the manufacturer's protocol. The following specific PCR primers were used for amplification: rat heparanase sense, 5′-CAAGAACAGCACCTACTCACGAAGC-3′; rat heparanase antisense, 5′-CCACATAAAGCCAGCTGCAAAGG-3′; human heparanase sense, 5′- GCTACTCCGAGAACACTACCAGA-3′; human heparanase antisense, 5′-TGAATCAATCACTTCTCCACCAG-3′; β-tubulin III sense, 5′-GGAACATAGCCGTAAACTGC-3′; β-tubulin III antisense, 5′-TCACTGTGCCTGAACTTACC-3′; GAP43 (growth-associated protein 43) sense, 5′-TGCTGTGCTGTATGAGAAGAACC-3′; GAP43 antisense, 5′-GGCAACGTGGAAAGCCGTTTCTTAAAGT-3′; 18S rRNA sense, 5′-GTGGAGCGATTTGTCTGGTT-3′; and 18S rRNA antisense, 5′-CGCTGAGCCAGTCAGTGTAG-3′. PCR amplification was conducted using the following conditions: denaturation for 2 min at 94°C, and 94°C for 50 s, annealing for 50 s at 55°C, and extension for 1 min at 72°C (28 cycles for 18S rRNA and 30 cycles for human and rat heparanase, β-tubulin III and GAP43).

Western blot analysis

Cell extracts were prepared by lysis of (1–5)×106 cells in 50–200 μl of RIPA buffer [25 mM Tris/HCl (pH 7.4), 150 mM KCl, 5 mM EDTA, 1% Nonidet P40, 0.5% sodium deoxycholate and 0.1% SDS] containing 1 mM PMSF and a protease inhibitor cocktail (Sigma–Aldrich) on ice for 30 min, followed by removal of DNA and cell debris by centrifugation at 12000 g for 5 min at 4°C. The resulting supernatants were collected and frozen at −80°C or used immediately. Protein concentrations were measured with the DC Protein Assay Kit (Bio-Rad) following the manufacturer's recommendations. Then, 50 μg of protein/lane were separated by SDS/PAGE (10% gels) (unless specified otherwise), transferred on to PVDF or nitrocellulose membranes (Bio-Rad) and immunoblotted with antibodies. Specific peroxidase-conjugated secondary antibodies or IRDye® 800CW-conjugated goat (polyclonal) anti-mouse (or rabbit) IgG were used to detect protein expression with an enhanced chemiluminescence kit (Pierce) or the Odyssey infrared imaging system (LI-COR Biosciences). Nuclear protein extraction and EMSAs (electrophoretic mobility-shift assays) were conducted as described previously [26]. Briefly, after NGF (50 ng/ml) treatment, the cells were harvested by scraping and resuspending in 25 mM Hepes, 1.5 mM EDTA and 1 mM DTT (dithiothreitol) (pH 7.6), and homogenized with 10% glycerol. The cell lysates were then centrifuged at 4°C, and protein concentrations of the nuclear extracts were then determined using a BCA (bicinchoninic acid) assay. Aliquots of nuclear protein were then frozen and stored at −80°C until use.

Neuritogenesis assay

In neuritogenesis experiments, cells were seeded on 24-well plates (Corning) at a density of 2×103 cells/well. After incubation overnight, the cells were treated with medium containing 0.5% horse serum and 25 ng/ml 2.5S NGF (Invitrogen). Morphology of PC12 cells treated with NGF (25 ng/ml) or NGF plus the p38 MAPK inhibitor SB203580 (20 μM) for 72 h was evaluated by counting the proportion of cells containing at least one neurite that was twice as long as the diameter of the cell body. Neurite outgrowth was observed with nine random images captured per well, using either an inverted fluorescence microscope (cells) or a phase-contrast microscope. The images were captured by a digital camera at ×200 magnification. All cells in each image were analysed, with over 200 cells assessed per cell culture plate. The experiments were repeated at least three times. Mean and S.D. values were determined for each treatment.

