The effect of pyridoxine deficiency on lysyl oxidase activity in the chick.

The activity of lysyl oxidase, the enzyme responsible for the production of the precursor of the crosslinks of collagen and elastin, was measured in the aortas and epiphyseal cartilage of chicks raised on pyridoxine-deficient and control diets. Lysyl oxidase activity was diminished in both tissues in the deficient chicks, although the activity declined more slowly in the aorta. Deficient chicks given a large dose of pyridoxine 14 hr before sacrifice showed a marked increase in lysyl oxidase levels in both tissues. It is concluded that a pyridoxine derivative as well as copper are essential cofactors for lysyl oxidase.     

Lysyl oxidase activity in lungs of copper-deficient hamsters

Lysyl oxidase, the enzyme responsible for mediating crosslink formation in collagen and elastin, requires copper for its activity. In this study, lysyl oxidase activity and insoluble elastin content were unchanged in lungs from copper-deficient hamsters compared to controls. The lack of dramatic diminution in lysyl oxidase activity in animals who demonstrate significant structural alterations in the lung suggests that other mechanisms in addition to inhibition of crosslink formation are operative in this model.     

Regulation of lysyl oxidase, collagen, and connective tissue growth factor by TGF-beta1 and detection in human gingiva

Gingival overgrowth is characterized by excess extracellular matrix accumulation and elevated levels of cytokines, including transforming growth factor-beta1 (TGF-beta1). The functional relationships between altered cytokine levels and extracellular matrix accumulation have not been extensively investigated in gingival cells and tissues. Lysyl oxidase catalyzes the final known enzymatic step required for cross-linking collagen and elastin in the synthesis of a functional extracellular matrix. This study investigated the regulation by TGF-beta1 of lysyl oxidase and its collagen and elastin substrates in early passage human gingival fibroblasts. In addition, TGF-beta1 regulation of connective tissue growth factor (CTGF) was assessed in human gingival cells and tissues. The results show that TGF-beta1 increases lysyl oxidase enzyme activity and mRNA levels for lysyl oxidase and alpha-1-type I collagen, but not elastin, in a dose- and time-dependent manner. Maximal stimulation of lysyl oxidase activity and mRNA levels for both lysyl oxidase and collagen occurs after 48 hours of treatment of gingival fibroblastic cells with 400 pM of TGF-beta1. This study shows for the first time that CTGF mRNA and protein are strongly and rapidly induced by TGF-beta1 in human gingival fibroblasts. Exogenous addition of 1 to 50 ng/ml CTGF to gingival fibroblasts stimulates production of lysyl oxidase enzyme activity up to 1.5-fold after 48 hours, and 50 ng/ml CTGF stimulated insoluble collagen accumulation 1.5- to 2.0-fold after 4, 11, and 18 days of treatment. It is interesting to note that the addition of CTGF-blocking antibodies in the presence of TGF-beta did not block TGF-beta stimulation of collagen mRNA levels. Thus, although CTGF itself contributes to increased insoluble collagenous extracellular matrix accumulation, CTGF does not mediate all of the effects of TGF-beta1 on stimulation of collagen mRNA levels in human gingival fibroblasts. Immunohistochemistry studies of gingival overgrowth tissue samples indicate for the first time detectable levels of CTGF protein in Dilantin-induced hyperplasia tissues also positive for TGF-beta1. CTGF was not found in TGF-beta1-negative samples. In addition, extracellular lysyl oxidase protein was detected in vivo. Taken together, these studies support mostly independent roles for TGF-beta1 and CTGF in stimulating collagenous extracellular matrix accumulation in human gingival fibroblasts and tissues.     

Immunohistochemical localization of lysyl oxidase with monoclonal antibodies

Lysyl oxidase initiates the cross-linking of collagen and elastin by catalyzing the formation of the lysine-derived aldehyde. We cloned three hybridoma cell lines which secrete monoclonal antibodies to human lysyl oxidase. The localization of lysyl oxidase was investigated in various tissues and in cultured cells using an immunofluorescent antibody method. Antibodies showed a strong immunostaining in the aorta and dermal connective tissue suggesting a close relation to elastin and collagen. Fibroblasts, chondrocytes, and smooth muscle cells also yielded a marked positive immunoreaction as did a variety of nonfibroblastic cells such as endothelial cells, basal cells, biliary epithelial cells, and glomerular epithelial cells. In cultured cells, including human fibroblasts, an intense immunoreaction manifested as fine, filamentous structures in the cytoplasm. It is suggested that lysyl oxidase is associated with cytoskeletal protein.     

