Portal de Periódicos da CAPES

Protein Gene Product 9.5 is Expressed by Fibroblasts in Human Cutaneous Wounds

1998; Elsevier BV; Volume: 111; Issue: 4 Linguagem: Inglês

10.1046/j.1523-1747.1998.00330.x

1523-1747

John E. Olerud, Diane S. Chiu, Marcia L. Usui, Nicole S. Gibran, John C. Ansel,

Wound Healing and Treatments

In a study initially designed to evaluate reinnervation of human cutaneous wounds using an antibody to the neuroneal marker protein gene product (PGP) 9.5, we observed marked immunostaining of cells with morphologic features of fibroblasts in the wounds. PGP 9.5 has recently been shown to be an important enzyme in the highly conserved ubiquitin system of proteolysis. Because the ubiquitin system is known to play an important role in regulating the cell cycle, the presence of PGP 9.5 in cells at a wound site was of considerable interest. Our objectives were to clarify the time frame for the appearance of PGP 9.5 and ubiquitin in wounds, to verify that PGP 9.5 is produced by wound fibroblasts, and to evaluate a potential role for these proteins in the tissue repair process. Standard incisional human wounds were stained with antibodies specific for PGP 9.5 and ubiquitin. At 7 d, stellate cells with morphologic features of fibroblasts stained for PGP 9.5, whereas earlier wounds were generally negative. In 14 and 21 d incised wounds and in chronic granulation tissue from nonhealing ulcers there was strong cellular staining for PGP 9.5 and for ubiquitin. These stellate cells also showed expression of mRNA for PGP 9.5 by reverse transcriptase-polymerase chain reaction in situ hybridization. PGP 9.5 was detected in cultured fibroblasts both by reverse transcriptase-polymerase chain reaction and by northern blot analysis. Confocal microscopy showed colocalization of antibodies to PGP 9.5 and prolyl-4-hydroxylase (a fibroblast marker) as well as colocalization of PGP 9.5 and the platelet derived growth factor β receptor. We conclude that ubiquitin and PGP 9.5 were expressed by fibroblasts during the granulation tissue and remodeling phases wound healing. The mRNA for PGP 9.5 was demonstrated in stellate cells in chronic wounds and in fibroblasts in culture. The appearance of these degradative proteins in later wounds suggests a downregulation function in the wound healing response. In a study initially designed to evaluate reinnervation of human cutaneous wounds using an antibody to the neuroneal marker protein gene product (PGP) 9.5, we observed marked immunostaining of cells with morphologic features of fibroblasts in the wounds. PGP 9.5 has recently been shown to be an important enzyme in the highly conserved ubiquitin system of proteolysis. Because the ubiquitin system is known to play an important role in regulating the cell cycle, the presence of PGP 9.5 in cells at a wound site was of considerable interest. Our objectives were to clarify the time frame for the appearance of PGP 9.5 and ubiquitin in wounds, to verify that PGP 9.5 is produced by wound fibroblasts, and to evaluate a potential role for these proteins in the tissue repair process. Standard incisional human wounds were stained with antibodies specific for PGP 9.5 and ubiquitin. At 7 d, stellate cells with morphologic features of fibroblasts stained for PGP 9.5, whereas earlier wounds were generally negative. In 14 and 21 d incised wounds and in chronic granulation tissue from nonhealing ulcers there was strong cellular staining for PGP 9.5 and for ubiquitin. These stellate cells also showed expression of mRNA for PGP 9.5 by reverse transcriptase-polymerase chain reaction in situ hybridization. PGP 9.5 was detected in cultured fibroblasts both by reverse transcriptase-polymerase chain reaction and by northern blot