Inhibition of prolyl 4-hydroxylase in vitro and in vivo by members of a novel series of phenanthrolinones.

Examples of a novel series of phenanthrolinones are shown to be potent competitive inhibitors of avian prolyl 4-hydroxylase, and of collagen hydroxylation, in embryonic chick tendon cells and human foreskin fibroblasts in vitro and in the oestradiol-stimulated rat uterus in vivo. Two compounds, Compound 1 (1,4-dihydrophenanthrolin-4-one-3-carboxylic acid) and Compound 5 [8-(N-butyl-N-ethylcarbamoyl)-1,4-dihydrophenathrolin-4-one-3-carboxylic acid], with comparable potencies in vivo, were chosen to investigate the effect of the inhibition of the hydroxylation of newly synthesized uterine collagen on the turnover of this protein in vivo. Inhibition of hydroxylation by more than 50% for approx. 8 h following single oral doses of the compounds was associated with significant losses of radiolabelled proline and 4-hydroxyproline from collagen during this period. Progressive hydroxylation of collagen over 48 h, as the inhibitory action of the compounds declined, was accompanied by a decreased loss of radiolabel from the uterine collagen. Earlier reports indicated that underhydroxylated collagen, accumulating within the endoplasmic reticulum in cells where prolyl 4-hydroxylase is inactivated, is slowly degraded, but is then rapidly hydroxylated and secreted when the activity of prolyl 4-hydroxylase is restored. Taken with the present results, this suggests that the potential use of inhibitors of prolyl 4-hydroxylase to control excessive collagen deposition in pathological fibrosis may be limited by the need to maintain continuous inhibition of collagen hydroxylation so as to facilitate intracellular degradation of the accumulated protein.

A novel series of non-nucleoside inhibitors of inosine 5'-monophosphate dehydrogenase with immunosuppressive activity.

Inhibitors of inosine 5'-monophosphate dehydrogenase (IMPDH, EC 1.1.1.205) are effective immunosuppressive drugs that may also have additional potential applications as antitumour and antimicrobial agents. The clinical value of the most potent and specific inhibitor of IMPDH, mycophenolic acid, is limited by its rapid metabolism in vivo to an inactive glucuronide derivative. There is, therefore, a considerable incentive to develop structurally novel, preferably non-nucleoside, inhibitors with greater metabolic stability than mycophenolic acid. Here, we describe a high throughput screen for inhibitors of IMPDH, which facilitated the discovery of a single novel non-nucleoside inhibitor from a collection of approximately 80,000 compounds. The inhibitor is a pyridazine, which, like mycophenolic acid, exerts uncompetitive inhibition of IMPDH. Analysis of the enzyme kinetics suggests that the inhibitory action of the pyridazine is similar to that of mycophenolic acid, which involves trapping of a covalent intermediate formed during the conversion of IMP to xanthosine monophosphate. Chemical modification of the lead compound resulted in pyridazine derivatives with enhanced potency against IMPDH and guanine nucleotide synthesis in cultured cells in vitro and also against guanine nucleotide synthesis in the mouse spleen in vivo. One of the compounds was available in sufficient quantity to demonstrate highly effective immunosuppressive activity in a model of delayed type hypersensitivity in mice. To our knowledge, the novel pyridazines described in this report represent the first non-nucleoside uncompetitive inhibitors of IMPDH with immunosuppressive activity since the discovery of the inhibitory activity of mycophenolic acid and its derivatives thirty years ago.

Human colorectal carcinoma cells in vitro as a means to assess the metabolism of analogs of mycophenolic acid.