Inhibition of chick embryo lysyl oxidase by various lathyrogens and the antagonistic effect of pyridoxal.

Lysyl oxidase, which cross-links collagen and elastin, was obtained from chick embryo bone and cartilage and its substrate, elastin, from aorta. The enzyme was studied using an improved assay which enabled the stability of the substrate to be monitored. The enzyme was fully inhibited in vivo by beta-aminopropionitrile, semicarbazide, thiosemicarbazide and isoniazid and in vitro by beta-aminopropionitrile and semicarbazide but only partially by thiosemicarbazide and isoniazid. Penicillamine, which solubilizes collagen by labilizing Schiff base cross-links in vivo and which prevents stable cross-link formation in vitro indirectly by binding to aldehyde groups on collagen, was shown to have no direct inhibitory effect on lysyl oxidase in vivo or in vitro. Homocysteine, which also solubilizes collagen by a mechanism similar to penicillamine does not inhibit lysyl oxidase either in vivo or in vitro. Pyridoxal reversed the inhibition of lysyl oxidase by semicarbazide and isoniazid in vivo but was unable to reverse that produced by either beta-aminopropionitrile or thiosemicarbazide. These results can be explained by the presence of a sulphydryl group near the active site of lysyl oxidase, which can form a complex with the nitrile group on beta-aminopropionitrile or with the thiol group on thiosemicarbazide leading to irreversible inhibition.     

The static elastic properties and chemical composition of the rat aorta in spontaneously occurring and experimentally induced hypertension: the effect of an anti-hypertensive drug

The static elasticity and scleroprotein content of the aorta have been measured in 24 Okamoto spontaneously hypertensive rats aged 22-25 weeks, and 24 Wistars of the same age in which hypertension had been induced by nephrectomy and treated with a steroid. From the age of 4 weeks half the animals in each group were treated with a diuretic drug. By the age of 15 weeks caudal artery systolic blood pressure was significantly lower than control values in both drug-treated groups and remained so until death. Both types of hypertension were associated with larger diameter, thicker-walled and heavier aortas than those in the drug-treated animals. Vessels from Okamoto animals contained more collagen than those from the Wistars, although the collagen content was unchanged by drug treatment. Neither drug nor strain had any clear-cut affect on elastin content. In spite of these differences in wall thickness and chemical composition, values of the functional stiffness of the aorta measured over a wide range of pressure were similar in all four groups of animals. Using a simple model of the aorta in which elastin and collagen bear stress in parallel we find that the relationship between vessel composition and static incremental elastic modulus (structural stiffness) is similar in both models of hypertension and is not changed by drug treatment in spite of the consequent reduction in blood pressure.     

The static elastic properties and chemical composition of the rat aorta in spontaneously occurring and experimentally induced hypertension: the effect of an anti-hypertensive drug.

The static elasticity and scleroprotein content of the aorta have been measured in 24 Okamoto spontaneously hypertensive rats aged 22-25 weeks, and 24 Wistars of the same age in which hypertension had been induced by nephrectomy and treated with a steroid. From the age of 4 weeks half the animals in each group were treated with a diuretic drug. By the age of 15 weeks caudal artery systolic blood pressure was significantly lower than control values in both drug-treated groups and remained so until death. Both types of hypertension were associated with larger diameter, thicker-walled and heavier aortas than those in the drug-treated animals. Vessels from Okamoto animals contained more collagen than those from the Wistars, although the collagen content was unchanged by drug treatment. Neither drug nor strain had any clear-cut affect on elastin content. In spite of these differences in wall thickness and chemical composition, values of the functional stiffness of the aorta measured over a wide range of pressure were similar in all four groups of animals. Using a simple model of the aorta in which elastin and collagen bear stress in parallel we find that the relationship between vessel composition and static incremental elastic modulus (structural stiffness) is similar in both models of hypertension and is not changed by drug treatment in spite of the consequent reduction in blood pressure.     

