Research article

Cell surface heparan sulfates mediate internalization of the PWWP/HATH domain of HDGF via macropinocytosis to fine-tune cell signalling processes involved in fibroblast cell migration
Chia-Hui Wang, Fabian Davamani, Shih-Che Sue, Shao-Chen Lee, Po-long Wu, Fan-Mei Tang, Chiaho Shih, Tai-huang Huang, Wen-guey Wu
Biochemical Journal Dec 15, 2010, 433 (1) 127-138; DOI: 10.1042/BJ20100589

Abstract

HDGF (hepatoma-derived growth factor) stimulates cell proliferation by functioning on both sides of the plasma membrane as a ligand for membrane receptor binding to trigger cell signalling and as a stimulator for DNA synthesis in the nucleus. Although HDGF was initially identified as a secretory heparin-binding protein, the biological significance of its heparin-binding ability remains to be determined. In the present study we demonstrate that cells devoid of surface HS (heparan sulfate) were unable to internalize HDGF, HATH (N-terminal domain of HDGF consisting of amino acid residues 1–100, including the PWWP motif) and HATH(K96A) (single-site mutant form of HATH devoid of receptor binding activity), suggesting that the binding of HATH to surface HS is important for HDGF internalization. We further demonstrate that both HATH and HATH(K96A) could be internalized through macropinocytosis after binding to the cell surface HS. Interestingly, HS-mediated HATH(K96A) internalization is found to exhibit an inhibitory effect on cell migration and proliferation in contrast with that observed for HATH action on NIH 3T3 cells, suggesting that HDGF exploits the innate properties of both cell surface HS and membrane receptor via the HATH domain to affect related cell signalling processes. The present study indicates that MAPK (mitogen-activated protein kinase) signalling pathways could be affected by the HS-mediated HATH internalization to regulate cell migration in NIH 3T3 fibroblasts, as judged from the differential effect of HATH and HATH(K96A) treatment on the expression level of matrix metalloproteases.

  • cell migration
  • HATH domain
  • heparan sulfate binding
  • hepatoma-derived growth factor (HDGF)
  • macropinocytosis
  • proliferation

INTRODUCTION

Heparin-binding HDGF (hepatoma-derived growth factor) was first isolated as a secretory protein from the human hepatoma cell line HuH-7 [1]. It became the prototype for a new family of HRPs (HDGF-related proteins) that are mitogenic for various cell lines including hepatoma cells, fibroblasts, smooth mus-cle cells and endothelial cells [1–8]. All HRPs share a conserved N-terminal PWWP motif-containing the HATH [homologous to amino terminus of hHDGF (human HDGF)] domain (residues 1–100) [9–12]. Several studies have shown that the N- and C-terminal regions of hHDGF play distinct roles; the N-terminal region promotes the entry of HDGF into the cell, whereas the C-terminal region is capable of stimulating DNA synthesis and might be responsible for the regulation of its mitogenic activity [1,4,10,13]. In addition, the segment comprising residues 81–100 of HATH has been shown to interact with cell surface receptors and induce proliferative activity, whereas a single mutation at K96A abolishes this effect [14]. More recently, in addition to growth factor activity, the strongly conserved HATH domain in HRP-3 has been shown to bind to tubulin and promote neurite outgrowth during neuronal development [15].

HDGF has distinct functions on both sides of the plasma membrane. Inside the cells, intracellular HDGF can target the nucleus by its nuclear localization signals and facilitate DNA binding [4,16]. When secreted out of the cells, extracellular HDGF binds to cell surface receptors and modulates downstream signalling, thereby successfully integrating multiple linked functions in intracellular and extracellular environments [1,17–19]. However, the molecular mechanism of the internalization process is still unknown at present.

Transport of exogenous growth factors and/or their receptors to the cytosol, and eventually to the nucleus, plays important physiological and pathological roles in angiogenesis, wound healing and tumour progression [20–23]. However, growth factor binding to the membrane receptor alone is sufficient to induce activation of intracellular signalling pathways [24,25]. We have previously demonstrated that hHDGF is a modular protein consisting of a structured heparin-binding N-terminal HATH domain and a disordered C-terminal domain [26]. However, the function of the heparin-binding activity of HATH remains unclear. In the present paper, we report our findings after probing the role of the HATH domain itself in cell surface binding and subsequent internalization, and its effect on cellular functions. We demonstrate that cell surface HS (heparan sulfate) is essential for the internalization of the HATH domain of HDGF via macropinocytosis to fine-tune cell proliferation and migration.

MATERIALS AND METHODS

Materials

AF488 (Alexa Fluor® 488)-labelled transferrin, CTB (cholera toxin subunit B) and dextran were purchased from Molecular Probes. High-molecular-mass heparin (~18 kDa), enzymes, culture media and other chemicals were purchased from Sigma–Aldrich. Chondroitin sulfate A (whale cartilage, ~19 kDa), chondroitin sulfate C (shark cartilage, ~43 kDa), dermatan sulfate (pig skin, ~17 kDa), hyaluronic acid (pig skin, ~100 kDa) and keratan sulfate (bovine cornea, ~13 kDa) were provided by Dr Keiichi Yoshida, Seikagaku Co., Tokyo, Japan. HATH81–100 peptide was chemically synthesized by Fmoc (9-fluorenylmethyloxycarbonyl) solid-phase peptide synthesis (MD Bio, Taipei, Taiwan), the product was further purified using a reverse-phase column to reach >90% purity. Inhibitors U0126 and SB203580 were purchased from Calbiochem, and wortmannin, nystatin and filipin were from Sigma.

Construction of plasmids for DsRed-chimaeric proteins

The coding sequences for the HATH domain (Met1 to Tyr100) and C-terminal domain (Gln101 to the C-terminus; 101-C) were amplified by PCR using pGEM-T-hHDGF as a template and cloned into the pET6H vector as described previously and expressed with a histidine tag, M(H)6AMA, at its N-terminus [27]. To produce the DsRed-chimaeric proteins, the cDNA of DsRed was amplified by PCR using appropriate synthetic oligonucleotides and subcloned at the 5′-end of 101-C cDNA as pET6H-101-C for expresion of the DsRed–101-C protein, and at the 3′-end of HATH domain as pET6H-HATH for expression of the HATH–DsRed protein. The HATH(K96A) mutation was created by site-directed mutagenesis using pET6H-HATH plasmid as a template in the PCR reaction. The forward primer 5′-CCTGATCCATGGCGATGTCGCGATCCAACCGGCAG-3′ and a modified reverse primer with the mutated codon 5′-CGCGGATCCTTAATAGCCGGAAGCTGGTAGG-3′ were used. All of the clones were verified by sequencing. The polypeptides and chimaeric proteins used in the present study are depicted in Figure 1(A).

Figure 1
Figure 1 Constructs of HDGF fusion protein, its derivatives and their internalization properties

(A) Schematic representation of different constructs for expression of hHDGF and its truncated or mutated derivatives as fusion proteins with DsRed. (B) Fluorescence micrographs of different cell lines incubated with various fusion proteins (1 μM) for 24 h to examine the internalization of the DsRed-fusion proteins. Scale bar, 10 μm. (C) MFI of cells incubated with different concentrations of HATH–DsRed (NIH 3T3 for 6 h, CHO-K1 and pgsA-745 for 24 h). Internalization was measured as the change in MFI. (D) Kinetics of internalization of HATH–DsRed. Cells were incubated with 1 μM HATH–DsRed for 24 h and the change in MFI (normalized) was measured by flow cytometry (n=5; means±S.E.M.).