Cultures of the human colorectal carcinoma line, HT29, were used to assess the susceptibility to glucuronidation of the cytostatic, immunosuppressive drug mycophenolic acid (MPA) and 19 of its analogs. Removal of the metabolically vulnerable 7-hydroxyl group or its replacement by a fluorine atom, amino group, or nitrile group resulted in compounds that were completely resistant to metabolism, but that had substantially lower antiproliferative potency against the EMT6 carcinoma line that is unable to glucuronidate MPA. In compounds retaining the 7-hydroxy function replacement of the lactone moiety of the phthalane ring of MPA by either a cyclopentanone or a 6-membered lactam afforded some protection against metabolism, but also partially or completely suppressed antiproliferative activity. Some lipophilic substituents at position 2 of the hexenoic side chain in analogs with the 7-hydroxy function resulted in increased metabolism, whereas several substituents with increased steric bulk in this position (including benzyl, p-hydroxyphenyl, trifluoroacetamidophenyl,S-methyl, and methoxymethyl) markedly inhibited metabolism. The last three of these derivatives also maintained or exceeded the antiproliferative potency of MPA. We suggest that cultures of human colorectal carcinoma cells lines may provide a rapid and convenient means of assessing the susceptibility of novel synthetic compounds to both phase I and phase II metabolic conversions.

Therapeutic approaches to organ fibrosis.

Scarring of superficial tissues and chronic fibroses of major organs present major medical problems ranging from disfigurement to progressive disability and death. Growing understanding of the cellular and molecular events, which are common to these intractable disorders, now provides a favourable basis for the development of effective drug therapies. Much attention is focussed on the roles of the many cytokines and growth factors, which contribute to the fibrogenic process. The transforming growth factor (TGF)-beta 1 and 2 isoforms are among the most significant of these and approaches to control their activity include blocking the activation of latent TGF-beta, preventing the ligand-receptor interactions and the inhibition of down-stream signal transduction. Concerns regarding possible risks of the long-term suppression of TGF-beta function point to connective tissue growth factor (CTGF) as a possible alternative target. CTGF is induced by and appears to mediate at least some of the fibrogenic actions of TGF-beta, although not its important antimitogenic activity on epithelial cells. The fibrogenic effects of endothelins and angiotensin II have aroused considerable interest in the anti-fibrotic potential of antihypertensive agents designed primarily to limit the vasoconstrictive activities of these peptides. Polypeptides including interferons alpha and gamma, relaxin, TGF-beta 3 and hepatocyte growth factor, all show an ability to limit fibrogenesis in either clinical or experimental situations. Finally, inhibitors of the enzymes required for the post-translational processing of collagens, including prolyl 4-hydroxylase, C-proteinase and lysyl oxidase provide a more direct means of reducing the deposition of fibrillar collagens into the extracellular matrix although the potentially adverse effects of sustained manipulation of collagen metabolism remain to be investigated.

Cell cycle specific induction of HL-60 cell differentiation and apoptosis by mycophenolic acid.

HL-60 cells undergo terminal differentiation and apoptosis in response to different types of sub-toxic and toxic perturbations respectively. The mechanism by which cells sense different amounts of perturbation to activate pathways that lead to the engagement of a relevant biological response is not known. The response of HL-60 cells to treatment with the immunosuppressant mycophenolic acid (MPA), a specific inhibitor of dGTP/GTP-synthesis, allowed quantitation of a metabolic perturbation which triggered a cellular response. 1.5 microM MPA induced 38% terminal differentiation to CD14 positive, early monocyte-like cells and 22% cell death by apoptosis, whereas 3 microM MPA induced 70% apoptosis but no differentiation. Despite the difference in biological outcomes, 72 h exposure to both 1.5 microM and 3 microM MPA caused a similar ( approximately 75%) depletion of total GTP levels. Cells synchronized by centrifugal elutriation were treated with MPA. Elutriated cells were overall less sensitive to the effects of MPA but 3 microM MPA induced significantly less apoptosis and more differentiation in an elutriation-enriched G1-population than in a population normally distributed in the cell cycle, suggesting that the effects of MPA in S-phase may subsequently lead to cell death. However, analysis of apoptosis by using a terminal deoxynucleotidyltransferase assay and measurement of bromodeoxyuridine incorporation showed that apoptosis was engaged in G1. Analysis of the phosphorylation status of the retinoblastoma protein demonstrated that Rb was hypophosphorylated prior to apoptosis and that in apoptotic cells, separated by flow cytometry, Rb protein was absent, presumably due to proteolysis. The loss of Rb protein did not appear to permit transit to S-phase, and was not accompanied by an expression of c-Myc. Surprisingly, therefore, an antimetabolite inducing a loss of GTP brought about cell death by apoptosis in the G1 phase of the cell cycle.