Lysyl oxidase gene expression in the stromal reaction to in situ and invasive ductal breast carcinoma.

Lysyl oxidase is involved in the main pathway of collagen and elastin cross-linking: it has a role in the maturation of fibrillar matrix proteins in fibrosing processes and dictates their stability against metalloproteases. The stromal reaction patterns in ductal breast carcinoma are known to be morphologically varied. This has raised the hypothesis that there might be a differential expression of the lysyl oxidase gene as a function of stromal reaction pattern. The present study investigates this potential correlation and the role of matrix protein cross-linking in stromal differentiation. Lysyl oxidase was detected by immunohistochemistry and lysyl oxidase gene expression by in situ hybridization. Maximal expression was observed in myofibroblasts and myoepithelial cells around in situ tumors and in the reactive fibrosis facing the invasion front of infiltrating tumors. The lysyl oxidase substrates were observed in parallel, resulting in the stabilization of a scar-like peritumor barrier. In contrast, a lack of lysyl oxidase was associated with the loose or scirrhous stroma accompanying invading tumors; here, in situ hybridization revealed type I collagen synthesis, resulting in the deposition of non-cross-linked matrix proteins susceptible to degradation. The early development of a cross-linked matrix around ductal breast carcinoma suggests a possible bost defense mechanism, whereas the synchronous or late stromal reaction lacking lysyl oxidase favors tumor dispersion.     

Skin lysyl oxidase activity is not rate limiting for collagen crosslinking in the glucocorticoid-treated rat

Lysyl oxidase activity in the skin of rats receiving triamcinolone diacetate (12 mg/kg) for three consecutive days was decreased by sixty-four percent as compared to control values. A decrease of lysyl oxidase activity was observed twelve hours after the initial glucocorticoid injection. The decreased lysyl oxidase activity was accompanied by a forty-nine percent decrease of acetic acid extractable collagen. There was also a forty-two percent decrease in the alpha/beta ratio of the acetic acid soluble skin collagen of glucocorticoid-treated animals. These data indicate that although skin lysyl oxidase activity is decreased by glucocorticoid treatment, the crosslinking of acid extracted collagen as measured by the alpha/beta ratio and collagen solubility is increased. Accordingly lysyl oxidase activity is not rate limiting for collagen crosslink formation in the skins of rats treated with glucocorticoids.     

Properties and function of lysyl oxidase.

Lysyl oxidase catalyzes the oxidation of peptidyl lysine to alpha-aminoadipic-delta-semialdehyde, the precursor to the covalent crosslinkages that stabilize fibers of elastin and collagen. This enzyme contains both copper and a carbonyl cofactor consistent with an o-quinone. The proposed mechanism of action is derived from available kinetic and chemical data and also can account for mechanism-based inhibition of the enzyme by specific monoamines and diamines. Recent evidence for biosynthetic precursors and for the regulation of lysyl oxidase in fibrotic and malignant diseases is discussed.     

Effects of chlorella phospholipid on the aortic collagen and elastin metabolism and on the serum lipid content in rats with experimental arteriosclerosis

Rats treated with 44,800 IU of vitamin D₂ for 4 consecutive days were fed an atherogenic diet in the presence or absence of 1% chlorella phospholipid. Control rats received a basal diet and were administered olive oil. After 2 months the animals were killed and aortic prolyl hydroxylase, lysyl oxidase activity, collagen, elastin, and serum lipid levels were determined. Aortic prolyl hydroxylase activity was significantly decreased in rats receiving the atherogenic diet in the absence of chlorella phospholipid. The aortic collagen and elastin content was lower in rats additionally treated with chlorella phospholipid. Aortic lysyl oxidase activity was significantly decreased in all rats receiving the atherogenic diet. The serum cholesterol level was significantly higher in rats on the atherogenic diet, especially the absence of chlorella phospholipid supplementation. The findings suggest that the administration of chlorella phospholipid may stimulate the degradation of collagen and elastin in the aorta of rats fed an atherogenic diet and that the serum cholesterol lowering effect of chlorella phospholipid is not ascribable to thyroid functions. Furthermore, the results suggest that the aortic degradation rate of elastin was reduced by the atherogenic treatment.     

