Sharpin is a key regulator of skeletal homeostasis in a TNF-dependent manner.

OBJECTIVES: SHARPIN is a subunit of LUBAC and regulates activation of NF-κB, a pivotal transcription factor in skeletal homeostasis. Mutated SHARPIN gene (cpdm) mice develop chronic proliferative dermatitis and systemic inflammation. Cpdm mice have an osteopaenic phenotype characterised by decreased cortical and trabecular bone volume, but whether this is a consequence of the hyper-inflammatory phenotype is unknown. The inflammatory phenotype of cpdm mice is prevented by Tnf deficiency so we examined cpdm.Tnf (-/-) mice to examine the role of SHARPIN in skeletal development. METHODS: This research determined the extent to which SHARPIN and TNF interact within the skeleton through analyses of gene expression, µCT and biomechanical properties of bones of control (CTRL), cpdm, Tnf (-/-) (TNF KO) and cpdm.Tnf (-/-) (cpdm/TNF KO) mice. RESULTS: Gene expression of IL-1ß, TNF and caspase-3 increased in cpdm mice but was comparable to control values in cpdm/TNF KO mice. Decreased cortical and trabecular bone in cpdm mice translated to a loss in bone strength (ultimate stress and peak force). Cpdm/TNF KO mice developed bones similar to, or stronger than, control bones. CONCLUSIONS: Our results suggest that SHARPIN plays a significant role in skeletal homeostasis and that this role is strongly regulated through TNF pathways.

Galnon, a galanin receptor agonist, improves intrinsic cortical bone tissue properties but exacerbates bone loss in an ovariectomised rat model.

OBJECTIVES: Previous studies have shown galanin (GAL) injections onto mouse calvaria increased bone thickness and osteoblast number. This study investigated the effects of the GAL receptor agonist galnon on bone loss using the ovariectomised (OVX) rat model. METHODS: OVX rats were treated with either vehicle or galnon for 6 weeks via mini-osmotic pumps. Plasma osteocalcin concentrations, osseous cell gene expression, morphological and biomechanical properties of the skeleton were compared between the two groups. RESULTS: Treatment with galnon increased RANKL:OPG gene ratio (p<0.001) plus expression of TNF-α (p<0.05) and cathepsin K (p<0.05). µCT analyses revealed galnon-treated OVX animals had reduced trabecular and cortical morphology compared to control animals. Biomechanically, galnon OVX animals required similar peak force to failure to that of control OVX animals although galnon treatment did enhance the mechanical properties of Young's modulus and ultimate tensile stress. CONCLUSIONS: Our research suggests that galnon, a GAL receptor agonist, may enhance osteoclastic bone resorption in OVX rats. Although galnon reduced bone volume, biomechanical testing revealed that bone of galnon-treated animals was mechanically superior per unit area. Taken together, galnon simultaneously improves the intrinsic quality of cortical bone whilst stimulating osteoclastic activity in the OVX rat model.

Galanin treatment offsets the inhibition of bone formation and downregulates the increase in mouse calvarial expression of TNFalpha and GalR2 mRNA induced by chronic daily injections of an injurious vehicle.