Expression and purification of HATH proteins

Escherichia coli strain BL21(DE3) harbouring one of the above constructs was grown in LB (Luria–Bertani) broth with 0.1 mg/ml ampicillin at 37 °C. The culture was grown to a D600 of 0.6 and induced with 1 mM IPTG (isopropyl β-D-thiogalactoside). For expression of HATH and HATH(K96A) fragments, the induction time was 4 h, and for expression of DsRed-chimaeric proteins, overnight induction was performed. After induction, the cells were harvested by centrifugation (6000 g), resuspended in lysis buffer (300 mM NaCl, 50 mM Na2HPO4, pH 8.0, and 10 mM imidazole), and lysed with a microfluidizer. The cell lysate was centrifuged at 15000 g for 20 min at 4 °C and the supernatant was loaded on to a HIS-select Ni2+-affinity column (Sigma), washed with 300 mM NaCl, 50 mM Na2HPO4, pH 8.0, and 20 mM imidazole buffer. The target protein was eluted with the same buffer with 250 mM imidazole. The proteins were further purified to homogeneity by passing through a Superdex 75 size-exclusion column in 10 mM phosphate buffer, pH 7.4, 150 mM NaCl and 1 mM EDTA connected to an ÄKTA FPLC system (GE Healthcare). The purity was checked by Coomassie Blue-stained PhastGel (homogenous 20, GE Healthcare) and found to be higher than 95%. The concentrations of HATH and HATH(K96A) proteins were estimated using the Bio-Rad Protein Assay Kit and the concentrations of DsRed-chimaeric proteins were estimated spectrophotometrically using a molar absorption coefficient of ϵ558=75000 M−1·cm−1.

Preparation of depolymerized heparin

Depolymerized heparin was prepared as described previously [28]. Low-molecular-mass porcine intestinal heparin (100 mg) was depolymerized with 0.4 IU heparinase I (EC 4.2.2.7) for 63 h at 37 °C in 30 mM HOAc-NaOH, pH 7.0, 3 mM Ca(OAc)2 and 0.1% BSA. Depolymerization was terminated by heating at 100 °C for 5 min and the mixture was then applied to a Bio-Gel P10 column (1.5 cm×150 cm) using 1 M NaCl and 10% (v/v) ethanol as eluent. The fractionated heparin fragments were desalted using Sephadex G-15 column. The depolymerized heparins were characterized by FAB (fast atom bombardment) MS, carbohydrate PAGE and FTIR (Fourier-transform infrared) spectroscopy. De-sulfated heparins were prepared as described previously [29].

Cell culture

NIH 3T3 cells were maintained in DMEM (Dulbecco's modified Eagle's medium). CHO (Chinese-hamster ovary)-K1 and pgsA-745 cells were maintained in McCoy's 5A medium. Both media were supplemented with 100 units/ml penicillin, 100 μg/ml streptomycin and 10% FBS (fetal bovine serum).

Cell surface modification

Cells were treated with the desired concentrations of heparinase I, heparinase III (EC 4.2.2.8) or chondroitinase ABC (EC 4.2.2.4) in HBSS (Hanks balanced salt solution) for 1 h and washed three times with HBSS to remove surface GAGs (glycosaminoglycans). For the internalization assay, additional heparinase I, heparinase III or chondroitinase ABC was added to the culture medium in the presence of HATH–DsRed to remove cell surface GAGs.

Internalization assay

Cells were seeded in 24-well plates at a density of 105 cells/well and cultured overnight. The desired amounts of DsRed-chimaeric protein were added to each well with or without pre-incubation at 4 °C followed by incubation at 37 °C for various lengths of time. The cells were washed twice with PBS and trypsinized with 0.25% trypsin to remove free DsRed-chimaeric protein. Internalized DsRed-chimaeric protein was monitored by flow cytometry.

HATH co-localization assay

Cells were seeded on to 12-mm-diameter glass coverslips at a density of 5×104 cells/cm2 and incubated with 1 μM DsRed–HATH or DsRed–HATH(K96A) for live cell imaging experiments. For tracking the internalization pathway of DsRed–HATH or DsRed–HATH(K96A), the markers for different internalization routes involving caveolin (AF488–CTB), clathrin (AF488–tranferrin) or macropinocytosis (AF488–dextran) were used in the co-localization assay. Cell images were captured with an Olympus FV-1000 confocal microscope. Macropinocytosis of HATH was tested with the inhibitors cytochalasin, wortmannin and amiloride, and their relative intensity was measured.

Proliferation assay

Cell proliferation was estimated by measuring cell number. Cells were plated in 12-well plates at a density of 1×104 cells and grown in DMEM supplemented with 2% FBS and 100 nM HATH, HATH(K96A) or HATH(81–100). At the indicated times, cells were trypsinized and counted using a Bright Line Counting Chamber (Hausser Scientific). In addition, cell proliferation was also determined by measuring the incorporation of [3H]thymidine into proliferating cells. Cells were plated in 96-well plates at a density of 5×103 cells/well supplemented with 10% FBS. After 3 h, medium was replaced with DMEM containing 2% FBS with the desired amount of HATH and HATH(K96A). After an additional 56 h, each well received 2.5 μCi of [3H]thymidine and cells were incubated for an additional 16 h. Cells were typsinized and harvested. Radioactivity was measured in a TopCount NXT Microplate Scintillation and Luminescence Counter. Serum-free and 1% serum supplemented DMEM could not support proliferation of NIH 3T3 fibroblasts.

Cell migration

Boyden chambers (8 μm pore size; 24 well insert; BD Biosciences) were used for the assay. The membranes were coated on the lower side with 10 μg of fibronectin and blocked with 1% BSA. NIH 3T3 cells (20000 cells/well) were incubated with the desired proteins in serum-free medium in the upper chamber. The cells were allowed to migrate to the lower chamber containing medium with serum for 5 h and stained with 0.1% Coomassie Blue and counted. During the use of the signalling inhibitors U0126 and SB203580, cells were pretreated with the inhibitors for 30 min and allowed to migrate in the presence of the inhibitors. For the macropinocytosis inhibitor amiloride, the cells were pretreated with amiloride for 1 h, the inhibitor was removed and the cells were allowed to migrate.

RT (reverse transcription)–PCR

Total RNA was extracted from NIH 3T3 cells treated with HATH, HATH(K96A) or HATH81–100 using the RNeasy mini kit (Qiagen). The RT–PCR reaction was performed using the One-Step RT–PCR Premix Kit from Intron (Gyeonggi-do, Korea) using the following primers: MMP (matrix metalloproteinase)-9F, 5′-GTCGTGATCCCCACTTACT-3′; MMP-9R, 5′-AACACACAGGGTTTGCCTTC-3′; MMP-2F, 5′-GTCGCCCCTAAAACAGACAA-3′; MMP-2R, 5′-GGTCTCGAGGTGTTCTGGT-3′; TIMP (tissue inhibitor of MMP)-1F, 5′-ATTCAAGGCTGTGGGAAATG-3′; TIMP-1R, 5′-CTCAGAGTACGCCAGGGAAC-3′; GADPH (glyceraldehyde-3-phosphate dehydrogenase) F, 5′-TGATGACATCAAGAAGGTGGTGAAG-3′; and GAPDH R, 5′-TCCTTGGAGGCCATGTAGGCCAT-3′ synthesized at ScinoPharm (Tainan, Taiwan). RT–PCR conditions were as follows: RT at 45 °C for 30 min; denaturation at 94 °C for 5 min followed by 35 cycles of 94 °C for 30 s, 61 °C for 30 s (MMP)/59.5 °C for 30 s (TIMP-1)/58 °C for 30 s (GAPDH), 72 °C for 30 s; and final extension at 72 °C for 7 min.

Western blot analysis

Cells were stimulated for 5 min, 15 min, 30 min or 1 h with the respective proteins with/without the inhibitors and lysed with lysis buffer [20 mM Tris/HCl, pH 7.5, 150 mM NaCl, 1mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM β-glycerol phosphate, a mixture of protease inhibitors (Roche, Basel Switzerland) and 0.2 mM sodium orthovanadate] and analysed by SDS/PAGE followed by Western blotting with antibodies against phospho-p38 MAPK (mitogen-activated protein kinase), p38 MAPK, phospho-p44/42 ERK (extracellular-signal-regulated kinase), p44/42 ERK, phospho-Ser473 Akt, phospho-Thr308 Akt, Akt and β-actin. The Western blots were developed using ECL (Pierce).