analysis. Confocal microscopy showed colocalization of antibodies to PGP 9.5 and prolyl-4-hydroxylase (a fibroblast marker) as well as colocalization of PGP 9.5 and the platelet derived growth factor β receptor. We conclude that ubiquitin and PGP 9.5 were expressed by fibroblasts during the granulation tissue and remodeling phases wound healing. The mRNA for PGP 9.5 was demonstrated in stellate cells in chronic wounds and in fibroblasts in culture. The appearance of these degradative proteins in later wounds suggests a downregulation function in the wound healing response. immunocytochemistry platelet derived growth factor receptor protein gene product 9.5 terminal deoxynucleotidyltransferasemediated nickend labeling Protein gene product 9.5 (PGP 9.5) is widely used as a marker for cutaneous innervation. The 26 kDa protein was isolated from brain tissue (Jackson and Thompson, 1981Jackson P. Thompson R.J. The demonstration of new human brain-specific proteins by high-resolution two-dimensional polyacrylamide gel electrophoresis.J Neurol Sci. 1981; 49: 429-438Abstract Full Text PDF PubMed Scopus (187) Google Scholar) and subsequently shown to be an enzyme of the ubiquitin system (ubiquitin C-terminal hydrolase) (Wilkinson et al., 1989Wilkinson K.D. Lee K. Deshpande S. Duerksen-hughes P. Boss J.M. Pohl J. The neuron-specific protein PGP 9.5 is a ubiquitin carboxyl-terminal hydrolase.Science. 1989; 246: 670-673Crossref PubMed Scopus (767) Google Scholar). Ubiquitin, an abundant 76-amino acid protein, is one of the most phylogenetically conserved proteins known. The sequence is identical in all animals and differs in yeast by only three amino acids (Ozkaynak et al., 1987Ozkaynak E. Finley D. Solomon M.J. Varshavsky A. The yeast ubiquitin genes: a family of natural gene fusions.Embo J. 1987; 6: 1429-1439Crossref PubMed Scopus (359) Google Scholar). Although phosphorylation is often a mechanism to change the functional state of proteins, ubiquitinization targets proteins for degradation. Ubiquitin mediated protein degradation plays a critical role in cellular functions such as cell cycle control, DNA repair, and stress responses (Finley et al., 1991Finley D. Chau V. Ubiquitination Annu Rev Cell Biol. 1991; 7: 25-69Crossref PubMed Scopus (421) Google Scholar). PGP 9.5 removes ubiquitin from proteins undergoing degradation and allows recycling of free ubiquitin from the conjugation products (Finley et al., 1991Finley D. Chau V. Ubiquitination Annu Rev Cell Biol. 1991; 7: 25-69Crossref PubMed Scopus (421) Google Scholar). Relevant to the study of wounds, ubiquitination has been shown to play a role in degradation of PDGF receptor-ligand complexes (Mori et al., 1992Mori S. Heldin C-H. Claesson-welsh L. Ligand-induced polyubiquitination of the platelet-derived growth factor β-receptor.J Biol Chem. 1992; 267: 6429-6434Abstract Full Text PDF PubMed Google Scholar). PDGF is known to stimulate both mitogenesis and fibrogenesis in fibroblasts (Ross et al., 1986Ross R. Raines E.W. Bowen-pope D.F. The biology of platelet-derived growth factor.Cell. 1986; 46: 155-169Abstract Full Text PDF PubMed Scopus (1583) Google Scholar), which are key functions in the wound healing process. Other cytokine receptor/ligand complexes known to be important in wound healing undergo similar ubiquitin mediated degradations (Mori et al., 1995Mori S. Claesson-welsh L. Okuyama Y. Saito Y. Ligand-induced polyubiquitination of receptor tyrosine kinase.Biochem Biophys Res Com. 1995; 213: 32-39Crossref PubMed Scopus (49) Google Scholar). These include platelet derived growth factor alpha-receptor, epidermal growth factor receptor, and fibroblast growth factor-receptor. In this study we observed marked immunostaining of PGP 9.5 in cells with morphologic