7. MATERIALS AND METHODS

The study was approved by the Ethics Committee of the Department of Dermatology, Helsinki University Central Hospital. All tissue samples were formalin-fixed and paraffin embedded. Informed consent was obtained from individual subjects for all procedures.

Chronic dermal wounds. Human skin samples (n=14) were collected from patients with chronic leg ulcers undergoing excision and skin grafting prodecures at the Department of Plastic Surgery, Helsinki University Central Hospital. The patients were aged between 50 to 91 years. The ulcers were 3 months to 10 years old and had not responded to conservative treatment. Archival specimens of chronic ulcers (n=6) / vasculitis (n=5) were obtained from the Department of Pathology, Helsinki University Central Hospital.

Normally healing human wounds. As controls for acute full thickness wounds (n = 25) biopsies of normally healing donor areas on the anterior thigh were obtained from patients of the Department of Dermatology, University of Helsinki, undergoing pinch grafting procedure (Ceilley et al, 1977). On day 0, several pieces of skin measuring 8 mm in diameter and including a portion of the dermis centrally, were excised from the donor area and transferred to the ulcer area. Biopsies containing the donor area wound were taken 1, 2, 3, 4, 5, 6, 7 and 9 days postwounding from patients aged between 55 to 94 years.

Suction blisters were obtained from Central Military Hospital, Helsinki. Blisters were induced on the abdominal skin of four healthy volunteers under 30 years of age using a Dermavac device as described previously (Kiistala & Mustakallio, 1968). Biopsies were taken 2, 4, and 9 days after induction of the blister.

Normally healing experimental pig wounds. 3-cm long full-thickness incisional wounds were made, using a surgical steel scalpel, on the skin of an anesthesized piglet. The wounds were closed with Prolene sutures and, therefore, healed by primary intent. To ensure adequate sampling, three wounds were excised at each time point: 1, 3, 5, 7, 10 and 14 days after wounding. The samples were fixed in 10% formaldehyde, bisected in a plane perpendicular to the long axis of the wound, and embedded in paraffin.

Chronic human gastrointestinal ulcers. Specimens of duodenal ulcer (n=3), gastric ulcer (n=5), ulcerative colitis (n=12), Crohn's disease (ileum, n=7; colon, n=8), ischemic colitis (n=8), histologically normal colon (n=3) and ileum (n=4) were obtained from the Department of Pathology, University of Helsinki. All disease samples represented active, ulcerative phase of the disease and were from adult patients.

Experimental model for intestinal anastomoses. Animal studies were approved by the Regional Committee for Ethics in Animal Research and Administrative Board at Helsinki University Central Hospital. Adult male Wistar Rats (300g) were anesthesthetized with a single intramuscular injection of ketamine (40mg/kg Vetalar®). The jejunum was cut at laparotomy with a steel scalpel approximately 10 cm from the Ligamentum of Treitz. Jejunal anastomoses were performed with one layer of interrupted 6-0 polypropylene sutures (Prolene®, Ethicon, Norderstedt, Germany). After the surgery the animals had free access to food and water. Animals were sacrificed at 1, 3, 4, 7, 9 and 14 days by an overdose of pentobarbital. To ensure adequate sampling, anastomoses were obtained from two separate animals for each time point. Tissues were embedded in paraffin and cut at 5 µm.