Collagen fibrillogenic activity in mouse mesenchymal tumor tissues and cells

Collagen fibrillogenic activity of mouse mesenchymal tumor cell lines m, Ht, and F was investigated. When nu/nu mice were subcutaneously inoculated with the cells, the production of argyllophilic fibers, hydroxyproline content, and lysyl oxidase activity were most prominent in the tumors produced by the m, and least in the tumors produced by the F. Prolyl hydroxylase activity was lowest in the m-cell tumors and highest in the F-cell tumors. Hydroxyprolien content and lysyl oxidase activity in the culture media correlated with the findings in the tumor tissues. However, prolyl hydroxylase activity in these cells did not correlate with hydroxyproline content or lysyl oxidase activity produced in the culture media. Thus, production of fibrous connective tissues is specific to each cell line, and the amount of collagen in the tumor tissues is significantly related to the amount of collagen secreted (r = 0.966) and lysyl oxidase activity synthesized by the tumor tissues (r = 1.000, P < 0.05). Lysyl oxidase activity is closely related to the amount of collagen in vitro (r = 0.970). These findings suggest that lysyl oxidase and not prolyl hydroxylase activity is an index of collagen fibrillogenesis and plays an important role in collagen deposition.     

Influence of endothelium in the in vitro vasorelaxant effect of magnesium on aortic basal tension in DOCA-salt hypertensive rat.

The objective of this study was to examine the influence of the endothelium on the extracellular magnesium induced relaxation of basal tension in isolated aortas from both mineralocorticoid-salt (DOCA-salt) hypertensive and control normotensive Sprague Dawley male rats. After incubation in magnesium-free physiological salt solution (PSS) (O mM magnesium), the increase of extracellular magnesium (1.2; 4.8 mM magnesium) caused a decrease in aortic tone which was significantly greater when endothelium was disrupted. Magnesium-induced relaxation was also more pronounced when endothelial NO production was blocked by 10(-4) M N omega-nitro-L arginine methyl ester (L-NAME). It is suggested that the vasorelaxation induced by extracellular magnesium is linked to the level of aortic basal tension developed in magnesium-free PSS. The endothelium does not seem to be directly implicated in magnesium-induced vasorelaxation in aortas from normotensive rats. However, in DOCA-salt hypertensive rats, the magnesium-induced relaxation of basal tension was less in the intact aorta (though not when the endothelium was disrupted) when the cyclo-oxygenase pathway was blocked by 10(-6) M indomethacin. These data therefore suggest that extracellular magnesium can promote relaxation by endothelium-dependent and cyclo-oxygenase-dependent mechanisms such as the production of relaxing prostacyclin in isolated aorta from DOCA-salt hypertensive rats.     

Renal and vascular oxidative stress and salt-sensitivity of arterial pressure

Oxidative stress occurs in a tissue or in the whole body when the total oxidant production exceeds the antioxidant capacity. Recent studies in human essential hypertension indicate that free radical production is increased and antioxidant levels are decreased, and more than one-half of these hypertensives have a salt-sensitive type of hypertension with progressive renal damage. Increased oxidative stress may also play a critical role in animal models of salt-sensitive hypertension. The stroke-prone spontaneously hypertensive rats (SHRSP) exhibits salt-sensitivity, vascular release of superoxide is increased, and total plasma antioxidant capacity is decreased. The superoxide release in the SHRSP rats inactivates nitric oxide, and superoxide dismutase (SOD) administration returns the bioactive nitric oxide levels to normal. The deoxycorticosterone acetate (DOCA)-salt hypertensive rat is salt-sensitive, aortic superoxide production is increased, and renal inflammation is significant. Treatment of the DOCA-salt rats with apocynin, an NADPH oxidase inhibitor, decreased aortic superoxide production and decreased arterial pressure. The Dahl salt-sensitive (S) rat has increased mesenteric microvascular and renal superoxide production and increased plasma levels of H2O2. The renal protein expression of SOD is decreased in the kidney of Dahl S rats, and long-term administration of Tempol, a superoxide mimetic, significantly decreased arterial pressure and renal damage. In conclusion, both human hypertension and experimental models of salt-sensitive hypertension have increased superoxide release, decreased antioxidant capacity and elevated renal damage.     