We have previously shown that after bone fracture, galanin (GAL) and GAL receptor expression is increased in osteoblast-like cells of callus; however, the role of elevated GAL/GAL receptors in this instance of bone injury is not known. We hypothesize that in injury, GAL may facilitate bone formation by suppressing the production of cytokines such as TNFalpha and IL-1alpha, thereby affecting bone collagen formation and collagenolysis by key matrix metalloproteinases (MMPs). In studies to explore this hypothesis, we used a mouse calvarial injection model to (1) investigate whether mild injury caused by a daily subcutaneous injection of a glycerol-containing vehicle onto calvaria affected osteoblast/bone formation-associated histomorphometric parameters and gene expression (mRNA encoding GAL, GAL receptors, TNFalpha, IL-1beta, collagen type I, MMP-2 and -13) compared to non-injected, control mice and (2) determine the effect of GAL+vehicle treatment on these entities. Five groups of 4-week-old mice were used: a non-injected control group; a vehicle (50/50 solution of 10 mM PBS+0.025% BSA/5.4 M glycerol)-treated group; and 3 GAL-treated groups (0.2, 2 and 20 ng doses). Solutions were injected subcutaneously onto calvaria in a 10 mul volume, every day for 2 weeks. Vehicle injection reduced calvarial periosteal osteoblast cell height (P<0.001), osteoblast number (P<0.001) and osteoid thickness (P<0.01), relative to values in non-injected animals at 2 weeks. Vehicle injection also inhibited BFR in this periosteal bone relative to values in non-injected animals at both 1 and 2 weeks (P<0.05 and P<0.001, respectively). Increasing concentrations of GAL reversed the above-listed inhibitory effects caused by vehicle. This reversal was demonstrated by a dose-dependent effect of GAL on osteoblast cell height (Pearson's r=0.330; P<0.05), osteoblast number (Pearson's r=0.715; P=0.000), osteoid thickness (Pearson's r=0.516; P=0.000) and BFR (Pearson's r=0.525; P<0.05) after 2 weeks of GAL+vehicle treatment; with the 20 ng/day GAL+vehicle injection schedule returning these measured parameters toward non-injected control values. All GAL+vehicle treatments had no effect on calvarial expression of GAL, GALR1, GALR3, collagen type 1 and MMP-2 mRNAs compared to levels in vehicle-injected controls. GAL treatment did, however, produce dose-dependent effects on calvarial expression of GALR2 (Pearson's r=0.763; P=0.000), MMP-13 (Pearson's r=0.806; P=0.000), IL-1beta (Pearson's r=0.807; P=0.000) and TNFalpha (Pearson's r=0.542; P=0.000) mRNAs with 20 ng/day of GAL+vehicle producing the strongest reversal of vehicle-associated changes. Thus, the 20 ng/day GAL+vehicle regimen offset the inhibition of osteoblastic activity, and therefore bone formation caused by daily glycerol-containing vehicle injection. This effect on bone formation may be due in part to the peptide suppressing the formation and associated activity of TNFalpha, IL-1beta and MMP-13, as TNFalpha and IL-1beta are known inhibitors of bone formation and MMP-13 is involved in collagenolysis. Furthermore, these effects may be due to the action of GAL via GALR2, as it was the only GAL receptor affected by this GAL treatment regimen. These results indicate that GAL can facilitate bone formation associated with injury and reveal potential efficacy for GAL in treating osseous conditions where bone formation may be inhibited due to excess TNFalpha and IL-1beta production.

The effects of hypothyroidism on nerve growth factor and norepinephrine concentrations in weight-bearing and non-weight-bearing bones of rats.

Thyroid hormones affect bone remodelling directly via receptors in osteoblasts. Previously, however, we have shown that the euthyroid and hyperthyroid states significantly influence the concentrations of both nerve growth factor (NGF) and norepinephrine (NE) in particular bones. Both NGF and NE directly affect bone metabolism and therefore it is possible that thyroid hormone action on bone may also be indirect via its actions on these two neural-related substances. In light of previous studies, the current experiments aimed to investigate whether hypothyroidism also influenced NGF and NE concentrations in weight-bearing and non-weight-bearing rat bones. Hypothyroidism was induced by oral ingestion of propylthiouricil (PTU; 3.8+/-0.2 mg/kg/day) for 21 days. Histological examination on distal femurs and microparticle enzyme immunoassayed plasma concentrations of T3 and T4 verified the hypothyroid status in treated rats. NGF concentrations were assayed via enzyme-linked immunosorbent assay (ELISA) and NE concentrations were measured via high performance liquid chromatography (HPLC) with electrochemical detection (ECD). NGF concentrations: Femoral NGF concentrations were 207% higher in hypothyroid rats (674.9+/-88.3 ng/g) than in euthyroid rats (326.7+/-63.6 ng/g; p < 0.05). Rib NGF concentrations in hypothyroid rats (3125.1+/-450.2 ng/g) were increased by 342% compared to euthyroid ribs (914.5+/-128.6 ng/g; p < 0.01). Rib NGF concentrations in hypothyroid rats were 463% higher than in femurs of hypothyroid rats (p < 0.001). NE concentrations: In hypothyroid rats, NE concentrations were reduced by approximately 50% in both ribs (38.9 ng/g) and calvaria (41.5 ng/g) compared to euthyroid rats (74.7 ng/g and 87.4 ng/g respectively; p < 0.05 for both). These findings on hypothyroid rats may be taken in conjunction with our companion work on hyperthyroid rats (Yao et al., 2002, JMNI 2:327-334) and put in context with other reports, to indicate that (i) there are several sources of NGF in bone, some of which are stimulated by hypothyroidism and others by hyperthyroidism and (ii) the concentrations of both NGF and NE in bone are sensitive to weight-bearing and thyroid hormone status.