The production and specificity of the anti-sense human collagenase-1, stromelysin-1, stromelysin-2, matrilysin, uPA, TIMP-1 and TIMP-3 (McDonnel et al, 1991; Busiek et al, 1992; Saarialho-Kere et al, 1992; Sudbeck et al, 1992; Saarialho-Kere et al, 1994; Airola et al, 1995; Airola et al, 1998) probes have been described. The MMP-13 cDNA plasmid MMP13HT3 (Johansson et al, 1997a) was linearized within the multiple cloning site with HindIII and EcoR1 to allow transcription of anti-sense (corresponding to nucleotides 1532-2042) and sense RNAs, respectively. The HME cDNA used as a template was a kind gift from Steven Shapiro (Pulmonary Department, Washington University, St Louis). The 650 bp fragment (600-1250) was generated by PCR using the primers CAT ACG ATT TAG GTG ACA CTA TAC and TAA TAC GAC TCA CTA TA, resulting in a product with SP 6 RNA polymerase recognition element at the 3'end and a T7 element at the 5'end. Both antisense and sense probes were transcribed from this polymerase chain reaction product. The specificity of the probes was confirmed by sequencing. The murine stromelysin-2 cDNA used as a was kindly provided by Sabine Werner (Max-Planck-Institut für Biochemie, Martinsried, Germany) (Madlener et al, 1996). As a control for nonspecific hybridization, sections in each experiment were hybridized with ³⁵S-labelled sense RNA from a bovine tropoelastin cDNA. The validity of this probe as a negative control has been confirmed by Northern and by in situ hybridization assays (Prosser et al, 1989; Saarialho-Kere et al, 1992; Saarialho-Kere et al, 1993b). In addition, negative control probes transcribed from collagenases-1 and -3, stromelysins-1, matrilysin, HME, uPA as well as TIMP-1 and -3 cDNAs in sense orientation were used.

Table 3. Probes and their sources

Probe (Study)	Transcribed from bases/ Genbank accession number	Source of the template
Collagenase-1 (I, II, III)	1-550 (550 bp)/ M13509	Dr Gregory Goldberg
Collagenase-3 (III, IV)	1532-2042 (511 bp)/NM_002427	Dr Carlos López-Otín
Stromelysin-1 (I, II, III)	1584-1801 (217 bp)/J03209	Dr Markku Kurkinen
Stromelysin-2 (II, IV)	1568-1743 (176 bp)/X07820	Dr Henning Birkedal-Hansen
Matrilysin (I, II)	14-813 (800 bp) /Z11887	Dr Lynn Matrisian
Macrophage metalloelastase (IV)	600-1250 (651 bp)/L23808	Dr Steven Shapiro
TIMP-1 (I, II, V)	1-313 (313 bp)/X03124	Dr David Carmichael
TIMP-3 (IV, V)	282-917 (636 bp)/U14394	Dr Veli-Matti Kähäri
UPA (II)	835-1661 (827 bp)/K03226	Dr Antti Vaheri
Stromelysin-2, murine (IV)	8-219 (212 bp)/X64020	Dr Sabine Werner

TIMP, tissue inhibitor of metalloproteinases; UPA, urokinase plasminogen activator

The cDNA probes for stromelysin-2 and HME were amplified by polymerase chain reaction (PCR). The template and primer concentrations were optimized for each probe. First, the reagents (primers, template, nucleotides, reaction buffer, and the DNA polymerase) were mixed on ice. The PCR was started with initial denaturing in 94 °C for 2,5 minutes. This was followed by 35 cycles consisting of 1. 30 sec of denaturing (94 °C) 2. 45 sec of annealing (55 °C) and 3. 45 sec of elongation (72 °C). The PCR-product was purified by chloroform-isoamylalcohol extraction and ethanol precipitation. The correct size of the PCR-product was checked by running it into an ethidium bromide-stained agarose gel, and by visualizing it under UV-light.

In situ hybridization was performed on 5- µm sections as described in detail (Prosser et al, 1989). All samples were treated with proteinase K and were washed in 0.1 M triethanolamine buffer containing 0.25% acetic anhydride. Sections were covered with 50 µl of hybridization buffer containing 2.5-5 x 10⁴ cpm/ µl of ³⁵S-labelled anti-sense or sense RNA probe. After hybridization at 50 °C to 55 °C for 18 hours in a humidified chamber, the slides were washed under stringent conditions, including treatment with RNAse A to remove unhybridized probe. Following 10 to 45 days of autoradiography, the photographic emulsion was developed, and slides were stained with hematoxylin and eosin. Samples previously positive for collagenase-1, stromelysin-1 and stromelysin-2 (cutaneous wounds), for matrilysin (normal skin), for collagenase-3, macrophage metalloelastase, TIMP-1 and -3 (breast and colon carcinomas) and for uPa (cutaneous wounds) were used as positive controls in each experiment.