features of fibroblasts, particularly in mature wounds (14 and 21 d). This was an unexpected finding because PGP 9.5 expression was originally thought to be quite specific for neuroneal and neuroendocrine tissue (Thompson et al., 1983Thompson R.J. Doran J.F. Jackson P. Dhillon A.P. Rode J. PGP 9.5 – a new marker for vertebrate neurons and neuroendocrine cells.Brain Res. 1983; 278: 224-228Crossref PubMed Scopus (661) Google Scholar). PGP 9.5 has, however, been detected using two-dimensional gel electrophoresis or immunocytochemistry in non-neuroneal cells from renal distal convoluted tubules, ovarian follicles, corpus luteum, and epididymal epithelium, as well as in spermatogonia and Leydig cells (Wilson et al., 1988Wilson P.O.G. Barber P.C. Hamid Q.A. et al.The immunolocalization of protein gene product 9.5 using rabbit polyclonal and mouse monoclonal antibodies.Br J Exp Path. 1988; 69: 91-104PubMed Google Scholar;Santamaria et al., 1993Santamaria L. Martin R. Paniagua R. Fraile B. Nistal M. Terenghi G. Polak J.M. Protein gene product 9.5 and ubiquitin immunoreactivities in rat epididymis epithelium.Histochemistry. 1993; 100: 131-138Crossref PubMed Scopus (31) Google Scholar). Prior to our observations in human fibroblasts, 1Olerud JE, Gibran NS, Usui ML, Chiu DS, Ansel JC: PGP 9.5 and ubiquitin immunostaining in human cutaneous wounds. J Invest Dermatol 4:421a 1995 (abstr.)1Olerud JE, Gibran NS, Usui ML, Chiu DS, Ansel JC: PGP 9.5 and ubiquitin immunostaining in human cutaneous wounds. J Invest Dermatol 4:421a 1995 (abstr.)2Karnovsky MJ: A formaldehyde-glutaraldehyde fixative of high osmolarity for use in electron microscopy. J Cell Biol 27:137a 1965 (abstr.) the only evidence that fibroblasts produce PGP 9.5 came from two-dimensional immunoblotting, comigration, and microsequencing of proteins recovered from human embryonal MRC-5 lung fibroblasts (Honoré et al., 1991Honoré B. Rasmussen H.H. Vandekerckhove J. Celis J.E. Neuronal protein gene product 9.5 (IEF SSP 6104) is expressed in cultured human MRC-5 fibroblasts of normal origin and is strongly down-regulated in their SV40 transformed counterparts.Febs Lett. 1991; 280: 235-240Abstract Full Text PDF PubMed Scopus (17) Google Scholar). More recently,DiPaolo et al., 1995DiPaolo B.R. Pignolo R.J. Cristofalo V.J. Identification of proteins differentially expressed in quiescent and proliferatively senescent fibroblast cultures.Exp Cell Res. 1995; 220: 178-185Crossref Scopus (30) Google Scholar used similar two-dimensional immunoblotting techniques to show the presence of PGP 9.5 in cultured human neonatal foreskin fibroblasts. We used immunocytochemistry (ICC) to document the presence of both PGP 9.5 and ubiquitin in standard incisional wounds and in chronic decubitus ulcers. Additionally, we wanted to determine if the stellate cells were fibroblasts and whether they produced mRNA for PGP 9.5 as well as expressing the protein. We questioned whether the stellate cells showed coexpression of PGP 9.5 and platelet derived growth factor β receptor (PDGFβ-r), because that would be consistent with a potential mechanism of action in the wound healing process involving degradation of one of the monomeric tyrosine kinase receptors known to be important in wound healing. We used confocal microscopy, transmission electron microscopy (TEM), northern blot analysis, and reverse transcriptase-polymerase chain reaction (RT-PCR) techniques for these studies. Simplate-II (General Diagnostics, Organon Teknika, NC) bleeding time devices were used to create wounds on the arms and legs of normal male volunteers 66 ± 6 y of age (mean ± SD). The Simplate-II is a spring loaded instrument that, when activated, projects a pair of blades 5 mm in length, 1 mm in depth, and 3 mm