RESULTS AND DISCUSSION

GAG-mediated HDGF internalization via the binding of the N-terminal HATH domain

Cell surface GAGs are known to play significant roles in cellular entry of various growth factors and proteins, thereby governing their physiological functions [30–32]. To study the role of various domains of HDGF in cellular internalization, different constructs of HGDF, including HATH, HATH(K96A) and the C-terminal portion of HDGF devoid of the HATH domain, 101-C, were tagged with DsRed at the N- or C-terminus and expressed as fusion proteins (Figure 1A). The K96A mutant of HATH was used because it has been demonstrated that substitution of this single amino acid is sufficient to abolish both binding to the cell surface receptor and proliferative activity of the HATH domain [14].

We studied the internalization of exogenously added fusion proteins using different cell lines, NIH 3T3 and CHO-K1, both expressing cell surface GAGs, and CHO pgsA-745, which is a GAG-deficient cell line. After 24 h incubation, both HATH and HDGF were found to be internalized in comparable amounts in both NIH 3T3 and CHO-K1 cells, whereas 101-C failed to internalize, as observed by fluorescent microscopy (Figure 1B). This suggests that the HATH domain alone is necessary and sufficient for cellular internalization of HDGF and its derivatives. In contrast, none of the fusion proteins could be internalized in the CHO pgsA-745 cell line that lacks the ability to synthesize cell surface HSPGs (HS proteoglycans) [33], signifying that GAGs play a crucial, deterministic role in the uptake of HDGF and HATH proteins.

Dose- and time-dependent internalization of HATH–DsRed in the three cell lines was also analysed under physiological conditions by flow cytometry (Figures 1C and 1D). Although NIH 3T3 and CHO-K1 cells showed significant internalization of HATH–DsRed in a dose- and time-dependent manner, CHO pgsA-745 cells failed to show any internalization. A difference in the total amount of internalized proteins (Figure 1C) and the internalization kinetics (Figure 1D) in CHO-K1 cells and NIH 3T3 cells were observed. Whereas HATH–DsRed internalization was barely detectable in CHO-K1 cells within the first 2 h, there was a significant internalization with ~50% of the saturation level in NIH 3T3 cells. This could be due to enhanced expression of GAGs in NIH 3T3 compared with CHO-K1 cells [34] and/or the differential availability of other cell surface receptors for HATH. NIH 3T3 fibroblasts have been reported to express the receptor that produces HATH-stimulated mitogenic activity [14].

GAG-binding specificity of the HATH domain

GAGs on the cell surface are critical environmental modulators and play important roles in cell differentiation and tissue morphogenesis [35]. In order to identify the specific GAGs involved in HATH binding, heparin and other GAGs, i.e. dermatan sulfate, chondroitin sulfate A, chondroitin sulfate C, keratin sulfate and hyaluronic acid, were added to the growth medium to compete with the cell surface GAGs for binding with HATH–DsRed. Only heparin was found to be capable of blocking the internalization of HATH–DsRed in both NIH 3T3 (Figure 2A) and CHO-K1 cells (Figure 2B), suggesting that HATH binds specifically to a cell surface heparin-like moiety such as HS.

Figure 2
Figure 2 GAG specificity and internalization of HATH–DsRed in NIH 3T3 and CHO-K1 cells

NIH 3T3 (A) or CHO-K1 (B) cells were co-incubated with 1 μM HATH–DsRed and different concentrations of exogenous heparin, dermatan sulfate (DS), keratin sulfate (KS), hyaluronic acid (HA), chondroitin sulfate A (CsA) or chondroitin sulfate C (CsC) and internalization was quantified by flow cytometry, expressed as MFI. Concentrations are shown in log scales in the upper panels in (A) and (B) (n=7, means±S.E.M.). (C and D) Effect of enzymatic treatments on cell internalization. NIH 3T3 (C) or CHO-K1 (D) cells were pretreated with heparinase I, heparinase III or chondroitinase for 1 h followed by incubation with 1 μM HATH–DsRed in the presence of the enzyme (6 h for NIH 3T3 and 24 h for CHO-K1). Internalized amounts of HATH–DsRed were measured by flow cytometry and the profiles were plotted as a function of enzyme concentration (n=7, means±S.E.M.).

GAG specificity for HATH–DsRed internalization was further investigated by exogenous treatment of NIH 3T3 and CHO-K1 cells with heparinase I, heparinase III or chondroitinase ABC to remove specific surface GAGs and to monitor HATH binding on the cell surface. Modification of cell surface HS by heparinase I and heparinase III abrogated the ability of the cells to internalize HATH–DsRed, whereas there was no detectable change in HATH internalization following chondroitinase treatment (Figures 2C and 2D). We conclude that the internalization of HATH–DsRed in both cell types is predominantly mediated through HS.

HS chain length and sulfation pattern for HATH binding

To investigate the role of HS chain length in the internalization of the HATH domain, exogenous heparin-derived mimetics with defined chain lengths were allowed to compete for HATH binding. In NIH 3T3 and CHO-K1 cells, exogenous heparin, especially long-chain heparin at high concentrations, was found to be effective in blocking HATH–DsRed internalization (Figure 3A). At heparin concentrations of 500 μg/ml, heparin-derived mimetics with a chain length of 12 carbohydrate moieties were capable of completely blocking HATH–DsRed internalization. This is found to be approximately twice the length of the heparin-binding site in the monomeric HATH domain. We have previously demonstrated that the HATH domain is capable of forming a domain-swapped dimer with two heparin-binding sites lining up to accommodate a heparin molecule with 12 carbohydrate moieties [26]. Although experimental evidence correlating PWWP/HATH domain dimerization and the biological function of hHDGF is still lacking, the present result on the chain length specificity suggests that dimerization of the PWWP/HATH domain may play a role in the internalization process. This observation is in line with the biochemical and physiological observations to support a direct role for HS in FGF (fibroblast growth factor) dimerization, and in the formation of an active FGF/FGF receptor signalling complex [36].

Figure 3
Figure 3 Heparan sulfate chain length specificity and effect of heparin-derived mimetics on binding and uptake of HATH—DsRed

(A) Heparan sulfate chain length-dependent internalization of the HATH domain, monitored by flow cytometry and expressed as MFI (n=3, means±S.E.M.). NIH 3T3 cells were incubated with HATH–DsRed at 1 μM concentration, along with soluble heparins of different chain lengths. (B) Effect of desulfated heparins [deN,OS, deN-sulfate (deNS), de6-O sulfate (de6OS) and de2-O sulfate (de2OS)] on the internalization of HATH domain, monitored by flow cytometry and expressed as MFI (n=3, means±S.E.M.). NIH 3T3 cells were incubated with HATH–DsRed at 1 μM concentration, along with desulfated heparins. (C) SPR sensograms of HATH binding to SPR chips coated with native or desulfated heparin.

We have observed an apparent inconsistency in the effective dose for blocking the HS-mediated internalization by heparin-derived mimetics and intact heparin. For instance, approx. 300 μg/ml of long-chain heparin-derived mimetics are required to compete for the internalization process; however, only 10 μg/ml is needed for intact heparin (Figure 3A). One possible explanation is that HS–HATH recognition may be related to its sulfation pattern. It has been reported that enzymatic digestion specifically targets the highly sulfated S domain for heparinase III and the low sulfated A domain by heparinase I treatment. Since the heparin-derived mimetics were prepared by enzymatic digestion, such treatment could alter the degree of sulfation of HS, thereby affecting its interaction with HATH. In addition, the SAS (sulfated-acetylated-sulfated) sequence optimal for the protein–HS interaction could be altered in the heparin-derived mimetics [36]. All of these factors could account for the high concentrations of mimetics required for blocking HATH internalization under our experimental conditions.

To investigate further the role of sulfation at specific positions of heparin, de-sulfated mimetics, specifically at N-sulfate, 6-O sulfate and 2-O sulfate positions were prepared and checked for heparin-binding specificity by flow cytometry (Figure 3B) and SPR (surface plasmon resonance) (Figure 3C). Although heparin-derived mimetics with de2-O sulfate and de6-O sulfate behave similarly with the native heparin, the de-N-sulfated heparin mimetics could not compete with heparin for its binding to HATH, suggesting that N-sulfation on heparin is crucial for HATH binding.