Immunohistochemistry was performed on sections serial to those used for in situ hybridization. The peroxidase-antiperoxidase technique was applied using Vectastain Elite ABC kit (Vector Laboratories, Inc., Burlingame, CA). After deparaffinazation and dehydration, the endogenous proxidase activity was blocked with 0.3%-0.6% hydrogen peroxide. Non-specific staining was blocked by treatment with normal serum. The monoclonal primary antibodies were incubated for one hour at 37 °C, and the polyclonal primary antibodies overnight at 4 °C in a humidified chamber. After incubation with the primary antibody (see Table 3 for the antibodies used), the secondary biotinylated antibody was added, followed by avidin-biotin-peroxidase complex. Diaminobenzidine or aminoethylcarbazole (Ki-67, CD3, CD31) were used as chromogenic substrates and Harris hematoxylin as counterstain, as described in detail (Saarialho-Kere et al, 1993b). If necessary, sections were pre-treated with 10 mg/ml trypsin, 1 mg/ml protease type XXVII (Sigma Chelmical Co., St. Louis, MO), or antigen retrieval as described (von Boguslawski, 1994). Controls were performed with mouse preimmune ascites fluid or with rabbit pre-immune serum.

Combined immunohistochemistry and in situ hybridization was performed as described in detail (Ranki et al, 1995). After RNAse-free immunohistochemical staining with CD68 antibody, the samples were treated with 4% paraformaldehyde, washed with diethylpyrocarbonate-treated phosphate-buffered saline, and hybridized for collagenase mRNA with ³⁵S-labeled RNA as described (Prosser et al, 1989). After autoradiography, the slides were developed and stained with Mayer hematoxylin.

Table 4. Antibodies and their sources.

Normal human skin fibroblasts from a 26-year-old healthy volunteer were maintained in Dulbecco's modification of Eagle's medium (DMEM, Flow Laboratories, Irvine, Scotland, UK), 2 mM glutamine, 100 IU penicillin-G per ml, and 100 µl streptomycin per ml. Collagen gels were prepared with bovine dermal collagen, Cellon (Strassen, France), containing 95% type I collagen and 5% type III collagen. Eight volumes of Cellon were mixed with one volume of 10´ concentrated DMEM and 1 volume of NaOH (0.05 M) in 0.2 M N-(2-hydroxyethyl)piperazine-N'-(2-ethanesulfonic acid) and kept on ice. Cells were trypsinised, resuspended in DMEM supplemented with 10% fetal bovine serum, mixed into a neutralized Cellon solution, and transferred into six-well plates. The plates were incubated at 37 °C for 1 h for collagen polymerization. After this, DMEM supplemented with 10% fetal bovine serum was added and the gels were detached from the sides of the wells. The gels were incubated for 48 h before releasing the cells from the gels. As a control, fibroblasts were plated as monolayer and cultured in DMEM supplemented with 10 % fetal bovine serum for 48 h. Promonocytic U937 cells and CRL-1995 fibroblasts were cultured as previously described (Wilhelm et al, 1997; Saarialho-Kere et al, 1993c). Human alveolar macrophages were isolated from healthy adult smokers by saline bronchoalveolar lavage and cultured as described (Campbell et al, 1991).

To isolate total RNA from the dermal fibroblasts inside collagen gels, the gels were briefly treated with 0.5 mg collagenase (type II, Sigma) per ml in phosphate-buffered saline (pH 7.4) with 1 mM CaCl₂. Total cellular RNA was isolated from all cell types studied by using the single-step method (Chomczynski & Sacchi, 1987). Aliquots of total RNA (10-17 µg) were fractionated on 0.8% agarose gel containing 2.2M formaldehyde, transferred to a Zeta-Probe filter (Bio-Rad, Richmond, CA) by vacuum transfer (VacuGene XL; LKB, Bromma, Sweden), and immobilized by heating at 80 °C for 30 min. The filter was prehybridized for 2 h and subsequently hybridized for 20 h with ³²P labeled cDNAs (Thomas, 1980) for human collagenase-1 (Goldberg et al, 1986) and human collagenase-3 (Johansson et al, 1997a). The [³²P]cDNA-mRNA hybrids were visualized with autoradiography.