apart. This human wound model has been previously described in detail (Olerud et al., 1995Olerud J.E. Odland G.F. Burgess E.M. Wyss C.R. Fisher L.D. Matsen III, F.A. A model for the study of wounds in normal elderly adults and patients with peripheral vascular disease or diabetes mellitus.J Surg Res. 1995; 59: 349-360Abstract Full Text PDF PubMed Scopus (32) Google Scholar). Volunteers were recruited using methods approved by the University of Washington Institutional Review Board for Human Subjects (UWIRBHS). All subjects were screened and shown to be free of neuropathy and diabetes mellitus as previously described (Olerud et al., 1995Olerud J.E. Odland G.F. Burgess E.M. Wyss C.R. Fisher L.D. Matsen III, F.A. A model for the study of wounds in normal elderly adults and patients with peripheral vascular disease or diabetes mellitus.J Surg Res. 1995; 59: 349-360Abstract Full Text PDF PubMed Scopus (32) Google Scholar). Two 4 mm punch biopsies were removed from wounds 1, 2, 3, 4, 7, 14, and 21 d following injury. Half of each sample was fixed in half strength Karnovsky's fixative 2Karnovsky MJ: A formaldehyde-glutaraldehyde fixative of high osmolarity for use in electron microscopy. J Cell Biol 27:137a 1965 (abstr.)and then embedded in Epon 812 (Luft, 1961Luft J.H. Improvements in epoxy resin embedding methods.J Biophys Biochem Cytol. 1961; 9: 409-414Crossref PubMed Scopus (6126) Google Scholar) for high quality light microscopy and TEM. The other half was either fixed in 4% paraformaldehyde/phosphate buffered saline (PBS) for 24 h and processed for paraffin embedding or frozen in OCT (Tissue Tek, Miles, IN). Chronic ulcers were studied. Six paraplegic subjects undergoing surgery to excise and repair chronic sacral decubitus ulcers agreed to allow the study of tissue removed during surgery. Consent was obtained using methods approved by the UWIRBHS. Ulcer tissue was taken immediately upon excision, fixed in 10% neutral buffered formalin, and processed for paraffin embedding or frozen in OCT (Tissue Tek). To detect PGP 95 and ubiquitin in standard wounds and ulcers Six micron tissue sections of the paraformaldehyde fixed wounds and the neutral buffered formalin fixed ulcers, mounted on Superfrost Plus (Labcraft) glass slides, were deparaffinized and processed for immunoperoxidase with the following solutions: Tris buffered saline (TBS) washes, H2O2/TBS blocking (1%, 30 min), goat serum/TBS blocking (13%, 30 min), polyclonal rabbit PGP 95 anti-serum (Accurate Chemical Scientific, Westbury, NJ) at 1:1500 dilution for 1 h at room temperature or polyclonal rabbit ubiquitin anti-serum (Dako, Carpenteria, CA) at 1:250 dilution for 1 h Diluent used for all antibodies was 01% bovine serum albumin in TBS and all incubations were done at room temperature The secondary antibody was biotinylated goat anti-rabbit antibody (Vector Laboratories, Burlingame, CA) at 1:200 dilution for 30 min, followed by avidin-biotin complex (ABC Universal Kit, Zymed Laboratories, San Francisco, CA) at 1:200 dilution for 30 min Sections were then visualized for immunoreactivity using 3,3′ diaminobenzidine (Sigma, St Louis, MO) as a chromogen (012% in H2O, 20 min) Frozen sections of wounds washed in TBS were post-fixed with 01% glutaraldehyde/TBS for 5 min The slides were processed as described above except that they were blocked with horse and goat sera/TBS (3% and 13%, respectively, 30 min) and incubated with a monoclonal mouse ascites PDGFβ-r antibody (Zymogenetics, Seattle, WA) at 1:1000 dilution for 1 h and secondary biotinylated horse anti-mouse (Vector Laboratories) at 1:200 dilution for 30 min Frozen sections of wounds were post-fixed in cold acetone for 5 min, washed in TBS, and processed as above using a mouse monoclonal anti-human