Role of receptor binding and pre-incubation in internalization kinetics

We have demonstrated so far that HATH–DsRed can be internalized in a cell-specific manner, depending on the availability of a specific HS structure. However, it has been known for some time that cellular entry of growth factors involves initial binding either to a high affinity tyrosine kinase receptor on the cell surface and/or to low affinity, but high capacity, GAGs [31,37,38]. Indeed, the difference in the internalization kinetics of HATH–DsRed between CHO-K1 and NIH 3T3 cells (Figure 1D) could be interpreted as being due to the difference in their cell surface receptor. Since the binding of cell surface receptor by HATH has been demonstrated to involve the specific amino acids 81–100 and that the single site mutation of K96A can abolish this activity, we carried out a competition experiment using HATH81–100 and HATH(K96A) to test the possible involvement of receptor binding in the internalization kinetics. Characterization of the HS-binding activity of HATH(K96A) by SPR has revealed a similar HS-binding strength between HATH(K96A) and HATH; however, the HATH81–100 peptide failed to bind HS (see Supplementary Figure S1 at http://www.BiochemJ.org/bj/433/bj4330127add.htm); thereby, it allows us to dissect the relative contribution of HS and receptor binding in triggering HATH internalization.

As shown in Figure 4(A), the competition experiment was monitored by flow cytometry as a function of time, HATH–DsRed was allowed to compete with HATH, HATH(K96A) and HATH81–100 peptide for internalization in NIH 3T3 cells. HATH, HATH(K96A) and HATH81–100 were all found to compete favourably with HATH–DsRed; however, their patterns of internalization were different. At the saturation level observed at 24 h, the relative strength of competition follows the order of HATH, HATH(K96A) and HATH81–100. This is consistent with the notion that the saturation level of HATH–DsRed internalization is mainly mediated by its HS-binding property. Although HATH could compete effectively with HATH–DsRed, under the same conditions HATH–DsRed could still undergo detectable internalization with a MFI (mean fluorescence intensity) value of ~100 in the presence of a similar amount of HATH(K96A). The slightly weaker competition for HATH(K96A) as compared with HATH can be understood as DsRed-labelled HATH(K96A) internalizes approx. 2-fold slower than DsRed-labelled HATH (Figure 4B).

Figure 4
Figure 4 Kinetics of heparan sulfate or receptor-mediated internalization of HATH—DsRed

(A) NIH 3T3 cells were incubated with 1 μM DsRed–HATH competing with HATH, HATH(K96A) or HATH81–100 for 24 h and the rate of internalization was monitored by flow cytometry (n=5, means±S.E.M.). (B) The kinetics of DsRed–HATH and DsRed–HATH(K96A) internalized under physiological conditions. (C) Effect of pre-binding on internalization. Cells were incubated with 1 μM DsRed–HATH or DsRed–HATH(K96A), without pre-binding or after pre-binding at 4 °C for 30 min. The effect of physiological conditions and prebinding effect on the internalization kinetics was monitored by flow cytometry and expressed as MFI. w/, with; w/o, without.

In the case of HATH81–100 peptide competition within the first 6 h, the receptor binding HATH81–100 peptide was found to be more effective than HS-binding HATH(K96A) and apparently as effective as the intrinsic HATH. However, in the time period of 6–12 h, HATH81–100 peptide appeared to lose its competition effect such that a significant amount of HATH–DsRed was detected, reaching a saturation level at 24 h. This is not due to the possible degradation of the HATH81–100 peptide during the course of experiment, since the result remained the same when additional amount of HATH81–100 peptide was added during the course of the competition experiment (results not shown). Although the HATH81–100 peptide has been proven to be able to bind to the cell surface receptor [14], HATH81–100–DsRed failed to internalize in NIH 3T3 cells (see Supplementary Figure S2 at http://www.BiochemJ.org/bj/433/bj4330127add.htm). This demonstrates that receptor binding without HS binding is not sufficient for internalization. Therefore we suggest that receptor binding plays an initiating role in facilitating HS-mediated HATH–DsRed internalization. This finding corroborates our observation that HATH is more efficiently internalized compared with HATH(K96A), despite the fact that both of the proteins bind to HS with a similar affinity. Moreover, the apparent delay in internalization of HATH–DsRed in CHO-K1 cells raises the possibility that CHO-K1 may be devoid of the HDGF receptor (Figure 1D).

The exact mechanism for the role of receptor binding in initiating HS-mediated internalization of HATH is not clear at the present time. A similar question arises regarding the relatively slow internalization rate of HATH under physiological conditions (Figure 1). We therefore investigated the internalization process of DsRed-labelled HATH and HATH(K96A) under pre-binding conditions, a strategy usually adopted in the investigation of the endocytic pathway. As shown in Figure 4(C), we found that when HATH and HATH(K96A) were pre-incubated at 4 °C to allow pre-binding to the cell surface HS and/or receptor, a large amount of internalization was observed for both of the proteins within a short time span of 30 min after the endocytic pathways were turned on by switching the temperature to 37 °C. In comparison, internalization of a similar amount without pre-binding was only observed at 6 h for HATH and at 24 h for HATH(K96A). The dramatic effect of pre-binding condition on enhancement of the internalization rates suggests that HS-mediated internalization of HATH and HATH(K96A) might require significant clustering of GAG-containing HSPG, such as is the case for HS-mediated macropinocytosis. The pre-incubation condition or the availability of membrane receptor in the cell lines studied may simply promote the clustering process via multivalent binding of the HATH domain.

Endocytic pathways of the HATH domain

We first investigated the endocytic pathway of the HATH domain by using confocal microscopy to examine the co-localization of DsRed–HATH with other fluorescently labelled cargos for the macropinocytosis pathway and for clathrin- and caveolin-mediated endocytosis (Figure 5A). Internalization experiments involving live NIH 3T3 cells demonstrated that both DsRed–HATH and DsRed–HATH(K96A) co-localize with dextran, indicating that both proteins could be internalized through macropinocytosis by involving bulk fluid uptake. However, the receptor and HS-binding HATH domain can be internalized via the caveolin-mediated pathway as well, similar to other functional growth factors [39,40]. Neither DsRed–HATH nor DsRed–HATH(K96A) co-localized with transferrin. The internalization of HATH and HATH(K96A) appeared as aggregated patches, not scattered, suggestive of complex formation during internalization [41]. HATH, therefore, appears to employ dual mechanisms of internalization, the high-capacity macropinocytosis and the high-specificity caveolin-dependent endocytosis. HATH(K96A), on the other hand, can be internalized only through HS-mediated macropinocytosis, as also shown by the quantification of the degree of co-localization of the various proteins with their pathway markers (Figure 5B).

Figure 5
Figure 5 Routes of internalization of DsRed–HATH and DsRed–HATH(K96A)

(A) Confocal live cell images of internalized DsRed constructs (1 μM) of HATH or HATH(K96A). HATH or HATH(K96A) was allowed to internalize with the co-localization markers CTB, dextran (DEX) or transferrin (TFN) to determine the internalization route. Scale bar, 10 μm. (B) Comparison of percentage of co-localization of DsRed–HATH and DsRed–HATH(K96A) as determined by confocal microscopy. (C) Effect of macropinocytosis inhibitors (cytochalasin, wortmannin and amiloride) on the internalization of HATH to study the internalization route of HATH domain. (D) The effect of HATH81–100 peptide on the internalization of DsRed–HATH and DsRed–HATH(K96A) as determined by confocal microscopy.

Various inhibitors for macropinocytosis were also investigated to further illustrate the HATH internalization route (Figure 5C). These include the inhibitor for actin polymerization (cytochalasin), the inhibitor for fluid-phase pinocytosis and PI3K (phosphoinositide 3-kinase) (wortmannin) and an ion channel blocker (amiloride). Since the macropinocytosis inhibitor amiloride could almost block all of the internalization of HATH and no significant effect can be detected by dynosore, a dynamin inhibitor (results not shown), the results also suggest macropinocytosis as a major internalization pathway for HATH. Finally, when DsRed–HATH and DsRed–HATH(K96A) were added with an equal ratio of HATH81–100, the latter could compete only with HATH, but not with HATH(K96A), suggesting dual targets for HATH internalization and one route via HS-mediated macropinocytosis for HATH(K96A) (Figure 5D).