Glucuronidation associated with intrinsic resistance to mycophenolic acid in human colorectal carcinoma cells.

The in vivo efficacy of the antitumor, immunosuppressive antibiotic mycophenolic acid is known to be limited by its rapid conversion to the biologically inactive 7-0-glucuronide, catalyzed by UDP-glucuronosyl transferase activity, which is widely distributed among normal tissues, including intestinal epithelium. We have found that mycophenolic acid is also converted to its glucuronide by several lines of human colorectal carcinoma cells, including HT29, Lovo, and Colo-205. In contrast, malignant cell lines not of colorectal origin, including EMT6, HeLa, and SKOV3, showed no ability to metabolize mycophenolic acid. The 7-amino derivative of mycophenolic acid was not metabolized by HT29 cells. This compound was less potent than mycophenolic acid versus EMT6 and HeLa cells but showed inhibitory activity against HT29 cells comparable with the parent antibiotic. The rapid metabolism of mycophenolic acid by HT29 cells was associated with a markedly lower sensitivity to both the antiproliferative activity of the drug and to its ability to inhibit GTP synthesis, compared with cells lacking the capacity for significant glucuronidation. After an initial decline in cellular GTP in HT29 cells induced by mycophenolic acid, there was a progressive recovery in GTP over 48 h, accompanying the metabolism of the antibiotic. This recovery process was not observed in EMT6 cells. It is suggested that glucuronosyl transferase activity may occur widely in colorectal cancer cells and could contribute to resistance to drugs that are susceptible to inactivation by glucuronide conjugation.

Glucuronidation by human colorectal adenocarcinoma cells as a mechanism of resistance to mycophenolic acid.

Mycophenolic acid (MPA), a potent and specific inhibitor of IMP dehydrogenase, exerts its anti-mitotic action by a rapid depletion of the cellular content of guanine nucleotides. Although MPA is a potent inhibitor of GTP synthesis in the HT29 line of human colorectal adenocarcinoma cells in short-term culture, its ability to depress the cloning efficiency of these cells was found to be markedly less than against the mouse mammary carcinoma line, EMT6. In vivo, MPA is efficiently converted to the biologically inactive O-glucuronide derivative thereby limiting its effectiveness as an anti-tumor agent. Investigation of the fate of MPA incubated with monolayer cultures of HT29 and EMT6 cells revealed that the compound is rapidly converted to the O-glucuronide derivative by HT29 cells, but not by EMT6 cells. Confirmation of the identity of the glucuronide formed by HT29 cells was obtained by its conversion to MPA after incubation with beta-glucuronidase and by comparison of the mass spectrum of its HPLC peak with that of synthetic MPA O-glucuronide. Cultures of two other lines of human colorectal adenocarcinoma cells, Colo-205 and LoVo, also depleted their culture media of MPA although we have not yet established whether these cells also synthesize the glucuronide. The intrinsic partial resistance of HT29 cells to MPA appears to be associated with the ability of these cells to convert MPA to the biologically inactive glucuronide. These results, in conjunction with other reports of the capacity of colorectal cancer cells for Phase I and II metabolism of xenobiotics, may have implications for the design of drugs intended for the treatment of colorectal cancer.

The complete cDNA derived sequence of the rat prolyl 4-hydroxylase alpha subunit.

A rat cDNA encoding the prolyl 4-hydroxylase alpha subunit (P4H alpha) was isolated and sequenced. The primary aa sequence deduced from the nucleotide sequence reveals a 534-aa protein that shows extensive aa identity with the human (88%) and chick (77%) P4H alpha.

Pharmacodynamics of the inhibition of GTP synthesis in vivo by mycophenolic acid.