prolyl-4-hydroxylase antibody (5B5) (Dako) at 1:100 dilution and biotinylated horse anti-mouse IgG secondary antibody Frozen sections were acetone fixed, washed in TBS, blocked with 13% goat and 03% horse sera for 30 min, then incubated with PGP 95 antibody at 1:1000 dilution and 5B5 antibody at 1:100 dilution for 1 h at room temperature The diluent for all antibodies contained 01% bovine serum albumin and 01% Tween 20 (polyoxyethlyenesorbitan monosaurate, Sigma) Sections were then sequencially immunolabeled with a fluorescein-labeled goat anti-rabbit (Vector Laboratories) at 1:200 dilution and biotinylated horse anti-mouse IgG (Vector Laboratories) at 1:400 dilution for 30 min Slides were rinsed in TBS and immunolabeled with Texas Red Streptavidin (Vector Laboratories) at 1:1600 dilution for 30 min, washed in TBS, then coverslipped using Vectashield (Vector Laboratories) as a mounting medium Immunolabeling was as above except PDGFβ-r antibody was used (1:500) in place of 5B5 To visualize fluorescent double labeling Image acquisition for the fluorescence double-labeling was on a BioRad MRC-600 confocal microscope equipped with a krypton/argon mixed gas laser The dual excitation mode was used, passing 488 nm wavelength for fluoroscein isothiocyanate excitation and 568 nm wavelength for Texas Red excitation Dual image files were first split and then imported into Adobe Photoshop for display purposes, with green for fluoroscein isothiocyanate and red for Texas Red, and then superimposed to show colocalization of label Color prints were output to a Tektronix dye sublimation printer at 300 dpi To evaluate ultrastructural features of stellate cells Standard 21 d wound specimens that had been fixed directly in half strength Karnovsky's fixative and processed by standard methods for TEM were studied using a Phillips 420 transmission electron microscope Oligonucleotide primers were synthesized (Bio-synthesis, Lewisville, TX) corresponding to bp 170–190 5′ primer, sense CAG CAT GAG AAC TTC AGG AAA, and the complementary reverse sequence against bp 488–508 3′ primer, anti-sense GCC ATC CAC GTT GTT AAA CAG of PGP 9.5 cDNA as previously described (Day and Thompson, 1987Day I.N.M. Thompson R.J. Molecular cloning of cDNA coding for human PGP 9.5 protein. A novel cytoplasmic marker for neurones and neuroendocrine cells.Febs Lett. 1987; 210: 157-160Abstract Full Text PDF PubMed Scopus (86) Google Scholar). RNA was isolated from cultured fibroblasts and reverse transcribed to make complementary cDNA to PGP9.5. PCR was performed on 0, 1, 2, and 5 μl of the preconfluent and confluent RT reaction products by adding equal amounts of the 3′ and 5′ primers, 1.25 × d4NTP mixture (Pharmacia LKB Biotechnology, Piscataway, NJ) and 10 × PCR buffer + MgCl2 (Boehringer, Indianapolis, IN). Taq polymerase (Boehringer) was added during the hot start at 85°C. Sterile mineral oil also was added to prevent evaporation. The PCR reaction was carried out in a PTC-100-CR machine (MJ Research, Watertown, MA) for 35 cycles using an annealing temperature of 60°C, denaturing temperature of 94°C, and extension temperature of 72°C. PCR reaction products were digested with restriction enzymes MboI (New England Biolabs, Beverley, MA) and EcoRI (Boehringer, Indianapolis, IN) at 37°C for 2 h. Gel electrophoresis was performed on uncut PCR product, MboI-cut PCR product, and EcoRI-cut PCR product to verify expected size. The uncut RT-PCR product was then sequenced in both directions using a PCR sequencing reaction containing the 5′ and 3′ NEP oligonucleotide primer and Dye Terminator with AmpliTaq DNA Polymerase FS (ABI Prism) followed by sequencing using a model 377 Fluorescence Sequencer (ABI Prism). Fifteen micrograms of newborn foreskin fibroblast