Statistical analysis

Results are expressed as the means±S.E.M. The data were analysed using Student's t test. P values of <0.05 indicated significant differences (*P<0.05; **P<0.01).

RESULTS

NGF-induced heparanase expression in PC12 cells is attenuated by the heparanase inhibitor suramin

Heparanase mRNA and protein are expressed in neuronal cells and glial cells [22], implying that heparanase may play roles in the normal function of the CNS [22,24]. We used a conventional neuronal cell model, the PC12 cell line responsive to NGF, to study its role in neural cell apoptotic death, proliferation and differentiation [24].

After treatment with NGF (50 ng/ml) for 16 h, heparanase mRNA was notably up-regulated in PC12 cells (Figure 1A), consistent with previously published results [24]. NGF also increased heparanase protein expression in PC12 cells in a time-dependent manner (Figure 1B). It was reported that NGF at least binds to its two receptors, p75NTR and TrkA [27]. However, only the activated form of TrkA, but not that of p75NTR, is significantly co-localized with heparanase [28]. This suggests that TrkA is probably involved in heparanase expression. To investigate whether the heparanase expression correlated with the NGF receptor, K252a, an antagonist of TrkA [29], was used to evaluate the effect of NGF on heparanase expression. K252a (100 nM) inhibited the heparanase expression induced by NGF, indicating the involvement of TrkA signalling (Figure 1C). It is known that NGF binding of TrkA leads to the activation of ERK signalling and rapid recruitment of EGR1 [30], a nuclear transcription factor for heparanase in PC12 cells [31,32]. Indeed, EGR1 was dramatically recruited into the nucleus when induced by NGF in PC12 cells, as demonstrated by Western blot analysis (Figure 1D) and EMSA (Supplementary Figure S1 at http://www.BiochemJ.org/bj/440/bj4400273add.htm). These results are consistent with those reported previously [33]. Thus we deduced that the NGF induction of heparanase in PC12 cells was likely to be mediated by the TrkA-EGR1 pathway.

Figure 1 NGF induced heparanase mRNA expression, whereas heparanase inhibitor blocked neuritogenesis induced by NGF

NGF was used at 50 ng/ml. (A) Heparanase (Hepa) expression in PC12 cells treated with or without NGF for 16 h was analysed by RT–PCR with 18S rRNA as the internal control. (B) Heparanase proteins were detected in cell extracts of PC12 cells treated with or without NGF for 0, 16, 24, 48 or 72 h by Western blot analysis with β-actin as a loading control. (C) Heparanase proteins were detected in PC12 cells treated with or without NGF combined with (+) or without (−) K252a (100 nM) by Western blot analysis. (D) Nuclear protein extracts of PC-12 cells treated with NGF for 0–8 h were analysed by Western blot with anti-EGR-1 and anti-(histone H1) antibodies. (E) Neuritogenesis of PC12 cells treated without or with NGF only or with NGF and suramin (20 μM) for 72 h under a phase-contrast microscope (×200 maginfication). The arrow indicates a typical neurite at least twice as long as the diameter of the cell body from the differentiated PC12 cell. (F) The effect of suramin (20 μM) on heparanase expression was measured by Western blot analysis in the presence (+) or absence (−) of NGF (25 ng/ml). (G) PC12 cell viability was detected by MTT assay after treatment with or without suramin at 20 μM or 200 μM for 72 h, with the viability of untreated cells set to 100%. (H) BrdU incorporation into the cell DNA was also used to evaluate the effect of suramin on PC12 cell viability. Cells were incubated with 100 μM BrdU for 1 h after treatment with or without suramin for 72 h. BrdU-labelled cells, as percentages of total DAPI-positive cells, were determined by counting five to ten different fields of each well after immunofluorescent staining as described in (I). Data are represented as means±S.D. in (G) and (H). (I) Immunofluorescent images of PC12 cells treated as described in (H) were fixed and probed with an anti-BrdU monoclonal antibody before being detected with an FITC-conjugated goat anti-mouse secondary antibody and stained with DAPI.

