Identification of cell types responsible for bone resorption in rheumatoid arthritis and juvenile rheumatoid arthritis.

Focal resorption of bone at the bone-pannus interface is common in rheumatoid arthritis (RA) and juvenile rheumatoid arthritis (JRA) and can result in significant morbidity. However, the specific cellular and hormonal mechanisms involved in this process are not well established. We examined tissue sections from areas of bone erosion in patients with RA and JRA. Multinucleated cells (MNCs) were present in resorption lacunae in areas of calcified cartilage and in subchondral bone immediately adjacent to calcified cartilage, as previously described. mRNA for the calcitonin receptor (CTR) was localized to these MNCs in bone resorption lacunae, a finding that definitively identifies these cells as osteoclasts. These MNCs were also positive for tartrate-resistant acid phosphatase (TRAP) mRNA and TRAP enzymatic activity. Occasional mononuclear cells on the bone surface were also CTR positive. Mononuclear cells and MNCs not on bone surfaces were CTR negative. The restriction of CTR-positive cells to the surface of mineralized tissues suggests that bone and/or calcified cartilage provide signals that are critical for the differentiation of hematopoietic osteoclast precursors to fully differentiated osteoclasts. Some MNCs and mononuclear cells off bone and within invading tissues were TRAP positive. These cells likely represent the precursors of the CTR-TRAP-positive cells on bone. Parathyroid hormone receptor mRNA was present in cells with the phenotypic appearance of osteoblasts, in close proximity to MNCs, and in occasional cells within pannus tissue, but not in the MNCs in bone resorption lacunae. These findings demonstrate that osteoclasts within the rheumatoid lesion do not express parathyroid hormone receptor. In conclusion, the resorbing cells in RA exhibit a definitive osteoclastic phenotype, suggesting that pharmacological agents that inhibit osteoclast recruitment or activity are rational targets for blocking focal bone erosion in patients with RA and JRA.

Expression of two human skeletal calcitonin receptor isoforms cloned from a giant cell tumor of bone. The first intracellular domain modulates ligand binding and signal transduction.

Two distinct calcitonin (CT) receptor (CTR)-encoding cDNAs (designated GC-2 and GC-10) were cloned and characterized from giant cell tumor of bone (GCT). Both GC-2 and GC-10 differ structurally from the human ovarian cell CTR (o-hCTR) that we cloned previously, but differ from each other only by the presence (GC-10) or absence (GC-2) of a predicted 16-amino acid insert in the putative first intracellular domain. Expression of all three CTR isoforms in COS cells demonstrated that GC-2 has a lower binding affinity for salmon (s) CT (Kd approximately 15 nM) than GC-10 or o-hCTR (Kd approximately 1.5 nM). Maximal stimulatory concentrations of CT resulted in a mean accumulation of cAMP in GC-2 transfected cells that was greater than eight times higher than in cells transfected with GC-10 after normalizing for the number of receptor-expressing cells. The marked difference in maximal cAMP response was also apparent after normalizing for receptor number. GC-2 also demonstrated a more potent ligand-mediated cAMP response compared with GC-10 for both human (h) and sCT (the EC50 values for GC-2 were approximately 0.2 nM for sCT and approximately 2 nM for hCT; EC50 values for GC-10 were approximately 6 nM for sCT and approximately 25 nM for hCT). Reverse transcriptase PCR of GCT RNA indicated that GC-2 transcripts are more abundant than those encoding for GC-10. In situ hybridization on GCT tissue sections demonstrated CTR mRNA expression in osteoclast-like cells. We localized the human CTR gene to chromosome 7 in band q22. The distinct functional characteristics of GC-2 and GC-10, which differ in structure only in the first intracellular domain, indicate that the first intracellular domain of the CTR plays a previously unidentified role in modulating ligand binding and signal transduction via the G protein/adenylate cyclase system.

Cloning and characterization of a mouse brain calcitonin receptor complementary deoxyribonucleic acid and mapping of the calcitonin receptor gene.