Inhibitory effect of HATH(K96A) on cell migration and proliferation

Both HATH and the HATH81–100 peptide have previously been reported to bind to cell surface receptors and promote proliferation. It has also been shown that the K96A mutant form of HATH81–100 does not bind receptors and is not able to enhance cell proliferation [14]. Since HATH(K96A) can bind to cell surface HS, it would be interesting to see whether HATH(K96A) behaves in a similar manner to the K96A mutant form of HATH81–100. NIH 3T3 cells showed different responses when treated with HATH, HATH(K96A) or HATH81–100. Using cell count as a measure, HATH and HATH81–100 were able to stimulate cell proliferation, whereas HATH(K96A) was not (Supplementary Figure S3 at http://www.BiochemJ.org/bj/433/bj4330127add.htm). We then used a [3H]thymidine incorporation assay to confirm the effect. At a low serum concentration, HATH promoted cell proliferation by 30% and 50% at concentrations of 100 nM and 1 μM respectively (Figure 6A). However, HATH(K96A) was unable to stimulate cell proliferation. The results indicate that the HS-binding ability of HATH(K96A) is insufficient to stimulate cell proliferation and that receptor binding is required.

Figure 6
Figure 6 The effect of HATH and HATH(K96A) on cell migration and proliferation

(A) Proliferation of NIH 3T3 cells in response to 100 nM or 1 μM HATH or HATH(K96A) stimulation for 72 h. Proliferation was determined by measuring [3H]thymidine incorporation. Medium containing 2% and 10% FBS were used as negative and positive controls respectively. (B) NIH 3T3 cells were stimulated with 100 nM HATH, HATH(K96A) or HATH81–100 and were allowed to migrate across the fibronectin-coated transwell membrane of a Boyden chamber for 5 h at 37 °C. The migrated cells were stained and photographed. The untreated cells and uncoated negative controls were processed as fibronectin-coated (FN+) and fibronectin uncoated (FN-). (C) Representation of the percentage of migrated cells in response to 100 nM HATH, HATH81–100 or HATH(K96A).

HDGF has been known to be an angiogenesis factor that supports cell migration and proliferation [5,8,42]. We therefore tested the effect of HATH, HATH(K96A) and HATH81–100 on cell migration. Cell migration assays using transwell chambers were carried out within 5 h. As shown in Figures 6(B) and 6(C), although HATH and HATH81–100 showed a marginal effect in promoting cell migration, HATH(K96A) clearly inhibited cell migration significantly as compared with the the control. We attribute it to HS-mediated internalization of HATH(K96A), since HATH(K96A) does not induce downstream signalling through receptor binding. In fact, endocytosis of extracellular/soluble growth factors has been known to play a key role in regulation of cell migration and intracellular signalling cascades [43]. Despite lacking the C-terminal domain of hHDGF, both HATH and HATH81–100 have been shown to bind to cell surface receptors to stimulate mitogenic activity. The inhibitory effect of HATH(K96A) on both cell migration and proliferation suggests that HS-mediated macropinocytosis of HATH(K96A) plays an additional regulatory role on the downstream cell signalling induced by the HATH domain.

Modulation of cell signalling by HATH internalization

MAPK pathways have been reported to play a role in cell migration and proliferation [44]. For instance, HDGF is known to induce phosphorylation of p44/42 ERK in NIH 3T3 cell proliferation [45,46]. Figure 7(A) shows that HATH was able to stimulate phosphorylation of p44/42 ERK, peaking at 5 min, after which the effect declined. Surprisingly, phosphorylation of p44/42 ERK was also stimulated by HATH(K96A) and HATH81–100, with the latter having an even stronger effect. The results demonstrate that HATH domain binding to either HS or receptor is able to stimulate the p44/42 ERK signalling pathway, but the effect due to receptor binding is prominent. Interestingly, stimulation of phosphorylation of p38 MAPK becomes apparent only much later (>1 h) (Figures 7B and 7C). The phosphorylation of FAK (focal adhesion kinase), involving the PI3K/Akt pathway, is known to support cell migration in NIH 3T3 fibroblasts [47]. However, neither of the phosphorylation sites in Akt was affected by incubation with HATH (Figure 7B). Thus Akt appears not to be involved in HATH-stimulated cell migration.

Figure 7
Figure 7 HATH domain of HDGF affects signalling through p38 MAPK and ERK 1/2 (p44/42) pathways

(A and B) NIH 3T3 cells were grown in DMEM with 10% FBS. Cells were serum starved in 0.5% FBS for 6 h prior to addition of HATH or HATH(K96A). After 0, 5, 15, 30 (A) or 60 (B) min, cell lysates were prepared, separated by SDS/PAGE and immunoblotted with antibodies against p38 and phospho-p38 MAPK, ERK and phospho-ERK, and Akt and phospho-Akt. (C) Expression levels of p38 MAPK, phospho-p38 MAPK, ERK and phospho-ERK intensities in three independent experiments in the presence of HATH, HATH(K96A) or HATH81–100. *P≥0.05 and **P≥0.01. Results show the means±S.E.M (n=3). (D) The effect of the macropinocytosis inhibitor amiloride on phosphorylation of p38 MAPK stimulated by HATH and HATH(K96A). (E) Graphical representation of the phospho-p38 MAPK intensity affected by amiloride (**P≥0.01). p, phospho.

We also examined the effect of the macropinocytosis inhibitor amiloride [48,49] on the phosphorylation level of p38 MAPK to understand the role of HS-mediated internalization of HATH in fine-tuning cell signalling processes (Figures 7D and 7E). The inhibitors for macropinocytosis affected HATH- and HATH(K96A)-induced p38 MAPK phosphorylation with opposite effect. Amiloride reduced phosphorylation of p38 MAPK (Figure 7D), whereas the caveolin inhibitors filipin and nystatin [48,49] had no effect on HATH(K96A) and only a partial effect on HATH (results not shown). Although quantitative understanding of the observed effect may be complicated, the increase in phosphorylation by HATH after amiloride treatment is suggestive that HATH binding to the cell surface receptor alone might trigger signalling when macropinocytosis is blocked. Nevertheless, it is clear that the cell signalling of HATH and HATH(K96A) are differentially affected when HS-mediated macropinocytosis is perturbed by amiloride treatment.

HATH migration involves MMP-9 expression

MMP-2, MMP-9 and TIMPs regulate tumour progression [50]. Our preliminary data from the microarray analysis of genes induced by HATH and HATH(K96A) in NIH 3T3 cells revealed differences in the expression of MMP-9 and TIMP-1 (results not shown). This was confirmed by semi-quantitative RT–PCR analysis (Figures 8A and 8B), which showed that although HATH treatment increased the expression of MMP-9 significantly, HATH(K96A) only induced a marginal increase in the expression of MMP-9. In contrast, the expression level of TIMP-1 was more significantly enhanced by HATH(K96A) when compared with HATH. MMP-2 levels remained unchanged (Figures 8A and 8B). These results are consistent with the cell migration ability of HATH and HATH(K96A), i.e. HATH promoted cell migration whereas HATH(K96A) inhibited it.

Figure 8
Figure 8 Regulation of MMP expression by HATH

(A) Cells were stimulated with HATH or HATH(K96A) for 3 h and total RNA was prepared. Expression levels of MMP-2, MMP-9 and TIMP-1 were determined by semi-quantitative RT–PCR using titrating amounts of RNA template (250 ng of RNA). GAPDH was used as an internal control. Lane 1, control; lane 2, HATH; lane 3, HATH(K96A). Unstimulated cells served as a control. (B) Relative mRNA expression levels of MMP-2, MMP-9 and TIMP-1. Results are means±S.E.M. (n=3) *P≥0.05. (C) The protein levels of secreted MMP-2, MMP-9 and TIMP-1 proteins after 48 h stimulation were analysed by Western blot. Samples were concentrated from the culture media. (D) The quantification of the result of Western blot shown in (C). All experiments were performed in serum-free medium. Lane 1, untreated control; lane 2, with HATH; and lane 3, with HATH(K96A) for (A) and (C). Results are means±S.E.M. are represented (n=3) **P≥0.01.