As NGF promotes PC12 cell neuritogenesis, we hypothesized that the increased expression of heparanase may contribute to the process of neurite outgrowth. Therefore suramin, a chemical inhibitor of heparanase, was used to explore the role of heparanase in neuritogenesis. PC12 cells were incubated with NGF in the presence or absence of suramin (20 μM) for 72 h. We found that suramin nearly completely blocked neuritogenesis induced by NGF (50 ng/ml) in PC12 cells compared with the control group (Figure 1E). Further investigation revealed that suramin decreased the heparanase expression in PC12 cells treated with NGF for 24 h (Figure 1F). To confirm that the inhibition of neuritogenesis by suramin was not due to cell toxicity, the effect of suramin on PC12 cell viability was tested using the MTT assay and BrdU incorporation method. In the MTT assay, suramin, even at a concentration of 200 μM, did not significantly reduce cell viability after treatment for 72 h (Figure 1G). Consistently, suramin also did not affect BrdU incorporation into DNA in PC12 cells after treatment at both concentrations of 20 μM and 200 μM (Figures 1H and 1I respectively) for 72 h. Therefore the inhibition of NGF-induced neuritogenesis by suramin observed at 20 μM in PC12 cells was unlikely to be due to any effect on cell viability. Because suramin as an inhibitor lacks appropriate specificity for heparanase activity, it could potentially affect heparanase expression through inhibiting the NGF-induced PC12 neuritogenesis. Taken together, these results implicated heparanase in the process of neuritogenesis.

Knock down of heparanase expression impairs NGF-induced neuritogenesis of PC12 cells

To further determine whether heparanase is required for the promotion of neuritogenesis by NGF, we then knocked down the expression of heparanase using shRNA. To achieve this, PC12 cells were stably transfected with heparanase shRNA. The heparanase expression at the mRNA (Figure 2A) and protein (Figure 2B) levels were clearly knocked down in three heparanase shRNA lines (Sh-2, Sh-7 and Sh-12). The expression of the pro-heparanase protein (65 kDa) in the above three shRNA cell lines was lower than that in the negative control cells and sham cells (Figure 2B).

Figure 2 Knock down of heparanase expression led to arrest of neuritogenesis induced by NGF

(A) mRNA levels of heparanase (Hepa) were detected by RT–PCR in cells stably transfected with three clonal PC12 cell lines carrying shRNA against heparanase (Sh-2, Sh-7 and Sh-12), control cells (Con) and vector control cells (GFP). 18S rRNA was used as the internal control. (B) Heparanase protein expression was detected by Western blot analysis in Sh-2, Sh-7 and Sh-12 cell lines, vector control (GFP) and control (Con) cells probed with an anti-heparanase polyclonal antibody. Arrows indicate the molecular mass of pro-heparanase (65 kDa) and the active form of the enzyme (50 kDa). (C) Histogram showing the proportion of cells bearing one neurite at least twice the length of the diameter of the cell body in the neuritogenesis assay. The three cell clones (Sh-2, Sh-7 and Sh-12), control cells (Con) and vector control (GFP) cells were seeded in 24-well plates at 2×103 cells/well and incubated overnight, followed by exposure to medium (0.5% horse serum) containing 25 ng/ml NGF for 72 h. Neurite outgrowth was observed with nine random fields per well, and the captured images are shown in Supplementary Figure S2 (at http://www.BiochemJ.org/bj/440/bj4400273add.htm). All cells in each image were analysed, with over 200 cells assessed per cell culture plate. Results are represented as means±S.D. for each treatment, and the experiments were repeated at least three times. **P<0.01 and *P<0.05 for Sh2, Sh7, and Sh12 compared with GFP, by Student's t test. (D) Histogram showing the proportion of cells bearing one neurite at least twice the length of the diameter of the cell body in each group in the rescue experiment. (E) Morphology images of neurite outgrowth in the rescue experiment are shown. In the rescue experiment, Sh-2 was transiently transfected with control vector pLVX-IRES-ZsGreen1 (ZS) (Sh-2+ZS), vector containing wild-type heparanase pLVX-IRES-ZsGreen1-HPSE (Sh-2+ZS-HPSE) or vector subcloned with mutant heparanase pLVX-IRES-ZsGreen1-M-HPSE (Sh-2+M-HPSE) lentiviral vectors as described in the Experimental section in the presence (+) or absence (−) of NGF (25 ng/ml) followed by morphological evaluation. The vector control of the Sh-2 (pGCsi-GFP) stably transfected PC12 cells (GFP) were transfected with pLVX-IRES-ZsGreen1 to serve as a control group. The neuritogenesis assay was performed as described in (C).