RNA was run on a formaldehyde gel and transferred to Hybond-N nylon membrane (Amersham, Arlington Heights, IL) using a Vacugene apparatus (Pharmacia LKB Biotechnology) The blot was cross-linked and the 28S and 18S bands were marked Twenty-five nanograms of gel-purified PGP 95 cDNA was labeled with 32P-αCTP (Amersham) using the Rediprime DNA labeling system (Amersham) The blot was prehybridized at 65°C for 1 h with Rapid-Hyb buffer (Amersham), then hybridized with the radiolabeled probe at 65°C for 2 h, washed at room temperature with 2 × standard saline citrate containing 01% sodium dodecyl sulfate, washed at 65°C with 01 × standard saline citrate containing 01% sodium dodecyl sulfate, and exposed to film for 16 h at room temperature prior to development RT-PCR in situ hybridization was performed to identify cells containing PGP 9.5 mRNA. Six micron, formalin-fixed, paraffin-embedded sections of chronic decubitus ulcers were deparaffinized and rehydrated in 0.5×standard saline citrate followed by rinsing in RNase buffer (500 mM NaCl, 10 mM Tris-HCl, pH 8). Sections were then digested with proteinase K (Sigma) in RNase buffer (20 μg per ml) at 37°C for 10 min, rinsed in 0.5×standard saline citrate and incubated with RQ1 RNase-free DNase (10 μl per ml, Promega, Madison, WI) for 10 min at room temperature, and rinsed in DEPC H2O. Slides were covered with RT reaction mixture containing the same ingredients as described above for fibroblast RNA RT reaction. Slides were then incubated at 42°C for 1 h in a humidified chamber, and followed by heat inactivation of the MMLV RT at 65°C for 5 min buffer. Slides were rinsed in sterile dH2O and placed in the PCR machine. A PCR reaction mixture consisting of ingredients identical to those described above for the fibroblast PCR reaction, with the exception of replacing d4NTP with biotin 16-dUTP (Boehringer) and 10 mM d3NTPs containing dATP, dGTP, and dCTP, was added to each of the slides during the 85°C hot start and coverslipped with rubber-sealed coverslips (Research Products International, Mount Prospect, IL). Slides were run for 10 cycles at temperatures described above, washed in dH2O, and rinsed in PBS. Sections were covered with avidin conjugated to alkaline phosphatase (Dako) in 1% bovine serum albumin/PBS (1:250 dilution) at 37°C for 1 h then washed in Triton/TBS, followed by equilibration in TBS pH 9.5. Colorimetric development was performed using NBT/BCIP (Boehringer) in TBS pH 9.5 for 10 min at room temperature. Reaction was stopped by rinsing in dH2O and slides were counterstained with methyl green. Two sets of controls were run in an identical fashion except that one set was run without addition of primers and the other was run without addition of MMLV RT and Taq Polymerase. To assess a role for the PGP 9.5/ubiquitin system in wound apoptosis Six micron sections of paraformaldehyde-fixed, paraffin embedded tissue were processed for TUNEL assay according to methods outlined in the ApopTag Plus Kit (Oncor, Gaithersburg, MD) with the following modifications: Histoclear II (National Diagnostics, Atlanta, GA) was used in place of xylene in the deparaffinizing process, Proteinase K (Sigma) was used at 20 μg PBS per ml, 20 min at 37°C or 50 μg PBS per ml at 32°C for 20 min, 0.00036% 4–6 diamidino-2 phenylindole (DAPI)/TBS was applied for 3 min to counterstain nuclei, and Prolong Antifade Kit (Molecular Probes, Eugene, OR) was used as slide mounting media. Serial tissue sections were labeled with PGP 9.5 at 1:1500 dilution for 1 h, biotinylated goat anti-rabbit (Vector Laboratories) at 1:400 dilution for 30 min and fluorescent Cy 5 conjugated streptavidin (Jackson ImmunoResearch Laboratories, West Grove, PA) at 1:1000 dilution for 30 min was used. PGP 9.5 was