Effects of centrally administered losartan on deoxycorticosterone-salt hypertension rats

To investigate whether brain AT1 receptor stimulation contributes as a hypertensive mechanism to deoxycorticosterone acetate (DOCA)-salt hypertension, losartan (1 mg/4 microL) or artificial cerebrospinal fluid (aCSF) was injected into the lateral cerebral ventricle in conscious control uninephrectomized Wistar rats or rats with DOCA-salt for 2 or 4 weeks, and mean arterial pressure (MAP) and heart rates (HR) were recorded. In rats with DOCA-salt treatment, resting MAP increased to 144+/-6 mmHg after 2 weeks and to 170+/-5 mmHg after 4 weeks versus 115- 120 mmHg in controls. In rats with 2 week DOCA-salt treatment, MAP started declining at 4 hr after intracerebroventricular (icv) injection of losartan, and significant decreases in MAP were found at 18 and 24 hr. In rats with 4 week DOCA-salt treatment, MAP was significantly decreased at 4, 18 and 24 hr. In both groups MAP decreased to that of control rats. In control rats, icv losartan had no effect on MAP and HR. Icv aCSF did not significantly change MAP and HR in either DOCA-salt hypertensive rats or control rats. Normalization of MAP after icv administration of the AT1 receptor antagonist suggests a significant role for brain AT1 receptor stimulation in the development and maintenance of hypertension in the DOCA-salt hypertensive rat model.     

Alterations of the connective tissue components induced by β-aminopropionitrile

The ultrastructural and biochemical alterations induced by β-aminopropionitrile on aorta, lung, and skin of 7-day-old chicks have been studied. The inhibition of elastin formation by β-aminopropionitrile was associated with: (i) apposition of elastin on the old fiber in the form of button-like appendices; (ii) absence of microfibrils around this abnormally deposited elastin; (iii) presence of ruthenium red and toluidine blue O-positive material within the lathyritic elastin; (iv) increase of proteoglycan content; (v) increase of the mean diameter of collagen fibers; (vi) increased vesiculation of the endoplasmic reticulum of both fibroblasts and smooth muscle cells in aorta. The role of lysyl oxidase in the assembly of elastin and collagen fibers is discussed. Particular attention has been paid to the relationship between elastin, microfibrils, and proteoglycans in the formation of the elastic fiber.     

X-linked cutis laxa: defective cross-link formation in collagen due to decreased lysyl oxidase activity

We studied several members of a family with an X-linked form of cutis laxa; the affected males have mild skin laxity, a characteristic facies, skeletal abnormalities, structural abnormalities of the genitourinary tract, and low serum copper levels. The activity of lysyl oxidase, a copper-dependent enzyme involved in cross-link formation in collagen, was decreased in skin-biopsy specimens (13 to 26 per cent of normal) and in culture medium from cells to two affected males (15 to 20 per cent of normal). Immunoreactive lysyl oxidase from skin of both patients was virtually undetectable by immunodiffusion assay. The amounts of lysyl-derived aldehydes (the product formed in collagen and elastin by lysyl oxidase) and of cross-links formed from these products were decreased in dermal fibroblasts in culture. Collagen extractability from these cells was increased in culture. These findings suggest that lysyl oxidase deficiency provides the biochemical basis of the X-linked form of cutis laxa.     

Building Elastin. Incorporation of recombinant human tropoelastin into extracellular matrices using nonelastogenic rat-1 fibroblasts as a source for lysyl oxidase.

