Arterial tissues from two subjects with and without a history of CVD were investigated by immune-histochemistry. Samples were collected 0.15-20 cm distal to the origin of the left descending coronary artery (LAD). The first subject (subject A) was a 77 yr female with an advanced atherosclerotic plaque (cross-sectional area of plaque size was 6.2 mm²). The second subject (subject B) was a 32 yr female with a minimal atherosclerotic lesion of 0.2 mm². Sample collection and handling was performed post-mortem in accordance with the guidelines of the medical ethical committee of the Wake Forest University Health Sciences as previously described[30].

For the clinical investigation, subjects were selected from a previously reported epidemiological study of 1356 elderly women aged 60 to 85 years examined at the Center for Clinical and Basic Research, Denmark[31]. Briefly, the participants in the present study were recruited by questionnaire surveys. In 2000 to 2001, these women were invited for a follow-up examination. All women gave informed consent to participation, and the study was carried out in accordance with the Helsinki Declaration II and the European Standards for Good Clinical Practice. The local ethics committee approved the study protocol. Participants in the present investigation were 102 women, including 51 women diagnosed with CVD and 51 age-matched controls. The CVD diagnosis was based on the presence of calcified deposits in the lumbar aorta visualized on lateral radiographs and graded with scores ranging from 0 to 24. Calcified deposits in the lumbar aorta adjacent to each lumbar vertebra (L1-L4) were assessed separately for the anterior and posterior wall of the aorta using the midpoint of the inter-vertebral space as the boundaries. Severity of aortic calcifications was described by 3 scores: the affected segments score (scale 0 to 4), the anterior and posterior affected score (scale 0 to 8), and the anteroposterior severity score (scale 0 to 24)[32]. Women graded with a score of 8 or more were classified as having CVD in the present investigation. Additional CVD risk factors were collected. Body weight and height were measured and BMI calculated. Arterial blood pressure was measured with a digital blood pressure monitor (UA-777, A&D Instruments LTD). Total cholesterol and triglyceride levels were measured by a Vitros-250 automatic blood analyzer (Johnson & Johnson). All serum samples were collected in the morning from fasted individuals. Demographic characteristics in the study population are indicated in Table 1.

Hearts were perfusion fixed at 100 mm Hg with formalin, coronary arteries were isolated, blocks were paraffin embedded, sectioned into 5 μM sections and mounted on glass. Pretreatment of sections in buffer 1 (10 mM Tris, 0.25 mM EDTA, pH 9.0) was used to enhance epitope presentation. The sections were blocked for nonspecific binding in Tris buffered saline (TBS) containing 5% casein before incubation with a monoclonal antibody to human CTX-I (in-house) or mouse anti-human Cathepsin K (Chemicon, MAB3324 directed against a synthetic peptide corresponding to amino acids 182-195 of human cathepsin K) or rabbit polyclonal antiserum LF-9 against Bovine alpha1(I) amino-propeptide of Collagen I with cross-reactivity to human, monkey pig and sheep (Larry W Fisher, Matrix Biochemistry Section, Cranofacial and skeletal diseases branch, NIDCR/NIH, Bethesda, Maryland). Peroxidase-labeled mouse or rabbit Envision (DAKO cytomation, DK) was used as secondary antibody. Immunoreactivity was visualized by liquid DAB substrate chromogen solution (3,30-diaminobenzidine chromogen). The nuclei were counterstained using hematoxylin (DAKO), and the slides were dehydrated, cleared and cover-slipped. Digital histographs were taken using an Olympus BX60 microscope and an Olympus CP71 camera.

Isoltion of CD14+ primary monocytes was performed using coated Dynabeads M-450 (111.11 Dynal Biotech). Human monocytes were isolated from peripheral blood from healthy female volunteers 18 to 45 years old. The blood was diluted 1:1 with PBS (BE17-512F Biowhittaker), carefully layered on a Ficoll-Paque gradient (17-1440-03 Amersham Pharmacia) and centrifuged at 2000 rpm for 20 minutes. The lymphocytes were collected from the interface between the plasma and the Ficoll-Paque, washed with ice-cold PBS followed by centrifugation at 2000 g rpm for 12 minutes. The wash was repeated twice, after which the cells were re-suspended in cold PBS containing 2% serum (S0415 Biochrom). The cells were kept on ice while the beads were prepared. 125 μl CD14+ magnetic beads (10 million beads) were washed 3 times in ice cold PBS. The beads were placed in the magnetic device for 2 minutes between each wash.

Three hundred million cells were added to the beads and incubated at 4°C with end-over-end homogenization for 20 minutes. Hereafter, the beads were washed 5 times in 5 ml PBS containing 2% serum with gentle re-suspension between each washing step. Finally, the cells were re-suspended in phenol-red free alpha-MEM containing 10% serum, 100 units/ml penicillin and 100 μg/ml streptomycin, as previously described[20, 33].