We have identified and characterized a mouse brain calcitonin receptor (CTR) complementary DNA (cDNA). This cDNA encodes a receptor protein that, after expression, has high affinity binding for salmon calcitonin (Kd approximately, 12.5 nM) and is coupled to adenylate cyclase. The binding affinity of this expressed receptor for salmon calcitonin is lower than that described for the previously cloned porcine renal and human ovarian CTRs, but is similar to that of the recently described rat brain CTR, designated the C1b form of the receptor. Analysis of the deduced structure of the mouse brain CTR reveals that it is highly related to the other CTR cDNAs that belong to a distinct family of G-protein-coupled receptors with seven transmembrane-spanning domains. The major structural feature that distinguishes the mouse cDNA clone from the other CTRs is the presence of a consecutive 111-basepair nucleotide sequence that encodes a 37-amino acid sequence which is predicted to localize to the first extracellular loop between the second and third transmembrane-spanning domains. We have mapped the CTR gene in the mouse to the proximal region of chromosome 6, which is homologous to the 7q region of human chromosome 7; only a single CTR gene was identified. Preliminary analysis of the mouse CTR gene reveals that it is complex, consisting of multiple exons separated by lengthy introns that would allow for splice variants consistent with the existence of multiple CTR isoforms predicted from the CTR cDNA clones. The differential cellular and tissue distribution of these functionally distinct CTR isoforms provides the molecular basis for the previously reported widespread distribution and functional heterogeneity of the CTR.

Characterization of the structural and functional properties of cloned calcitonin receptor cDNAs.

We have recently cloned CTRs from cDNA libraries prepared from porcine renal and human ovarian cell lines. In situ hybridization and Northern analysis confirm the widespread distribution of CTR mRNA in numerous tissues. Hydropathy plots of the predicted amino acid sequence of the receptors demonstrate multiple hydrophobic regions that could generate 7 transmembrane spanning domains, similar to other G protein-coupled receptors. Searches of databanks for proteins with related amino acid sequences reveals that the CTRs are closely related to the receptors for parathyroid hormone/parathyroid hormone related peptide, secretin, vasoactive intestinal peptide, growth hormone releasing hormone, glucagon-like peptide-1 and glucagon. These receptors have no significant sequence homology to other G protein-coupled receptors, and therefore, appear to comprise a distinct receptor family. Expression of the hCTR or pCTR in COS cells results in expression of high affinity CTRs which are coupled to adenylate cyclase (AC). The hCTR, however, demonstrates higher affinity for human and salmon CT compared to the pCTR. Both CTRs demonstrate low affinity binding and AC activation in response to calcitonin gene related peptide, amylin or secretin, providing a possible explanation for the cross-reactivity among these peptides in vivo. Stable transfectants expressing the pCTR increase cAMP levels and increases in cytosolic free Ca2+ concentration consistent with dual coupling to AC and phospholipase C. Additional studies will help to establish the structural basis for this functional property as well as the evolutionary relationship of the members of this newly identified family of receptors.

Cloning, characterization, and expression of a human calcitonin receptor from an ovarian carcinoma cell line.

A human ovarian small cell carcinoma line (BIN-67) expresses abundant calcitonin (CT) receptors (CTR) (143,000 per cell) that are coupled, to adenylate cyclase. The dissociation constants (Kd) for the CTRs on these BIN-67 cells is approximately 0.42 nM for salmon CT and approximately 4.6 nM for human CT. To clone a human CTR (hCTR), a BIN-67 cDNA library was screened using a cDNA probe from a porcine renal CTR (pCTR) that we recently cloned. One positive clone of 3,588 bp was identified. Transfection of this cDNA into COS cells resulted in expression of receptors with high affinity for salmon CT (Kd = approximately 0.44 nM) and for human CT (Kd = approximately 5.4 nM). The expressed hCTR was coupled to adenylate cyclase. Northern analysis with the hCTR cDNA probe indicated a single transcript of approximately 4.2 kb. The cloned cDNA encodes a putative peptide of 490 amino acids with seven potential transmembrane domains. The amino acid sequence of the hCTR is 73% identical to the pCTR, although the hCTR contains an insert of 16 amino acids between transmembrane domain I and II. The structural differences may account for observed differences in binding affinity between the porcine renal and human ovarian CTRs. The CTRs are closely related to the receptors for parathyroid hormone-parathyroid hormone-related peptide and secretin; these receptors comprise a distinct family of G protein-coupled seven transmembrane domain receptors. Interestingly, the hCTR sequence is remotely related to the cAMP receptor of Dictyostelium discoideum (21% identical), but is not significantly related to other G protein-coupled receptor sequences now in the data bases.