observed by ICC to stain cutaneous nerve fibers in the epidermis and dermis of normal skin adjacent to the standard wounds of all ages (Figure 1). No cellular staining was observed in 1, 2, or 3 d wounds, although staining was occasionally observed in stellate cells adjacent to 4 d wounds. PGP 9.5 was clearly observed in the bed of 7 d wounds in cells with morphologic features of fibroblasts (Figure 2a). By 14 d, the cellular staining was strong in these stellate cells (Figure 2b). Similar intensity persisted in the 21 d wounds (not shown). PGP 9.5 staining was observed in most, but not all, of the stellate cells in the granulation tissue of chronic ulcers (Figure 3a). Stellate cells in the dermis adjacent to the ulcers did not show immunostaining (Figure 3b).Figure 2Cells in wound beds of 7 and 14 d wounds stain positively with antibodies to PGP 9.5, ubiquitin, and PDGFβ-r. Arrows indicate the positive cellular staining. Controls without primary antibodies do not show any cellular staining. Arrows in parts (g) and (h) indicate cells only showing the nuclear counterstain without antibody staining.View Large Image Figure ViewerDownload (PPT)Figure 3Cells in the granulation tissue from a chronic decubitus ulcer stain positively for both PGP 9.5 and ubiquitin. Arrows indicate positive cellular staining in the ulcer bed (a, c) for PGP 9.5 and ubiquitin, respectively. Cells in the tissue adjacent to the ulcer (b) do not show PGP 9.5 positive staining of stellate cells (small arrows); however, the large nerve (arrowhead) shows positive PGP 9.5 staining. Control (d) without primary antibodies do not show any cellular staining. Arrows in (d) indicate cells only showing the nuclear counterstain without PGP 9.5 staining.View Large Image Figure ViewerDownload (PPT) Similar to PGP 9.5, ubiquitin staining was observed in stellate cells in the standard human wounds at 7, 14 (Figure 2c, d), and 21 d (not shown), but not in earlier wounds. Likewise, ubiquitin staining of granulation tissue from chronic decubitus ulcers was also observed (Figure 3c). Immunostaining for PDGFβ-r was observed in stellate cells of 7, 14 (Figure 2e, f), and 21 d old wounds (not shown) but was not observed in earlier wounds. PDGFβ-r staining was noted in endothelial cells and pericytes adjacent to the wounds at all time points studied (not shown). Confocal microscopy of a 14 d wound (Figure 4) showed further evidence that the stellate cells are fibroblasts. Some stellate cells in the wound bed shown in Figure 4 are stained green by immunofluorescence for PGP 9.5 (Figure 4a), whereas other stellate cells stained red with the fibroblast marker 5B5 (Figure 4b). Some cells appear orange on the merged confocal micrograph, indicating costaining for PGP 9.5 and 5B5. It should be noted that not all cells staining with antibodies to 5B5 are positive for PGP 9.5. Figure 5 is a confocal micrograph of the same 14 d wound stained with antibodies to PGP 9.5 (green) and PDGFβ-r (red). Coexpression of the two proteins is demonstrated by costaining of the stellate cells in this merged confocal micrograph. Stellate cells in 14 d wounds prepared for TEM include extensive rough endoplasmic reticulum, Golgi membranes, and lack lysosomal vacuoles consistent with fibroblasts (Figure 6). These cells are closely associated with fibrillar collagen. A 339 bp segment of cDNA was amplified from newborn foreskin fibroblast RNA (Figure 7, lane 1). Restriction enzyme digestions of the PCR product were consistent with the predicted products corresponding to the PGP 9.5 cDNA sequence 170–508 bp. Gel electrophoresis on the EcoRI-cut PCR product yielded two bands at the expected sizes of 249 and 90 bp (Figure 7, lane 2), and the MboI-cut PCR product yielded a doublet at the