We further examined the effect of knock down of heparanase on the NGF-induced neurite outgrowth in PC12 cells using the Sh-2, Sh-7 and Sh-12 clonal cell lines. These stable heparanase shRNA cells incubated with NGF for 72 h had reduced early neurite extension compared with the wild-type counterpart and vector control (Supplementary Figure S2 at http://www.BiochemJ.org/bj/440/bj4400273add.htm). Evaluation of the morphology revealed that the proportion of cells bearing at least one neurite twice as long as the diameter of the cell body in the three cell clones was significantly less than that of the control group and sham group (Figure 2C). Furthermore, the rescue experiment showed that the impaired differentiation induced by NGF of cells with knock down of heparanase was restored by transfection of heparanase (wild-type or its enzymatically inactive mutant) into these cells (Figures 2D and 2E). On the basis of these results, we deduced that heparanase (or perhaps the pro-heparanase), at least in part, played a positive role in NGF-induced PC12 cell neuritogenesis.

Overexpression of heparanase enhances NGF-induced PC12 cell neuritogenesis

Since NGF-induced PC12 neuritogenesis was attenuated by knock down of the heparanase gene, we further evaluated how overexpression of heparanase would affect this process by generating the construct pcDNA3.1/myc-HisB-hHPSE. PC12 cells were stably transfected with this heparanase overexpression plasmid using the same method as described in the construction of the knock down cell clones, which ultimately resulted in three cell clones, designated 9406, 9415 and 9416. Expression of heparanase at the mRNA (Figure 3A) and protein (Figure 3B) levels in the three chosen clones was much higher than in the control and vector sham control cells when detected using an antibody against heparanase or the Myc tag (results not shown), indicating that heparanase was stably overexpressed in these cells.

Figure 3 Heparanase promoted neuritogenesis induced by NGF

(A) Human heparanase (Hepa) mRNAs were detected by RT–PCR in wild-type (Con) cells, vector sham control (pcDNA3.1) cells and in three PC12 cell clones (9406, 9415 and 9416) stably transfected with human heparanase. (B) Heparanase protein expression in wild-type cells, vector sham control cells and the three chosen clones were determined by Western blot analysis using a polyclonal anti-(human heparanase) antibody. Arrows indicate the pro-heparanase (65 kDa) and enzyme-active (50 kDa) forms. (C) Histogram showing the proportion of cells bearing at least one neurite twice as long as the diameter of the cell body in the 9406, 9415 and 9416 clones, wild-type (Con) and vector sham control (pcDNA3.1) cells. Heparanase-overexpressing cells, control cells and sham cells were seeded at 2×103 cells/well and then treated with NGF and evaluated for morphology as described in Figure 2. Data are represented as means±S.D. for triplicate treatments, and the experiments were repeated at least three times. The data were analysed by Student's t test. *P<0.05, 9406 compared with pcDNA3.1; **P<0.01, 9415 compared with pcDNA3.1; *P<0.05, 9416 compared with pcDNA3.1. Morphology images are shown in Supplementary Figure S3 at http://www.BiochemJ.org/bj/440/bj4400273add.htm.