The purpose of this study was to assess the feasibility of crosslinking exogenously produced tropoelastin, the precursor of insoluble elastin, into existing elastin. Tritiated recombinant human tropoelastin (rhTE) was added to neonatal rat aorta smooth-muscle cell (NNRSMC) cultures. As much as 12% of the added rhTE was incorporated into the NNRSMC-derived insoluble elastin with the formation of the elastin crosslinks desmosine (DES) and isodesmosine (IDES) in a time-dependent fashion. The ratio of radioactivity found in DES and IDES crosslinks to that found in lysyl residues increased from 0.18 immediately after incubation with rhTE to 0.76 after 14 d. Crosslinking of rhTE into elastin and the accompanying formation of tritiated water was inhibited by beta-aminoproprionitrile, a potent inhibitor of lysyl oxidase, an enzyme critical for the post-translational processing of elastin and collagen. Acellular NNRSMC matrices were produced and replated with Rat-1 fibroblasts, cells that were found to express lysyl oxidase but not tropoelastin. At 14 d after incubation with rhTE, the ratio of DES and IDES radioactivity to that of lysine in the insoluble elastin was 0.38. We show for the first time that cells expressing lysyl oxidase, but not elastin, as well as elastogenic cells can incorporate rhTE into insoluble elastin with the formation of elastin crosslinks. This novel approach might be used to augment elastin repair in certain pathologic states.     

Alterations of elastin fibrogenesis by inhibition of the formation of desmosine crosslinks. Comparison between the effect of beta-aminopropionitrile (beta-APN) and penicillamine

Experimental lathyrism was induced by feeding newborn chicks a diet containing 0.2 and 0.4% DL-Penicillamine, with or without CuSO4 (10 mg/Kg diet) and Vitamin B6 (100 mg/Kg diet), or 0.015 and 0.1% beta-aminopropionitrile fumarate (beta-APN). After 7, 15, 25 and 55 days of treatment the animals were killed, the aortas removed and processed for electron microscopy in the presence of markers for proteoglycans, and the elastic fibers were carefully examined. Penicillamine, which prevents the formation of desmosine crosslinks by binding to precursors, induced the production of numerous new elastin fibers which appeared normal from the ultrastructural point of view. It seems, therefore, that at least in chick aortas, desmosine crosslinks are not necessary for the aggregation of tropoelastin molecules into structurally normal fibers. On the contrary, beta-APN, a classical inhibitor of lysyl oxidase, induced the tropoelastin molecules to aggregate into abnormal protuberances on the old fibers. Moreover, the elastin deposited during beta-APN treatment was always permeated by cytochemically revealed proteoglycans, which were never observed after penicillamine treatment. It is speculated that, at least in the system under study, the epsilon-amino groups of tropoelastin molecules may offer the binding sites for matrix proteoglycans until they are removed by lysyl oxidase, and that matrix proteoglycans might play a role in elastin fibrogenesis by preventing spontaneous tropoelastin aggregation in areas far from growing elastin fibers.     

Effect of DL-penicillamine on the aorta of growing chickens. Ultrastructural and biochemical studies.

The effect of DL-penicillamine on the architecture of the aortic wall of growing chickens was studied, with particular attention to elastin and collagen. Penicillamine was added to the diet (0.2% and 0.4%, in the presence or not of 10 mg/kg CuSO4 and 100 mg/kg vitamin B6) from hatching, for periods from 7 days up to 2 months. The same regions of the thoracic aortas were examined and compared in all the different experimental conditions. The results showed that penicillamine induced relevant modifications in the process of elastin fibrogenesis. The alterations consisted of an increase of elastin in the extracellular space, associated with an increase in the number of elastin fibers per unit area, and a decrease of the mean profile area of the fibers. Interestingly, penicillamine induced the formation of numerous bundles of microfibrils associated or not with elastin fibers. After prolonged treatment, elastin tended to diminish and the fibers tended to fuse into polymorphic syncytia. Collagen fibrils were larger, showed more heterogeneous cross diameters, were less numerous, and were more spread out within the tissue. All the other components of the aortic wall appeared not to be altered by the chemical. Penicillamine did not modify the copper content of chick aortas, whereas it induced a 40-50% reduction of the activity of both salt and 4 M urea-soluble peptidyl lysyl oxidases in the same tissue. These data may help in understanding some of the pathologic manifestations in human beings during D-penicillamine treatment.