Nerve growth factor and norepinephrine concentrations in weight-bearing and non-weight-bearing bones of euthyroid and hyperthyroid rats.

UNLABELLED: Thyroid hormone, nerve growth factor (NGF) and norepinephrine (NE) and weight-bearing affect bone metabolism, yet interactions between these factors and osseous tissue have not been investigated. Therefore, the aims of the study were to measure NGF and NE concentrations in weight-bearing and non-weight-bearing bones from euthyroid (control) and hyperthyroid (HT) rats. Hyperthyroidism was induced by oral intake of triiodothyronine (90 mg/kg/day) for 21 days. Histomorphometry on distal femurs verified significant trabecular bone loss in HT rats compared to euthyroid animals. NGF concentrations were assayed via ELISA, whilst NE concentrations were measured via HPLC and ECD. In euthyroid rats: (i) the concentration of NGF in ribs (914 ng/g) was almost 3-fold greater than in femurs (326 ng/g wet weight of tissue) (ii) the concentrations of NE in ribs (74.7 ng/g) and calvaria (87.4 ng/g) were 2.5-3.5-fold greater than either femurs (24.0 ng/g) or tibiae (30.5 ng/g) and (iii) NE concentrations were comparable between ribs (74.7 ng/g) and calvaria (87.4 ng/g) and similar between tibiae (30.5 ng/g) and femurs (24.0 ng/g). In HT rats: (i) the concentration of NGF in ribs (1802 ng/g) was 4-fold greater than in femurs (402 ng/g) (ii) NE concentrations in ribs (23.3 ng/g) and calvaria (13.6 ng/g) were 4.5-fold and 2.6-fold greater respectively than in tibiae (5.2 ng/g), while ribs had almost a 2-fold higher concentration of NE than calvaria. In HT rats compared to euthyroid animals: (i) NGF concentrations almost doubled in ribs but there was little change in the NGF concentration in femurs (ii) there was a reduction in NE concentrations in calvaria by 84%, in ribs by 69% in tibiae by 83% and 55% in femur (NS). CONCLUSIONS: (i) Non-weight-bearing is associated with higher concentrations of NGF and NE than weight-bearing in bones in euthyroid and HT rats; (ii) Hyperthyroidism exerts opposite effects on NGF and NE in bone and (iii) Hyperthyroidism interacts with weight-bearing to determine NGF and NE concentrations in bone. Therefore, the influence of thyroid hormone on NGF and NE in bone may need to be taken into account when considering the action of thyroid hormone on bone in either euthyroid or hyperthyroid states.

Immunohistochemical localization of nerve growth factor in fractured and unfractured rat bone.