Neurite outgrowth in the heparanase-overexpressing PC12 cells after stimulation with NGF for 72 h was indeed enhanced compared with the control groups (Supplementary Figure S3 at http://www.BiochemJ.org/bj/440/bj4400273add.htm). Morphology analysis showed that the proportion of cells bearing at least one neurite twice as long as the diameter of the cell body in the three cell clones was significantly greater than that of the control groups (Figure 3C). On the basis of the results above, we suggest that heparanase (or perhaps pro-heparanase) enhanced NGF-induced PC12 neuritogenesis.

p38 MAPK phosphorylation is required for neuritogenesis promoted by heparanase

In order to uncover the mechanism underlying the enhancement of neuritogenesis by heparanase, we evaluated the phosphorylation levels of p38 MAPK, ERK and JNK in all of the stable cell lines generated above. Phosphorylation of p38 MAPK was induced in the cell lines stably expressing heparanase (Figure 4A and Supplementary Figure S4C at http://www.BiochemJ.org/bj/440/bj4400273add.htm), whereas it was impaired in those with heparanase stably silenced (Figure 4B and Supplementary Figure S4D). However, there was no obvious change in the phosphorylation levels of ERK and JNK in those stable cells (Supplementary Figures S4A and S4B respectively). The results suggested that phosphorylation of p38 MAPK may be involved in NGF-induced PC12 cell neuritogenesis augmented by heparanase. To test this hypothesis, PC12 cells stably overexpressing heparanase were treated with the p38 MAPK phosphorylation inhibitor SB203580 (20 μM) for 72 h, followed by analysis of p38, ERK and JNK MAPK signalling components by Western blot. Indeed, although p38 MAPK phosphorylation was induced by heparanase, this effect was completely blocked by SB203580 (Figure 4C), whereas ERK and JNK phosphorylation were not disturbed by this inhibitor (Supplementary Figure S4E). To explore further the role of the p38 MAPK pathway in the heparanase enhancement effect on NGF-induced neurite outgrowth of PC12 cells, neuritogenesis was examined using the p38 MAPK phosphorylation inhibitor SB203580. The promoting effect of heparanase on NGF-induced PC12 cell neuritogenesis was blocked when the cells were treated with 20 μM SB203580 (Figures 4D and 4E). From these results, we deduced that heparanase could play a positive role in NGF-induced PC12 neuritogenesis through the p38 MAPK pathway. To further confirm the role of heparanase in the NGF-induced neuritogenesis in PC12 cells, the expression of the differentiation markers β-tubulin III and GAP43 was detected by RT–PCR in the presence of NGF in the heparanase stable cell lines. After NGF was added to the culture medium for 16 h, the mRNA expression levels of β-tubulin III and GAP43 were significantly higher in cells with overexpression of heparanase, whereas their expression was significantly lower in cells with heparanase stably silenced compared with the control and corresponding vector control cells (Figure 4F and Supplementary Figure S4F). These data suggest that heparanase may promote the differentiation induced by NGF.

Figure 4 p38 MAPK phosphorylation was required for heparanase to augment neuritogenesis induced by NGF

Phospho-p38 MAPK (Thr180/Tyr182) was detected by Western blot analysis in human heparanase-overexpressing (9406, 9415 and 9416) (A) and stably silenced (Sh-2, Sh-7, Sh-12) (B) PC12 cell clones compared with wild-type (Control in A and B) and vector control cells (pcDNA3.1 in A; GFP in B) respectively. Total p38 MAPK and β-actin proteins were used as loading controls. Phosphorylation of p38 MAPK normalized to total p38 MAPK in (A) and (B) is summarized in Supplementary Figure S4C and S4D (at http://www.BiochemJ.org/bj/440/bj4400273add.htm) respectively. (C) Phospho-p38 MAPK (Thr180/Tyr182) was detected by Western blot analysis in heparanase stably overexpressing clones, wild-type (Control) and vector control (pcDNA3.1) PC12 cells treated with (+) or without (−) the p38 MAPK inhibitor SB203580 (20 μM). (D) Heparanase stably overexpressing cell clones, wild-type (Con) and vector control (pcDNA3.1) cells were seeded on to 24-well plates at a density of 2×103 cells/well and cultured overnight, followed by treatment with medium (0.5% horse serum) containing 25 ng/ml NGF in the presence or absence of SB203580 (20 μM) for 72 h. The histogram shows statistical analysis of the proportion of cells bearing at least one neurite twice as long as the diameter of the cell body in human heparanase stably overexpressing cell clones, wild-type (Con) and vector control (pcDNA3.1) cells induced by NGF (25 ng/ml) treated with or without SB203580 (20 μM). Data were from at least triplicate experiments and are means±S.D. *P<0.05, **P<0.01 by Student's t test. (E) Neurite outgrowth observation and images in (D) were taken as described in Figure 2. (F) Neuron differentiation markers β-tubulin III and GAP-43 were measured by RT–PCR in heparanase stable silenced or overexpressing stable cell clones induce by NGF (25 ng/ml) compared with wild-type (Control) and vector control cells (GFP or pcDNA3.1). 18S rRNA was used as an internal control.