Mycophenolic acid is effective against a wide range of experimental tumors in rodents when given orally, despite rapid metabolism to the inactive glucuronide derivative and rapid clearance from plasma. In the light of this, the pharmacodynamic action of mycophenolic acid on the radiolabelling of GTP and ATP by [14C]hypoxanthine in spleen and heart has been investigated in vivo in the rat as a preliminary to studies in tumor tissue. The data indicate that inhibition of GTP, and more surprisingly, ATP synthesis in spleen was sustained for at least 24.25 hr after single oral doses of the disodium salt of mycophenolic acid, indicating that the inhibitor is retained in sensitive cells for considerably longer than might be expected from the pharmacokinetic profile in the plasma in this species. GTP and ATP levels became depressed in rat spleen subsequent to the inhibition of nucleotide radiolabelling. The persistence of mycophenolic acid in proliferating cells may account for the effectiveness of once daily dosing against rapidly growing experimental tumors. In contrast with spleen, there was no incorporation of radiolabel from [14C]hypoxanthine into either GTP or ATP in rat heart and mycophenolic acid had no effect on the cardiac content of either nucleotide. The lack of effect of mycophenolic acid on cardiac GTP levels is consistent with the absence of any pharmacological action on cardiac functions associated with receptor-G-protein-GTP interactions. The ability of the morpholinoethyl ester of mycophenolic acid (a clinically effective immunosuppressive agent) to inhibit GTP synthesis and depress GTP levels in rat spleen in vivo was compared with that of mycophenolic free acid and its disodium salt. The ester derivative was clearly more effective than the poorly water-soluble free acid but showed comparable activity with the freely soluble disodium salt.

Novel inhibitors of prolyl 4-hydroxylase. 3. Inhibition by the substrate analogue N-oxaloglycine and its derivatives.

N-Oxaloglycine (3) is an alpha-ketoglutarate (1) analogue that is a competitive inhibitor of prolyl 4-hydroxylase (EC 1.14.11.2). A study of the structure-activity relationships of some other oxalo derivatives shows that substitution on the glycine moiety modulates activity stereoselectively and that if the omega-carboxylate is homologated or replaced by either acylsulfonamides or anilide, then activity is sharply reduced. This sensitivity to these changes is contrasted with the relative insensitivity of another putative alpha-ketoglutarate analogue, pyridine-2,5-dicarboxylic acid (2), and the implication is discussed that compounds of both series are unlikely to bind to prolyl hydroxylase in the same way even though both inhibit the enzyme competitively.

Inhibition of collagen hydroxylation by 2,7,8-trihydroxyanthraquinone in embryonic-chick tendon cells.

Prolyl 4-hydroxylase (EC 1.14.11.2) is an essential enzyme in the post-translational modification of collagen. Inhibitors of this enzyme are of potential interest for the treatment of diseases involving excessive deposition of collagen. 2,7,8-Trihydroxyanthraquinone (THA) is an effective inhibitor of prolyl 4-hydroxylase by virtue of its ability to compete with the co-substrate 2-oxoglutarate (Ki = 40.3 microM). Using a simple and reproducible assay for collagen hydroxylation, we show that THA inhibits the hydroxylation of collagen in embryonic-chick tendon cells in short-term culture, with an IC50 value (concn. giving 50% inhibition) of 32 microM. In comparison, the ethyl ester of 3,4-dihydroxybenzoic acid has an IC50 value of 0.1 mM against collagen hydroxylation by chick tendon cells, whereas its Ki versus isolated prolyl 4-hydroxylase is 49 microM.

Assay of prolyl 4-hydroxylase by the chromatographic determination of [14C]succinic acid on ion-exchange minicolumns.

An assay for prolyl 4-hydroxylase (EC 1.14.11.2) is described which measures succinic acid produced during the decarboxylation of 2-oxoglutaric acid in the presence of poly(L-Pro-Gly-L-Pro). [1-14C]Succinic acid was separated from its precursor 2-oxo[5-14C]glutaric acid by using ion-exchange minicolumns. The contamination of succinic acid by 2-oxoglutaric acid was approx. 1%, and the recovery of succinic acid was 100%. Kinetic parameters of prolyl 4-hydroxylase measured by the assay showed good agreement with published values. Our experience indicates that the measurement of prolyl 4-hydroxylase by the production of succinic acid is especially suited to investigations involving large numbers of assays.

Inhibition of prolyl 4-hydroxylase by hydroxyanthraquinones.