We detected nerve growth factor (NGF) by immunohistochemical localization in both fractured and unfractured rat rib. In unfractured bone, periosteal mesenchymal osteoprogenitor cells appeared to be the only skeletal cells which stained for NGF. Adjacent skeletal muscle fibers exhibited NGF staining both in fractured and unfractured bone. Fracture callus periosteal osteoprogenitor cells, marrow stromal cells, osteoblasts, young osteocytes and endothelial cells of new capillaries had moderate to heavy staining for NGF at 1 and 3 weeks after fracture. Deeply positioned osteocytes and osteoclasts showed no NGF staining. Most chondrocytes of fracture calluses stained for NGF, however, some chondrocytes did not stain which may indicate that NGF is produced at particular stages of chondrocytic differentiation. In calluses, periosteal matrix stained heavily for NGF when juxtaposed to cartilage and less obviously when associated with new bone at both 1 and 3 weeks post-fracture. However, other fibrous, cartilaginous and osseous matrices did not stain for NGF at any time. At 6 weeks post-fracture, NGF staining was largely confined to periosteal osteoprogenitor cells. The detection of NGF in periosteal osteoprogenitor cells of unfractured rib points to these cells having a role in nerve maintenance in intact bone. Furthermore, the localization of NGF in osteoprogenitor cells, marrow stromal cells, osteoblasts, certain chondrocytes, endothelial cells, periosteal matrix of the fracture callus and skeletal muscle may mean that these entities participate in fracture innervation. The presence of NGF in the callus may also indicate a direct, as yet undefined action of this neurotrophin on skeletal cell metabolism.

Topical application of nerve growth factor improves fracture healing in rats.

The aim of the present study was to examine the effects of nerve growth factor on the healing of unsplinted fractured ribs. After fracture of a rib in male rats, nerve growth factor was delivered by a miniosmotic pump to the fracture site for 7 days at the rate of 1.4 micrograms/day. Callus catecholamine concentrations, bone callus size, histomorphometry, and biomechanical properties of the repairing rib were measured at 7, 21, and 42 days after fracture. After 21 days, concentrations of norepinephrine and epinephrine were significantly increased in the group treated with nerve growth factor compared with those in the control group (211% norepinephrine and 322% epinephrine). Also, the midline longitudinal area of non-osseous (fibrous tissue and cartilage) callus of the fracture was significantly smaller (54%) and had a higher proportion of cartilage in the treated group than in the controls. By 42 days, there was only bony callus between the fracture ends in both the control group and the treated group. The treated group, however, again showed significantly elevated concentrations of norepinephrine and epinephrine (286 and 382%, respectively) and significantly elevated breaking stress (50%) and Young's modulus (51%), together with a reduction in the transverse cross-sectional area of the repair site (57%). The resultant increases in effectiveness and rate of repair of bone with administration of nerve growth factor suggest that it may play an important role in the healing processes of fractured bone.

Immunity to nerve growth factor and the effect on motor unit reinnervation in the rabbit.

The trophic effects of nerve growth factor (NGF) on sympathetic, peripheral afferent, and other neural crest-derived cells have been intensively investigated. More recently, NGF has been shown to have an influence on motoneurons. This study was undertaken to investigate whether NGF had any influence on the mechanical or histological properties of reinnervated motor units. Three groups of rabbits were used: normal rabbits, rabbits in which the nerve to medial gastrocnemius (MG) was cut and allowed to reinnervate for 56 days, and rabbits in which the MG nerve reinnervated in the presence of immunity to NGF. Immunity to NGF did not affect the ability of motor axons to reinnervate a muscle, nor were the contractile characteristics of the motor units altered. The size of horseradish peroxidase-labeled motoneurons was not influenced by immunization against NGF; however, the distribution of afferent neuron sizes was altered. Conduction velocity of motor axons proximal to the neuroma was significantly faster after immunization against NGF. Transection and subsequent reinnervation by a peripheral nerve normally causes an increase in myelin thickness proximal to the neuroma. However, immunization against NGF appeared to decrease the magnitude of myelin thickening. It was concluded that immunization against NGF affects motor axonal conduction velocity via an influence on the neural crest-derived Schwann cells.

Effect of thyroidectomy on cardiovascular responses to hypoxia and tyramine infusion in fetal sheep.