Enzymatically inactive mutant heparanase enhances NGF-induced neuritogenesis through p38 MAPK phosphorylation

To understand whether the latent form of heparanase (65 kDa) or the enzyme-active form (50 kDa) was responsible for differentiation of PC12 cells induced by NGF, an heparanase Tyr156 to alanine mutant was created and transfected into PC12 cells. Indeed, the latent heparanase mutant was highly expressed in PC12 cells (Figure 5A). In addition, only p38 phosphoryation of the components (ERKs, JNK and p38) of the MAPK pathway was induced by the mutant (Figure 5B). Because only pro-heparanase proteins (65 kDa) were notably increased in the 9406, 9415 and 9416 stable cells (Figure 3B) and decreased in the stable knock-down cells (Figure 2B), we further overexpressed the latent heparanase mutant to evaluate its effect on NGF-induced PC12 cell neuritogenesis. The results showed that forced expression of the pro-heparanase mutant in PC12 cells also enhanced the NGF-induced neurite outgrowth (Figure 5C and Supplementary Figure S5 at http://www.BiochemJ.org/bj/440/bj4400273add.htm). It seems that this effect was also dependent on p38 MAPK phosphorylation, since 20 μM SB203580 blocked the promotion of the mutant heparanase on NGF-induced PC12 cell neuritogenesis (Figure 5C and Supplementary Figure S5). The proportion of cells bearing at least one neurite twice as long as the diameter of the cell body in the cells which overexpressed the mutant pro-heparanase was significantly greater than that in the control groups (Figure 5C). Because suramin, the enzyme inhibitor of heparanase, blocked neuritogenesis induced by NGF, and the latent heparanase mutant was still able to augment such NGF-induced neuritogenesis, this finding suggested that heparanase promotes neuritogenesis at least in PC12 cells independently of its enzymatic activity. This finding may be partially explained by the fact that heparanase expression as both the latent (65 kDa) and active enzymatic forms (50 kDa) was inhibited by suramin (Figure 1F).

Figure 5 Mutant heparanase enhanced NGF-induced neuritogenesis through p38 MAPK phosphorylation

(A) After the lentiviral vector was prepared for 48 h in HEK-293T cells, the latent heparanase (Hepa) mutant was infected into PC12 cells for 24 h and detected by Western blot analysis with an anti-heparanase antibody. (B) Phospho-p38 MAPK (Thr180/Tyr182), phospho-p44/42 MAPK (ERK1/2) and phospho-SAPK/JNK (Thr183/Tyr185) were detected by Western blot analysis in mutant human heparanase-overexpressing cells (ZS-M-HPSE), compared with wild-type (Control) and vector control cells (pLVX-ZS). Total p38 MAPK, p44/42 MAPK (ERK1/2) and β-actin proteins served as loading controls. (C) The histogram shows the proportion of cells bearing at least one neurite twice as long as the diameter of the cell body in ZS-M-HPSE, control (Con) and pLVX-ZS cell groups in the absence or presence of NGF (25 ng/ml). Morphology evaluation was performed as described in Figure 2. Detailed morphology images are shown in Supplementary Figure S5 (at http://www.BiochemJ.org/bj/440/bj4400273add.htm).