Prolyl 4-hydroxylase (EC 1.14.11.2) is an essential enzyme in the post-translational modification of collagen. Inhibitors of this enzyme are of potential interest for the treatment of diseases involving excessive deposition of collagen. We have found that anthraquinones with at least two hydroxy groups ortho to each other are potent inhibitors of this enzyme. Kinetic studies revealed that 2,7,8-trihydroxyanthraquinone (THA) competitively inhibited the co-substrate, 2-oxoglutarate, but was non-competitive with regard to ascorbate and was tentatively considered to be uncompetitive with regard to protocollagen. The inhibition by THA was greatly enhanced in the absence of added Fe2+ and was partially reversed by the addition of concentrations of Fe2+ in excess of the optimum for the enzymic reaction. Binding studies indicated that THA is an effective chelating agent for Fe2+. Several non-quinoidal compounds bearing the catechol moiety also inhibited the enzyme. The results suggest that THA inhibited prolyl 4-hydroxylase by binding to the enzyme at the site for 2-oxoglutarate possibly involving the Fe2+ atom, rather than by complexing with Fe2+ in free solution. The inhibition of prolyl 4-hydroxylase by THA exhibited strong positive co-operativity and may involve three distinct but non-independent binding sites.

Acceleration of wound healing by recombinant human urogastrone (epidermal growth factor).

Urogastrone, a human peptide growth factor that is closely related by its amino acid sequence to mouse epidermal growth factor, has been prepared by recombinant DNA techniques. The powerful mitogenic activity of urogastrone led us to evaluate its potential as an aid to wound healing. We have used the repair reaction that follows transection of the Achilles tendon in the rat to evaluate the action of urogastrone in this model. Twice-daily injections of recombinant urogastrone (20 micrograms/kg) close to the site of the wound led to more rapid increases in the dry weight of the repair lesion and its collagen and DNA contents compared with those in saline-injected controls (5 days). At 15 and 30 days after surgery the urogastrone-treated lesion continued to show enhanced dry weights and collagen contents, whereas the DNA content declined to that of saline-treated lesions. When urogastrone treatment was stopped after 15 days, the repair lesions examined at 30 days were little different from those in animals untreated throughout. Our results provide encouragement for an eventual clinical assessment of recombinant urogastrone in humans as an aid to wound healing provided that satisfactory answers are found to several outstanding questions.

Energetics of ligand binding to receptors.

The primary events in the biological signals elicited by hormonal ligands are the initial binding of the ligands to their receptors and the consequent changes in the receptors that this induces. Although the effects of chemical modifications of ligands on their affinity for receptors have been intensively studied in several systems, the influence of such changes upon the ability of ligands to induce the conformational changes in receptors that are necessary for the expression of agonism has been largely ignored. This paper considers some theoretical and experimental approaches to the problem of the relationships between ligand structure, binding energy and the induction of conformational change in receptors. The significance of the concept of productive and non-productive ligand binding to proteins previously applied with considerable success by Jencks to the understanding of enzyme catalysis, is highlighted. Given a series of closely related ligands it is possible to estimate the energetic contributions of substituent groups not only to the binding affinity to the receptor but also to the induction of conformational change in the receptor provided that the rate of conformational change, can be monitored.

Effects of inhibitors of collagen maturation on hypertension in rats.

The effects of beta-aminopropionitrile (BAPN), an inhibitor of collagen maturation, have been examined in three models of hypertension in rats. It is shown for the first time that BAPN has a rapidly developing antihypertensive action in rats with renal (one kidney-one clip) hypertension. The compound also lowered blood pressure in spontaneously hypertensive rats. In both these experiments although BAPN lowered blood pressure it did not affect the hypertension-induced hypertrophy of the aorta. BAPN suppressed the rise in blood pressure during the developing phase of deoxycorticosterone acetate-salt-induced hypertension and in this model the compound also prevented the accompanying aortic hypertrophy. However, another inhibitor of collagen synthesis and maturation, 3,4-dehydroproline, completely suppressed aortic hypertrophy but failed to prevent the rise in blood pressure in deoxycorticosterone acetate-salt hypertension. It is concluded that: (1) BAPN may have a direct antihypertensive action, and (2) inhibition of aortic hypertrophy alone cannot prevent the rise in blood pressure in the developing phase of deoxycorticosterone acetate-salt hypertension.