In NIH 3T3 fibroblasts, suppression of invasion is reported to be due to a decrease in MMP-9, but not MMP-2, levels [51,52]. Consistent with this observation, we found that only the MMP-9 protein level increased in the presence of HATH, but both TIMP-1 and MMP-9 protein levels increased in the presence of HATH(K96A) (Figures 8C and 8D). It is to be noted that MMPs and TIMPs are secretory proteins secreted out of the cells to modulate the extracellular matrix affecting migration. The secreted protein levels of MMP-9 are well above the untreated controls in both HATH and HATH(K96A); however, TIMP-1 levels were enhanced only by HATH(K96A), leading to inhibition of MMP-9 activity thereby inhibiting cell migration in HATH(K96A) treatment. Elevated expression of MMP-9 is associated with increased metastatic potential in many cancer types, including breast cancer, prostate cancer, brain cancer and melanoma cells. An increase in MMP-9 is indicative of tumour progression [32,53]. Interestingly, the mutant HATH(K96A) could exhibit anti-proliferative and anti-invasive ability, countering the effect of HATH. The HATH(K96A) mutant showed decreased MAPK phosphorylation and a decreased MMP/TIMP ratio to function as a suppressor of cell invasion.

In summary, in addition to the proliferative activity of HDGF as a transcription factor by binding to DNA, we show that the HATH/PWWP domain of HDGF can bind to cell surface HS for its internalization (Figure 9). Although HATH is known to bind to cell surface receptors to induce cell signalling for its cell proliferation and migration activity, the HS-mediated internalization of HATH can further fine-tune the related cell signalling processes and also its routes of internalization. Thus the HS-binding activity of HATH functions as an additional route of regulation to the two existing pathways for the proliferative activity of HDGF via either cell-surface-receptor binding or nuclear localization. The heparin-binding activity of the HATH domain alone could then fine-tune the mobilization of HDGF in the extracellular matrix and also its internalization pathway for both signal modulation and nuclear localization.

Figure 9
Figure 9 Schematic representation of the internalization mechanisms of HATH and HATH(K96A) and a possible signalling pathway for the functional role of HDGF

(A) HATH/HATH(K96A) binds to cell surface receptors/HS and internalizes through macropinocytosis- and/or caveolin-mediated mechanisms, thus exerting differential effects on the p38 MAPK pathway and cellular functions. (B) Scheme of unconventional routes for exogenous HATH(K96A)/HATH cell surface binding and internalization.

AUTHOR CONTRIBUTION

Chia-Hui Wang performed and analysed the experiments. Fabian Davamani performed and analysed experiments and contributed to manuscript writing. Shih-Che Sue designed experiments. Shao-Chen Lee performed SPR experiments and prepared heparin derivatives. Po-long Wu performed the Western blot experiments and Fan-Mei Tang performed the cell proliferation assay. Chiaho Shih was involved in data discussion. Tai-huang Huang and Wen-guey Wu supervised the research and wrote the manuscript.

FUNDING

This work was supported partially by an NSC grant to both T.-h.H. and W.-g.W. Preliminary characterization of the protein was conducted in the High-field NMR Center supported by the National Research Program for Genomic Medicine.

Acknowledgments

We thank Hong-Lin Chan for technical assistance with Western blot experiments. The confocal microscopy was performed at the Nano-Imaging Laboratory, National Synchrotron Radiation Research Centre (NSRRC), Taiwan.

Abbreviations: 101-C, peptide encoding Gln101 to the C-terminus of hepatoma-derived growth factor; AF488, Alexa Fluor® 488; CHO, Chinese-hamster ovary; CTB, cholera toxin subunit B; DMEM, Dulbecco's modified Eagle's medium; ERK, extracellular-signal-regulated kinase; FBS, fetal bovine serum; FGF, fibroblast growth factor; GAG, glycosaminoglycan; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; HATH, homologous to amino terminus of human hepatoma-derived growth factor; HDGF, hepatoma-derived growth factor; hHDGF, human HDGF; HRP, HDGF-related protein; HS, heparan sulfate; HSPG, HS proteoglycan; MAPK, mitogen-activated protein kinase; MFI, mean fluorescence intensity; MMP, matrix metalloproteinase; PI3K, phosphoinositide 3-kinase; RT, reverse transcription; SPR, surface plasmon resonance; TIMP, tissue inhibitor of MMP