expected sizes of 160 bp and 179 bp (Figure 7, lane 3). The DNA sequence of the uncut PCR product was identical with the predicted sequence corresponding to the PGP 9.5 cDNA at 170–508 bp (Day and Thompson, 1987Day I.N.M. Thompson R.J. Molecular cloning of cDNA coding for human PGP 9.5 protein. A novel cytoplasmic marker for neurones and neuroendocrine cells.Febs Lett. 1987; 210: 157-160Abstract Full Text PDF PubMed Scopus (86) Google Scholar). A single band of ≈1 kb was visualized with northern blotting of newborn foreskin fibroblast RNA with the 339 bp PGP 9.5 probe. This band corresponds to the predicted intact cDNA of PGP 9.5 (1014 bp) (Day and Thompson, 1987Day I.N.M. Thompson R.J. Molecular cloning of cDNA coding for human PGP 9.5 protein. A novel cytoplasmic marker for neurones and neuroendocrine cells.Febs Lett. 1987; 210: 157-160Abstract Full Text PDF PubMed Scopus (86) Google Scholar). RT-PCR in situ hybridization performed on chronic decubitus ulcers from patients with spinal cord injuries demonstrated marked PGP 9.5 mRNA localization in stellate cells with morphologic features of fibroblasts within the wound bed (Figure 8b). PGP 9.5 mRNA was not detected in similar cells in the dermis distant from the wound bed (Figure 8c). It should be noted that not all of the stellate cells in the wound bed exhibited PGP 9.5 mRNA staining, consistent with confocal microscopy that showed similar results for PGP 9.5 protein. PGP 9.5 mRNA was also found in the granular cell layer of the epidermis adjacent to wounds. Additionally, cells surrounding many blood vessels (pericytes) both within and away from the wound bed (Figure 8b, c) were noted to express mRNA for PGP 9.5. This experiment was performed to evaluate the role of PGP 9.5 in the apoptosis pathway. As expected for normal epidermis, a few cells in the granular layer of the epidermis adjacent to the 7, 14, and 21 d wounds showed TUNEL positive staining (Figure 9e). There was only one or two TUNEL positive cells for all time points in the dermal component of the wound bed (Figure 9c, d), whereas PGP 9.5 staining cells are abundant (Figure 9a, b). Hence, our findings do not support a role for PGP 9.5 in wound apoptosis during the time periods studied. The staining for PGP 9.5 and ubiquitin appeared in the stellate cells of wounds at day 7 and became more intense in day 14 and 21 wounds. Earlier wounds of 1, 2, and 3 d showed no cellular immunostaining for PGP 9.5. It was also noted that protein staining was limited to the wound area, whereas adjacent skin on the same microscopic section did not show staining. Subsequent studies of tissue from patients with chronic sacral ulcers showed stellate cells staining for both PGP 9.5 and ubiquitin in a pattern similar to the incised wounds. The stellate cells expressing PGP 9.5 in wounds were identified as fibroblasts, using the following lines of evidence: (i) the stellate cells had morphologic features of fibroblasts by light microscopy and ultrastructural features of fibroblasts by TEM; (ii) confocal microscopy of immunostained sections of wound tissue from a 14 d wound showed colocalization in the stellate cells of PGP 9.5 and 5B5, a relatively specific marker for fibroblasts (Esterre et al., 1992Esterre P. Melin M. Serrar M. Grimaud J-A. New specific markers of human and mouse fibroblasts.Cell Mol Biol. 1992; 38: 297-301PubMed Google Scholar;Bosseloir et al., 1994Bosseloir A. Heinen E. Defrance T. Bouzhazha F. Antoine N. Simar L.J. Moabs MAS516 and 5B5, two fibroblast markers, recognize human follicular dendritic cells.Immunol Lett. 1994; 42: 49-54Crossref PubMed Scopus (23) Google Scholar); (iii) we established that newborn foreskin fibroblasts are capable of making mRNA for PGP 9.5, using RT-PCR and north