DISCUSSION

Heparanase, commonly used as a HS-degrading enzyme, is considered to function in acidic environments of the ECM to facilitate tumour metastasis, angiogenesis and inflammation [34]. Apart from the well-studied catalytic feature of the enzyme, heparanase has also demonstrated functions apparently independent of its enzymatic activity. Non-enzymatic heparanase enhances cell adhesion [19,20,35,36] and induces p38 MAPK and Src phosphorylation [36] associated with induction of VEGF [37] and tissue factor [38] genes.

Although heparanase is often considered to regulate autoimmunity, inflammation and angiogenesis, the role of heparanase in normal development and tissue remodelling has not been well-investigated. To date, cell-surface heparanase has been implicated both in cell migration at early stages of embryogenesis and in subsequent morphogenesis of the cardiovascular and nervous systems in the chick embryo [39]. Similarly, knock down of both XHpaL and XHpaS (heparanase genes in Xenopus), results in the failure of Xenopus embryogenesis to proceed [6]. Moretti et al. [40] reported that augmented heparanase activity in the human olfactory epithelium may represent a physiological mechanism involved in neural cellular differentiation. In addition, it is of note that heparanase protein is detected in the dendrites of pyramidal neurons in the CA1 area of the brain, which indicates that heparanase might be implicated in the morphological maturation of spines [41].

In the present study, we first report that heparanase enhanced neuritogenesis independently of its enzymatic activity in the PC12 cell model, which has been widely used to investigate the cellular and molecular mechanisms underlying neuronal differentiation [23], because it alters in morphology into neuron-like cells and extends long and branching neuritis in response to NGF. We found that NGF mainly increased latent heparanase (Figures 1B, 1C and 1F) in PC12 cells, and the latent heparanase mutant also promoted neuritogenesis induced by NGF. This observation indicated that non-enzymatic heparanase has a predominant enhancing role in neuritogenesis in PC12 cells induced by NGF, which is distinct from that reported by Navarro et al. [24]. The results of the present study enrich the role of heparanase [19] and further pave the way for studies focusing on non-enzymatic activities of the heparanase molecule. Neurite outgrowth is a fundamental event in the development of the brain, as well as in the regeneration of damaged nervous tissue [42]. Cell-adhesion molecules, such as NCAM (neuronal cell-adhesion molecule) [43] and cadherins [44], transfer signals from the extracellular space which regulate the growth of the neurite. The non-enzymatic heparanase was reported to mediate cell attachment by recruiting paxillin with increased tyrosine phosphorylation [19], which has been linked to a remodelling of the actin cytoskeleton that leads to cell spreading and neurite formation [45]. Thus non-enzymatic heparanase may, as an adhesion molecule, modulate neuritogenesis signals from the extracellular space. The present study revealed that NGF also moderately increased enzymatic heparanase in PC12 cells (Figures 1B and 1C). The enzymatic heparanase can release GAG fragments from HSPGs [9], which was reported to promote neuritogenesis in vitro and stimulate nerve re-growth and muscle re-innervations [46,47]. Although enzymatically inactive mutant heparanase can still promote NGF-induced neuritogenesis (Figure 5C and Supplementary Figure S5), we cannot exclude the role for enzymatic heparanase in NGF-induced PC12 neuritogenesis. This suggestion is based on the experimental facts that (i) NGF not only induces latent heparanase expression, but also mildly increases enzymatic heparanase expression; (ii) forced latent heparanase expression is associated with expression of its enzymatic active form (Figure 3B), whereas knockdown of latent heparanase by siRNA (small interfering RNA) also slightly decreases enzymatic heparanase expression in PC12 cells; (iii) overexpressed mutant enzymatic heparanase enhances an heparanase enzymatic active form expression in PC12 cells (Figure 5A); and (iv) suramin dramatically disrupted PC12 neuritogenesis induced by NGF (Figure 1E) and also inhibited expression of both enzymatic and non-enzymatic heparanase (Figure 1F), indicating that this protein may promote NGF-induced PC12 differentiation as a zymogen and in an active form. All of these results suggest that heparanase augmentation of neuritogenesis induced by NGF might be through both the enzymatic and non-enzymatic form. Since more EGR1, a transcription factor of heparanase, is indeed recruited into the nucleus by NGF (Figure 1D and Supplementary Figure S1) for heparanase transcription, heparanase could be a component of the TrkA-EGR1 pathway which modulates NGF-induced PC12 neuritogenesis.