References

  1. ↵
    1. Nakamura H.,
    2. Izumoto Y.,
    3. Kambe H.,
    4. Kuroda T.,
    5. Mori T.,
    6. Kawamura K.,
    7. Yamamoto H.,
    8. Kishimoto T.
    (1994) Molecular cloning of complementary DNA for a novel human hepatoma-derived growth factor. Its homology with high mobility group-1 protein. J. Biol. Chem. 269:25143–25149.
    OpenUrlAbstract/FREE Full Text
    1. Abouzied M. M.,
    2. Baader S. L.,
    3. Dietz F.,
    4. Kappler J.,
    5. Gieselmann V.,
    6. Franken S.
    (2004) Expression patterns and different subcellular localization of the growth factors HDGF (hepatoma-derived growth factor) and HRP-3 (HDGF-related protein-3) suggest functions in addition to their mitogenic activity. Biochem. J. 378:169–176.
    OpenUrlAbstract/FREE Full Text
    1. Dietz F.,
    2. Franken S.,
    3. Yoshida K.,
    4. Nakamura H.,
    5. Kappler J.,
    6. Gieselmann V.
    (2002) The family of hepatoma-derived growth factor proteins: characterization of a new member HRP-4 and classification of its subfamilies. Biochem. J. 366:491–500.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Kishima Y.,
    2. Yamamoto H.,
    3. Izumoto Y.,
    4. Yoshida K.,
    5. Enomoto H.,
    6. Yamamoto M.,
    7. Kuroda T.,
    8. Ito H.,
    9. Yoshizaki K.,
    10. Nakamura H.
    (2002) Hepatoma-derived growth factor stimulates cell growth after translocation to the nucleus by nuclear localization signals. J. Biol. Chem. 277:10315–10322.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Okuda Y.,
    2. Nakamura H.,
    3. Yoshida K.,
    4. Enomoto H.,
    5. Uyama H.,
    6. Hirotani T.,
    7. Funamoto M.,
    8. Ito H.,
    9. Everett A. D.,
    10. Hada T.,
    11. Kawase I.
    (2003) Hepatoma-derived growth factor induces tumorigenesis in vivo through both direct angiogenic activity and induction of vascular endothelial growth factor. Cancer Sci. 94:1034–1041.
    OpenUrlCrossRefPubMed
    1. Everett A. D.,
    2. Lobe D. R.,
    3. Matsumura M. E.,
    4. Nakamura H.,
    5. McNamara C. A.
    (2000) Hepatoma-derived growth factor stimulates smooth muscle cell growth and is expressed in vascular development. J. Clin. Invest. 105:567–575.
    OpenUrlCrossRefPubMedWeb of Science
    1. Hu T. H.,
    2. Huang C. C.,
    3. Liu L. F.,
    4. Lin P. R.,
    5. Liu S. Y.,
    6. Chang H. W.,
    7. Changchien C. S.,
    8. Lee C. M.,
    9. Chuang J. H.,
    10. Tai M. H.
    (2003) Expression of hepatoma-derived growth factor in hepatocellular carcinoma. Cancer 98:1444–1456.
    OpenUrlCrossRefPubMedWeb of Science
  8. ↵
    1. Lepourcelet M.,
    2. Tou L.,
    3. Cai L.,
    4. Sawada J.,
    5. Lazar A. J.,
    6. Glickman J. N.,
    7. Williamson J. A.,
    8. Everett A. D.,
    9. Redston M.,
    10. Fox E. A.,
    (2005) Insights into developmental mechanisms and cancers in the mammalian intestine derived from serial analysis of gene expression and study of the hepatoma-derived growth factor (HDGF). Development 132:415–427.
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Izumoto Y.,
    2. Kuroda T.,
    3. Harada H.,
    4. Kishimoto T.,
    5. Nakamura H.
    (1997) Hepatoma-derived growth factor belongs to a gene family in mice showing significant homology in the amino terminus. Biochem. Biophys. Res. Commun. 238:26–32.
    OpenUrlCrossRefPubMedWeb of Science
  10. ↵
    1. Nameki N.,
    2. Tochio N.,
    3. Koshiba S.,
    4. Inoue M.,
    5. Yabuki T.,
    6. Aoki M.,
    7. Seki E.,
    8. Matsuda T.,
    9. Fujikura Y.,
    10. Saito M.,
    (2005) Solution structure of the PWWP domain of the hepatoma-derived growth factor family. Protein Sci. 14:756–764.
    OpenUrlCrossRefPubMedWeb of Science
    1. Qiu C.,
    2. Sawada K.,
    3. Zhang X.,
    4. Cheng X.
    (2002) The PWWP domain of mammalian DNA methyltransferase Dnmt3b defines a new family of DNA-binding folds. Nat. Struct. Biol. 9:217–224.
    OpenUrlPubMedWeb of Science
  12. ↵
    1. Slater L. M.,
    2. Allen M. D.,
    3. Bycroft M.
    (2003) Structural variation in PWWP domains. J. Mol. Biol. 330:571–576.
    OpenUrlCrossRefPubMedWeb of Science
  13. ↵
    1. Ge Y. Z.,
    2. Pu M. T.,
    3. Gowher H.,
    4. Wu H. P.,
    5. Ding J. P.,
    6. Jeltsch A.,
    7. Xu G. L.
    (2004) Chromatin targeting of de novo DNA methyltransferases by the PWWP domain. J. Biol. Chem. 279:25447–25454.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Abouzied M. M.,
    2. El-Tahir H. M.,
    3. Prenner L.,
    4. Haberlein H.,
    5. Gieselmann V.,
    6. Franken S.
    (2005) Hepatoma-derived growth factor. Significance of amino acid residues 81–100 in cell surface interaction and proliferative activity. J. Biol. Chem. 280:10945–10954.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    1. El-Tahir H. M.,
    2. Abouzied M. M.,
    3. Gallitzendoerfer R.,
    4. Gieselmann V.,
    5. Franken S.
    (2009) Hepatoma-derived growth factor-related protein-3 interacts with microtubules and promotes neurite outgrowth in mouse cortical neurons. J. Biol. Chem. 284:11637–11651.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Everett A. D.,
    2. Stoops T.,
    3. McNamara C. A.
    (2001) Nuclear targeting is required for hepatoma-derived growth factor-stimulated mitogenesis in vascular smooth muscle cells. J. Biol. Chem. 276:37564–37568.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Lotze M. T.,
    2. Tracey K. J.
    (2005) High-mobility group box 1 protein (HMGB1): nuclear weapon in the immune arsenal. Nat. Rev. Immunol. 5:331–342.
    OpenUrlCrossRefPubMedWeb of Science
    1. Maxwell C. A.,
    2. McCarthy J.,
    3. Turley E.
    (2008) Cell-surface and mitotic-spindle RHAMM: moonlighting or dual oncogenic functions? J. Cell Sci. 121:925–932.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Radisky D. C.,
    2. Stallings-Mann M.,
    3. Hirai Y.,
    4. Bissell M. J.
    (2009) Single proteins might have dual but related functions in intracellular and extracellular microenvironments. Nat. Rev. Mol. Cell. Biol. 10:228–234.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Hsu S. C.,
    2. Hung M. C.
    (2007) Characterization of a novel tripartite nuclear localization sequence in the EGFR family. J. Biol. Chem. 282:10432–10440.
    OpenUrlAbstract/FREE Full Text
    1. Le Roy C.,
    2. Wrana J. L.
    (2005) Clathrin- and non-clathrin-mediated endocytic regulation of cell signalling. Nat. Rev. Mol. Cell. Biol. 6:112–126.
    OpenUrlCrossRefPubMedWeb of Science
    1. Lin S. Y.,
    2. Makino K.,
    3. Xia W.,
    4. Matin A.,
    5. Wen Y.,
    6. Kwong K. Y.,
    7. Bourguignon L.,
    8. Hung M. C.
    (2001) Nuclear localization of EGF receptor and its potential new role as a transcription factor. Nat. Cell Biol. 3:802–808.
    OpenUrlCrossRefPubMedWeb of Science
  23. ↵
    1. Poon I. K.,
    2. Jans D. A.
    (2005) Regulation of nuclear transport: central role in development and transformation? Traffic 6:173–186.
    OpenUrlCrossRefPubMedWeb of Science
  24. ↵
    1. Simons M.,
    2. Horowitz A.
    (2001) Syndecan-4-mediated signalling. Cell. Signalling 13:855–862.
    OpenUrlCrossRefPubMedWeb of Science
  25. ↵
    1. Klagsbrun M.,
    2. Baird A.
    (1991) A dual receptor system is required for basic fibroblast growth factor activity. Cell 67:229–231.
    OpenUrlCrossRefPubMedWeb of Science
  26. ↵
    1. Sue S. C.,
    2. Lee W. T.,
    3. Tien S. C.,
    4. Lee S. C.,
    5. Yu J. G.,
    6. Wu W. J.,
    7. Wu W. G.,
    8. Huang T. H.
    (2007) PWWP module of human hepatoma-derived growth factor forms a domain-swapped dimer with much higher affinity for heparin. J. Mol. Biol. 367:456–472.
    OpenUrlCrossRefPubMedWeb of Science
  27. ↵
    1. Zhan X.,
    2. Hu X.,
    3. Friesel R.,
    4. Maciag T.
    (1993) Long term growth factor exposure and differential tyrosine phosphorylation are required for DNA synthesis in BALB/c 3T3 cells. J. Biol. Chem. 268:9611–9620.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    1. Lin Y. H.,
    2. Lee S. C.,
    3. Chang P. Y.,
    4. Rajan P. K.,
    5. Sue S. C.,
    6. Wu W. G.
    (1999) Heparin binding to cobra basic phospholipase A2 depends on heparin chain length and amino acid specificity. FEBS Lett. 453:395–399.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Lee S. C.,
    2. Guan H. H.,
    3. Wang C. H.,