Fetal sheep were thyroidectomized at 80 days' gestation and reoperated at 118-122 days for insertion of vascular catheters. The effects of hypoxaemia and intravenous tyramine infusion on plasma catecholamine concentrations, blood pressure and heart rate were then determined in experiments at 125-135 days' gestation. Age matched intact fetuses were also studied. Thyroidectomy was associated with increased concentrations of noradrenaline, adrenaline and dopamine in some thoracic and abdominal organs, increased noradrenaline concentrations in the cerebellum, and decreased adrenaline concentrations in the hypothalamus, cervical spinal cord, and superior cervical and inferior mesenteric ganglia. Arterial pressure was significantly lower in the thyroidectomized fetuses (34.0 +/- 0.15 mmHg) than in intact fetuses (44.7 +/- 0.2 mmHg; p less than 0.001). In contrast, plasma noradrenaline concentrations were significantly higher in the thyroidectomized fetuses (2.04 +/- 0.25 ng/ml) compared to the intact fetuses (0.99 +/- 0.08 ng/ml; P less than 0.001). In the intact fetuses there was a significant increase in plasma noradrenaline concentration and blood pressure during hypoxaemia, and bradycardia at the onset of hypoxaemia. In contrast, in the thyroidectomized fetuses hypoxaemia did not cause significant change in plasma catecholamine concentrations, blood pressure or heart rate. Infusion of tyramine produced a 1.9-fold increase of plasma noradrenaline in thyroidectomized fetuses compared to a 9.2-fold increase in the intact fetuses (P less than 0.05). Tyramine infusion caused a similar proportional increase of blood pressure in both thyroidectomized and intact fetuses. Heart rate decreased during the tyramine-induced hypertension in the intact fetus, but increased in the thyroidectomized fetuses.(ABSTRACT TRUNCATED AT 250 WORDS)

Peripheral and brain tissue catecholamine content in intact and anti-NGF-treated fetal sheep.

Fetal lambs were treated with a single dose of anti-mouse nerve growth factor (anti-NGF) at 80 days gestational age. The catecholamine content of tissues was determined at 135 days gestational age. There was a reduction of either norepinephrine, epinephrine, or both, in the thymus, thyroid, atrium (but not ventricle), lung, liver, kidney, and jejunum when compared with age-matched control fetuses. The spleen, ileum, colon, and the adrenal glands were not affected by anti-NGF. In treated fetuses there was a reduction in catecholamine content of the thalamus, hypothalamus, hippocampus, medulla, cerebellum, and cervical spinal cord. These results show that some tissues are sensitive to, and some are refractory to, the action of anti-NGF at 80 days gestation. Also the results suggest that NGF may play a role in the development of catecholamine-containing neurons within the central nervous system.

Effect of hypoxemia on plasma catecholamines in intact and immunosympathectomized fetal lambs.

Fetal lambs treated with sheep anti-mouse nerve growth factor antibodies (anti-NGF) at 80 days gestation subsequently showed a diminished cardiovascular response to intravenous infusion of tyramine (1 mg/min over 10 min) and no significant change in plasma norepinephrine concentrations as measured by reversed-phase high-pressure liquid chromatography and electrochemical detection. At autopsy at 135 days gestation, catecholamine content of the heart, thyroid, kidney, and ileum was reduced by greater than 70% compared with age-matched control fetuses. The anti-NGF-treated fetuses were thus judged to have impaired sympathetic function. When made hypoxemic (arterial PO2 12-14 Torr) for 1 h at 120 or 130 days gestation, anti-NGF-treated fetuses showed no significant change of arterial pressure, heart rate, or plasma catecholamines, whereas control fetuses showed bradycardia, a 28% increase in arterial pressure, and a 670% increase of plasma norepinephrine concentrations. These results suggest that the increase of arterial pressure that occurs during hypoxemia in fetal lambs between 120 and 135 days gestation is attributable to increased activity of the peripheral sympathetic nervous system. The adrenal medulla appears to contribute to the plasma catecholamine pool at rest, but the lack of increased plasma catecholamines during hypoxia after anti-NGF implies that the adrenal medulla was unable to release catecholamines. Possible reasons for this are discussed.