Previous studies have demonstrated that sustained activation of the ERK MAPK pathway is crucial for neuronal neuritogenesis of PC12 cells [48,49]. However, the p38 MAPK signalling pathway was also implicated in NGF-induced neuritogenesis of PC12 cells [50]. NGF induces sustained activation of the p38 MAPK signalling pathway in PC12 cells, and selective blockade of this cascade results in inhibition of neurite outgrowth [50]. Interestingly, neurite outgrowth induced by constitutively activated MEK (MAPK/ERK kinase), considered a regulator of ERK activity, also depends partly on p38 MAPK activity, although MEK activates both ERK and p38 MAPKs. In fact, MEK-induced PC12 cell neurite outgrowth is sensitive to inhibition of p38 MAPK by SB203580 [50]. Therefore a well-balanced activation of the ERK and p38 MAPK pathways may be necessary for neuritogenesis of PC12 cells in response to NGF. In the present study, we found that overexpression of heparanase (both wild-type and mutant latent heparanase molecules) promoted the constitutive phosphorylation of p38 MAPK, whereas silencing of heparanase attenuated the signal. However, phosphorylation of ERK and JNK was not obviously changed in cells with heparanase (wild-type and mutant) overexpressed or silenced. Meanwhile, SB203580, an inhibitor of p38 MAPK, nearly completely abolished the promotion of heparanase (wild-type and mutant) on PC12 neuritogenesis. On the basis of these findings, we deduced that both wild-type and mutant heparanase enhanced phosphorylation of p38 MAPK in PC12 cells, which promoted NGF-induced PC12 cell neuritogenesis.

In conclusion, the present study demonstrated that heparanase was involved in differentiation of PC12 cells via p38 MAPK phosphorylation, whereas the latent form of heparanase might play an independent role in neuritogenesis. Further studies are necessary to determine whether this is a neural-restricted or a more general mechanism in order to extend its relevance to adult stem/multipotent cell differentiation.

AUTHOR CONTRIBUTION

Hengxiang Cui and Kan Ding designed this project; Hengxiang Cui, Chenghao Shao, Qin Liu, Wenjie Yu, Jianping Fang, Weishi Yu and Amjad Ali performed experiments; Hengxiang Cui and Chenghao Shao analysed data; Weishi Yu carried out the BrdU analysis; Hengxiang Cui and Kan Ding wrote the paper.

FUNDING

The work was supported by the National Natural Science Foundation of China [grant number 30770484]; and the National Science and Technology Major Project “Key New Drug Creation and Manufacturing Program” [grant numbers 2009ZX09301-001, 2009ZX09501-011, 2009ZX09103-071].

Abbreviations: BM, basement membrane; BrdU, bromodeoxyuridine; CNS, central nervous system; DAPI, 4′,6-diamidino-2-phenylindole; DMEM, Dulbecco's modified Eagle's medium; ECM, extracellular matrix; EGR1, early growth response 1; EMSA, electrophoretic mobility-shift assay; ERK, extracellular-signal-regulated kinase; GAG, glycosaminoglycan; GAP43, growth-associated protein 43; GFP, green fluorescent protein; HEK, human embryonic kidney; HS, heparan sulfate; HSPG, HS proteoglycan; JNK, c-Jun N-terminal kinase; MAPK, mitogen-activated protein kinase; MEK, MAPK/ERK kinase; M-MLV, Moloney murine leukaemia virus; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide; NGF, nerve growth factor; RT, reverse transcription; SAPK, stress-activated protein kinase; shRNA, short hairpin RNA; VEGF, vascular endothelial growth factor

References

View Abstract