A cloned porcine renal calcitonin receptor couples to adenylyl cyclase and phospholipase C.

The signal transduction pathways of the recently cloned porcine kidney calcitonin (CT) receptor were evaluated. This receptor, when stably transfected into MC-3T3 cells, avidly bound salmon CT (SCT) [dissociation constant (Kd) = 4 nM]. Incubation with SCT resulted in a dose-dependent accumulation of adenosine 3',5'-cyclic monophosphate (cAMP) [50% effective concentration (EC50) = 0.02 nM] in transfected cells (referred to as PC-1 cells). Binding kinetics and cAMP dose response relationships were similar to those of the native receptor in LLC-PK1 cells. PC-1 cells also responded to calcitonin gene-related peptide (CGRP), but the EC50 value for cAMP accumulation was more than three orders of magnitude higher than for SCT. Exposure of PC-1 cells to SCT (5 nM to 1 microM) produced a dose-dependent rise in cytosolic free Ca2+ concentration ([Ca2+]i), whereas CGRP did not. The initial rise in [Ca2+]i was not dependent on extracellular Ca2+, suggesting that SCT induced release of Ca2+ from intracellular stores. SCT also increased inositol trisphosphate production in PC-1 cells. In conclusion, the cloned, transfected porcine CT receptor functionally couples to and activates both adenylyl cyclase and phospholipase C. This dual coupling is also a characteristic of the parathyroid hormone receptor, which has significant homology in amino acid sequence with the CT receptor.

Calcitonin stimulates bone formation when administered prior to initiation of osteogenesis.

The influence of calcitonin (CT) on various stages of bone formation was investigated. A demineralized collagenous bone matrix-induced bone forming system in rats was used to temporally segregate chondrogenesis and osteogenesis. Administration of CT (15 Medical Research Council Units [MRCU]) daily) at the initiation of matrix-induced bone formation (BF) resulted in a 76% stimulation of BF as measured by 45Ca incorporation and alkaline phosphatase activity. This increase was due, in part, to a stimulation of cartilage and bone precursor cell proliferation monitored by the rate of [3H]thymidine incorporation and ornithine decarboxylase activity. Chondrogenesis on day 7 as measured by 35SO4 incorporation was increased by 52% with CT treatment. To rule out the possibility of a secondary response due to parathyroid hormone, similar studies were done in parathyroidectomized animals and CT stimulation of BF was still observed. However, when CT injections were started after cartilage formation (day 8) there was no stimulation of BF but a significant decrease in 45Ca incorporation was observed. These results indicate CT has two actions: (a) when CT is administered during the initial phases of bone formation, it increases BF due to a stimulation of proliferation of cartilage and bone precursor cells; and (b) when CT is administered after bone formation has been initiated, subsequent bone formation is suppressed.

Abnormalities in the biosynthesis of cartilage and bone proteoglycans in experimental diabetes.

Proteoglycans synthesized in developing cartilage and bone were investigated in control and streptozotocin-induced (65 mg/kg, i.v.) diabetic rats. Ten days after streptozotocin injection, animals were implanted subcutaneously with demineralized bone matrix particles. This system induces formation of cartilage and bone on days 7 and 14, respectively. Two hours before they were killed, animals were injected with 35SO4 and the labeled proteoglycans were extracted from the explants and metaphyses by either a direct associative extraction (0.5 M GuCl2) or a direct dissociative extraction (4.0 M GuCl2). These procedures extract 80-90% of the total counts incorporated. To characterize the proteoglycans, extracts were subjected to cesium chloride density gradient centrifugation and molecular sieve chromatography. These data showed that (1) there is less proteoglycan made in diabetic bone; (2) the proteoglycan aggregate is of a smaller molecular weight in bone than in cartilage; (3) 10% of the proteoglycan synthesized in diabetic bone was in the form of aggregates compared with 48% of the control bone; (4) aggregates did form in the diabetic cartilage, and their molecular weight was smaller than in normal cartilage. This investigation shows that proteoglycans, structurally important macromolecules of cartilage and bone, are altered in experimental diabetes. This metabolic abnormality may be an important factor contributing to decreased bone formation observed in diabetes.