α-Hydroxypyridones inhibit hydroxylases linked with collagen synthesis and cell multiplication.
详细信息 查看全文
文摘
Metalloenzymes rank prominently among the potential targets for chelators used therapeutically to alleviate the tissue iron overload that occurs in chronically transfused patients. Non-heme enzymes providing more than one coordination site at their metal center can be expected to be particularly susceptible. The protein hydroxylases figure prominently among this group. We have studied the effect of several α-hydroxypyridones (α-HPs), agents now widely investigated as oral iron chelators, on the two most pivotal of these enzymes, prolyl 4-hydroxylase [PPH; E.C.1.14.11.2] and deoxyhypusyl hydroxylase [DOHH; E.C.1.14.99.29] which catalyze the posttranslational formation of peptidyl hydroxyproline and hypusine, respectively. Hydroxyproline occurs in all collagenous proteins. Its formation determines their triple helical conformation and controls the biosynthesis of the collagens and collectins; inhibition of prolyl hydroxylation results in selective suppression of their secretion. Hypusine occurs exclusively in the eukaryotic initiation factor 5A (eIF-5A). Its formation is associated with exit from the G1 compartment of the cell cycle and with initiation of DNA replication; reversible inhibition of deoxyhypusyl hydroxylation coincides with reversible arrest at the G1-S boundary and with reversible disappearance of a subset of mRNAs from polysomes, leading some to speculate that eIF-5A might govern the selective translation of G1 exit-gating transcripts. PPH and DOHH are dioxygenases which stereospecifically hydroxylate residues within -Gly-X-Y-Gly- motifs, and provide three cisoid coordination sites at their active site metal ion. Under turnover conditions, this ion is tightly bound to each apoenzyme, accessible only to chelators which fit the active site. The eight α-HPs tested, including CP20, CP94 and CP51, were inhibitory for both human enzymes at concentrations achievable in patients. Based on the concentrations giving half-maximal inhibition, DOHH was more sensitive than PPH. In fact, the ID50's obtained for suppression of cellular DOHH activity were consistently lower than the apparent Ki's obtained forpurified PPH. Using purified human PPH recombinantly produced in Sp. frugiperda insect cells, inhibitory efficiency was optimal if an electron-rich substituent was spatially oriented to mimic the C5 carboxyl group of the co-substrate, 2-oxoglutarate. α-HP analogs, for example, 3,4-dihydroxybenzoic acid (3,4-DHBA) and its ethyl ester, exhibited Ki's of 5 and 225 μM, respectively. By contrast, 3,4-DHBA inhibited DOHH with an apparent Ki of ≤ 1500 μM, that of the ethyl ester being significantly lower (500 μM). As expected, the α-HPs selectively inhibited collagen secretion by cultured human fibroblasts and smooth muscle cells. Using human lymphoblastoid cells, the α-HPs caused arrest at the G1-S boundary, rapidly reversible upon their removal. In the cells studied, flow cytometry revealed no other cytokinetic effects. In particular, we did not observe an arrest in S-phase, a finding that argues against a significant effect of α-HPs on the elongation phase of DNA replication, as one would expect from an inhibitory interaction with the catalytic subunit of deoxyribonucleotide reductase. It should be noted that this metalloprotein subunit is only expressed upon entry into S-phase and therefore is not a target for the α-HPs during G1. Based on these studies of the two human enzymes and of their role in cultured human cells, we propose that chronically transfused patients receiving adjuvant oral iron chelator therapy with α-HPs require careful clinical and biochemical monitoring for insidiously evolving connective tissue abnormalities, and for anti-proliferative side-effects. Indeed, one of the serious side-effects reported for oral CP20 is a reversible agranulocytosis and thrombocytopenia. We suggest that the susceptibility of DOHH to inhibition by α-HPs and the associated proliferative arrest in late G1 provide a plausible mechanistic explanation for this clinical observation. Our investigation suggests that novel α-HPs can be designed so as not to interact with either protein hydroxylase while retaining the capacity to function as effective, orally active iron chelators.