    4. Huang W. N.,
    5. Tjong S. C.,
    6. Chen C. J.,
    7. Wu W. G.
    (2005) Structural basis of citrate-dependent and heparan sulfate-mediated cell surface retention of cobra cardiotoxin A3. J. Biol. Chem. 280:9567–9577.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Tkachenko E.,
    2. Lutgens E.,
    3. Stan R. V.,
    4. Simons M.
    (2004) Fibroblast growth factor 2 endocytosis in endothelial cells proceed via syndecan-4-dependent activation of Rac1 and a Cdc42-dependent macropinocytic pathway. J. Cell Sci. 117:3189–3199.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Roghani M.,
    2. Moscatelli D.
    (1992) Basic fibroblast growth factor is internalized through both receptor-mediated and heparan sulfate-mediated mechanisms. J. Biol. Chem. 267:22156–22162.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    1. Yao J.,
    2. Xiong S.,
    3. Klos K.,
    4. Nguyen N.,
    5. Grijalva R.,
    6. Li P.,
    7. Yu D.
    (2001) Multiple signaling pathways involved in activation of matrix metalloproteinase-9 (MMP-9) by heregulin-β1 in human breast cancer cells. Oncogene 20:8066–8074.
    OpenUrlCrossRefPubMedWeb of Science
  33. ↵
    1. Fritz T. A.,
    2. Lugemwa F. N.,
    3. Sarkar A. K.,
    4. Esko J. D.
    (1994) Biosynthesis of heparan sulfate on β-D-xylosides depends on aglycone structure. J. Biol. Chem. 269:300–307.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    1. Goodfellow I. G.,
    2. Sioofy A. B.,
    3. Powell R. M.,
    4. Evans D. J.
    (2001) Echoviruses bind heparan sulfate at the cell surface. J. Virol. 75:4918–4921.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Schwartz N. B.,
    2. Cortes M.,
    3. King L. A.
    (2007) Got sulfate? Luring axons this way and that. Chem. Biol. 14:119–120.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Kreuger J.,
    2. Spillmann D.,
    3. Li J. P.,
    4. Lindahl U.
    (2006) Interactions between heparan sulfate and proteins: the concept of specificity. J. Cell Biol. 174:323–327.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. DiGabriele A. D.,
    2. Lax I.,
    3. Chen D. I.,
    4. Svahn C. M.,
    5. Jaye M.,
    6. Schlessinger J.,
    7. Hendrickson W. A.
    (1998) Structure of a heparin-linked biologically active dimer of fibroblast growth factor. Nature 393:812–817.
    OpenUrlCrossRefPubMedWeb of Science
  38. ↵
    1. Ornitz D. M.
    (2000) FGFs, heparan sulfate and FGFRs: complex interactions essential for development. BioEssays 22:108–112.
    OpenUrlCrossRefPubMedWeb of Science
  39. ↵
    1. Citores L.,
    2. Wesche J.,
    3. Kolpakova E.,
    4. Olsnes S.
    (1999) Uptake and intracellular transport of acidic fibroblast growth factor: evidence for free and cytoskeleton-anchored fibroblast growth factor receptors. Mol. Biol. Cell 10:3835–3848.
    OpenUrlAbstract/FREE Full Text
  40. ↵
    1. Gleizes P. E.,
    2. Noaillac-Depeyre J.,
    3. Dupont M. A.,
    4. Gas N.
    (1996) Basic fibroblast growth factor (FGF-2) is addressed to caveolae after binding to the plasma membrane of BHK cells. Eur. J. Cell Biol. 71:144–153.
    OpenUrlPubMed
  41. ↵
    1. Kusumi A.,
    2. Nakada C.,
    3. Ritchie K.,
    4. Murase K.,
    5. Suzuki K.,
    6. Murakoshi H.,
    7. Kasai R. S.,
    8. Kondo J.,
    9. Fujiwara T.
    (2005) Paradigm shift of the plasma membrane concept from the two-dimensional continuum fluid to the partitioned fluid: high-speed single-molecule tracking of membrane molecules. Annu. Rev. Biophys. Biomol. Struct. 34:351–378.
    OpenUrlCrossRefPubMedWeb of Science
  42. ↵
    1. Everett A. D.,
    2. Narron J. V.,
    3. Stoops T.,
    4. Nakamura H.,
    5. Tucker A.
    (2004) Hepatoma-derived growth factor is a pulmonary endothelial cell-expressed angiogenic factor. Am. J. Physiol. Lung Cell. Mol. Physiol. 286:L1194–L1201.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    1. Doherty G. J.,
    2. McMahon H. T.
    (2009) Mechanisms of endocytosis. Annu. Rev. Biochem. 78:857–902.
    OpenUrlCrossRefPubMedWeb of Science
  44. ↵
    1. Yu J.,
    2. Bian D.,
    3. Mahanivong C.,
    4. Cheng R. K.,
    5. Zhou W.,
    6. Huang S.
    (2004) p38 mitogen-activated protein kinase regulation of endothelial cell migration depends on urokinase plasminogen activator expression. J. Biol. Chem. 279:50446–50454.
    OpenUrlAbstract/FREE Full Text
  45. ↵
    1. Mao J.,
    2. Xu Z.,
    3. Fang Y.,
    4. Wang H.,
    5. Xu J.,
    6. Ye J.,
    7. Zheng S.,
    8. Zhu Y.
    (2008) Hepatoma-derived growth factor involved in the carcinogenesis of gastric epithelial cells through promotion of cell proliferation by Erk1/2 activation. Cancer Sci. 99:2120–2127.
    OpenUrlCrossRefPubMed
  46. ↵
    1. Behren A.,
    2. Binder K.,
    3. Vucelic G.,
    4. Herberhold S.,
    5. Hirt B.,
    6. Loewenheim H.,
    7. Preyer S.,
    8. Zenner H. P.,
    9. Simon C.
    (2005) The p38 SAPK pathway is required for Ha-ras induced in vitro invasion of NIH3T3 cells. Exp. Cell Res. 303:321–330.
    OpenUrlCrossRefPubMed
  47. ↵
    1. Balklava Z.,
    2. Verderio E.,
    3. Collighan R.,
    4. Gross S.,
    5. Adams J.,
    6. Griffin M.
    (2002) Analysis of tissue transglutaminase function in the migration of Swiss 3T3 fibroblasts: the active-state conformation of the enzyme does not affect cell motility but is important for its secretion. J. Biol. Chem. 277:16567–16575.
    OpenUrlAbstract/FREE Full Text
  48. ↵
    1. Okamoto Y.,
    2. Ninomiya H.,
    3. Miwa S.,
    4. Masaki T.
    (2000) Cholesterol oxidation switches the internalization pathway of endothelin receptor type A from caveolae to clathrin-coated pits in Chinese hamster ovary cells. J. Biol. Chem. 275:6439–6446.
    OpenUrlAbstract/FREE Full Text
  49. ↵
    1. Nishimura S.,
    2. Takahashi S.,
    3. Kamikatahira H.,
    4. Kuroki Y.,
    5. Jaalouk D. E.,
    6. O'Brien S.,
    7. Koivunen E.,
    8. Arap W.,
    9. Pasqualini R.,
    10. Nakayama H.,
    11. Kuniyasu A.
    (2008) Combinatorial targeting of the macropinocytotic pathway in leukemia and lymphoma cells. J. Biol. Chem. 283:11752–11762.
    OpenUrlAbstract/FREE Full Text
  50. ↵
    1. Seo Y. S.,
    2. Park J. J.,
    3. Kim J. H.,
    4. Kim J. Y.,
    5. Yeon J. E.,
    6. Kim J. S.,
    7. Byun K. S.,
    8. Bak Y. T.
    (2007) Usefulness of MMP-9/TIMP-1 in predicting tumor recurrence in patients undergoing curative surgical resection for gastric carcinoma. Dig. Dis. Sci. 52:753–759.
    OpenUrlCrossRefPubMed
  51. ↵
    1. Wang I. K.,
    2. Lin-Shiau S. Y.,
    3. Lin J. K.
    (2000) Suppression of invasion and MMP-9 expression in NIH 3T3 and v-H-Ras 3T3 fibroblasts by lovastatin through inhibition of ras isoprenylation. Oncology 59:245–254.
    OpenUrlCrossRefPubMedWeb of Science
  52. ↵
    1. Takahashi C.,
    2. Sheng Z.,
    3. Horan T. P.,
    4. Kitayama H.,
    5. Maki M.,
    6. Hitomi K.,
    7. Kitaura Y.,
    8. Takai S.,
    9. Sasahara R. M.,
    10. Horimoto A.,
    (1998) Regulation of matrix metalloproteinase-9 and inhibition of tumor invasion by the membrane-anchored glycoprotein RECK. Proc. Natl. Acad. Sci. U.S.A. 95:13221–13226.
    OpenUrlAbstract/FREE Full Text
  53. ↵
    1. Bjorklund M.,
    2. Heikkila P.,
    3. Koivunen E.
    (2004) Peptide inhibition of catalytic and noncatalytic activities of matrix metalloproteinase-9 blocks tumor cell migration and invasion. J. Biol. Chem. 279:29589–29597.
    OpenUrlAbstract/FREE Full Text
View Abstract
Back to top

 

 

January 2011

Volume: 433 Issue: 1

Biochemical Journal: 433 (1)

Actions

Email
Cell surface heparan sulfates mediate internalization of the PWWP/HATH domain of HDGF via macropinocytosis to fine-tune cell signalling processes involved in fibroblast cell migration
Chia-Hui Wang, Fabian Davamani, Shih-Che Sue, Shao-Chen Lee, Po-long Wu, Fan-Mei Tang, Chiaho Shih, Tai-huang Huang, Wen-guey Wu
Biochemical Journal Jan 2011, 433 (1) 127-138; DOI: 10.1042/BJ20100589
Permalink:
del.icio.us logo Digg logo Reddit logo Technorati logo Twitter logo CiteULike logo Facebook logo Mendeley logo
Citation Tools
Print
Alerts
Request Permissions
Save to my folders

View Full PDF

Jump To

Keywords

cell migration
HATH domain
heparan sulfate binding
hepatoma-derived growth factor (HDGF)
macropinocytosis
proliferation

Related Articles

Cited By...

Subjects