To generate human monocyte-macrophage cultures, the CD-14 positive cells were cultured in 75 cm² bottles in phenol-red free alpha-MEM containing 10% serum, 100 units/ml penicillin, 100 μg/ml streptomycin, 25 ng/ml M-CSF for 3 days. Cells were released using trypsin and replated on 48,8% Matrigel^®Basement Membrane Matrix (cat.no 40234, lot no. 900618) from Engelbreth-Holm-Swarm (EHS) mouse tumor rich in ECM proteins. Major component of the Matrigel is laminin, followed by collagen IV, heparin sulfate proteoglycan, entactin and nidogen. In our experiments, Matrigel was enriched with 41,5% type I collagen (3 mg/mL), 4,9% 10 × alpha-MEM and 4,9% reconstitutions buffer. After one day allowing the cells to incorporated into the matrix the cells were incubated in either absence (macrophages) or presence (foam cells) of 2 mg/ml LDL (L8292 Sigma-Aldrich) and M-CSF, and with or without selective cysteine protease inhibitor E64 (Calbiochem). Following the culture, cells were fixed in 4% formaldehyde, and stained for lipid droplets by oil red staining. Medium was collected for assessment of the presence of CTX-I indicative of cathepsin K mediated type I collagen degradation.

Serum samples were analyzed for degradation fragments of the C-telopeptides of type I collagen by the Serum CrossLaps ELISA (Nordic Bioscience, Denmark). Briefly, the Serum CrossLaps ELISA is based on monoclonal antibodies recognizing the isomerised amino acid sequence EKAHD-β-GGR specific for the C-telopeptide of type I collagen and D-β-G representing the isomerised bond between aspartate and glycine. The sandwich assay which detects cross-linked peptides (CTX-I), has been reported to have intra-assay and inter-assay coefficient of variation of 4.9% and 7.9%, respectively and is useful for the quantitative assessment of bone resorption[34].

Results shown are mean ± SEM. Significant differences between means of study groups were determined using the Student's two-tailed unpaired t-test. Differences were considered statistically significant if p < 0.05.

Left anterior descending (LAD) coronary arteries with different degrees of atherosclerotic lesions were assessed for the expression and localization of collagen type I, cathepsin K and the CTX-I cleavage product of type I collagen, generated by the action of cathepsin K (Figure 1). Immune-histochemical analysis revealed co-localization of collagen type I and cathepsin K in the advanced atherosclerotic lesions, especially at the plaque shoulders (Figure 1 D1 and 1 E1). In close proximity to the cathepsin staining there was a corresponding accumulation of CTX-I, presumably resulting from the cathepsin K activity on collagen type I. In the relatively unaffected coronary arteries there was little or no positive staining for the cathepsin K and CTX-I (Figure 1 A1, B2 and 1 C1).

Human monocyte-macrophage cultures were generated using CD-14 positive monocytes cultured on plastic in the presence of M-CSF or M-CSF in combination with LDL. When incubated in the presence of LDL, cells assumed a foam cell like appearance and were oil red positive as illustrated on Figure 2.A and 2.B.

To investigate whether these cells were able to degrade a collagen-rich ECM in part resembling that of the arterial wall, human monocyte-macrophages were generated and cultured on a matrigel (MG) enriched with 48.8% collagen type I. Collagen type I degradation by cathepsin K was measured by assessing the appearance of CTX-I in the conditioned medium after the culture period[23]. As presented in Figure 3A, macrophages (MF) alone did not generate CTX-I fragments above lowest detection limit of the assay. MG alone released to a lesser amount CTX-I fragments. In contrast, MF cultured on MG released 100-fold (p < 0.01) more cathepsin K generated collagen type I fragments (CTX-I), relative to control (matrigel alone). We also assessed the amount of CTX-I released when culturing monocytes in the presence of 2 mg/ml of LDL. Additionally, we also tested the levels of CTX-I released, when cells were cultured in the presence of selective cysteine protease inhibitor - E64. As illustrated in Figure 3B foam cells (FC) did not release any considerable amount of CTX-I when cultured on plastic. Matrigel (MG) released to a lesser amount CTX-I to the culture medium. FC cultured on matrigel released 30-fold (p < 0.01) increased levels of CTX-I relative to MG control. The cysteine protease inhibitor E64, resulted in a similar amount of CTX-I released as in MG control. All cell culture experiments have been validated for cell viability by Alamar blue (data not shown).

Women assigned to the CVD group based on radiographic evidence of aortic calcification had calcification scores between 9 and 24, with an average of 12.6 ± 0.5. This group had higher plasma triglyceride concentrations (p = 0.016) than age-matched control group with no aortic calcification did not differ in systolic blood pressure or total plasma cholesterol, (Table 1).

We assessed serum CTX-I levels in this population to determine if levels might be elevated in subjects with aortic calcification relative to those without, perhaps as evidence of extensive vascular degradation of type I collagen.

The level of CTX-I was found to be significantly lower in patients with CVD (0.35 ± 0.03 ng/ml) compared to the control group (0.46 ± 0.03 ng/ml), p = 0.014 as presented on Figure 4. Exclusion of individuals receiving anti-resorptive osteoporosis treatments such as bisphosphonates and hormone replacement therapies which lower CTX-I levels by approximately 50% eliminated these differences (0.40 ± 0.04 ng/ml patient group (n = 31), vs. 0.50 ± 0.03 ng/ml (n = 43) control group, p = 0.061). (Figure 4)