Biochemical Investigation of Progeroid Disease-Associated Mutations in Human ZMPSTE24
ZMPSTE24 is a unique intramembrane zinc metalloprotease that plays critical roles in the lamin A maturation pathway. Lamin A comprises a dense network underlying the inner nuclear membrane that maintains the structural integrity and proper function of the nucleus. The precursor of lamin A, prelamin A, terminates with a CAAX motif, where “C” is cysteine, “A” is typically an aliphatic amino acid, and “X” is one of several different amino acids. Like all CAAX proteins, prelamin A undergoes a series of post-translational modifications, including farnesylation of the cysteine by farnesyltransferase, endoproteolysis of the AAX residues by ZMPSTE24 or possibly a related protease RCE1, and carboxyl methylation by isoprenylcysteine carboxyl methyltransferase (ICMT). After CAAX processing, an additional cleavage event by ZMPSTE24 occurs to remove 15 residues from the C-terminus, including the farnasylated and carboxyl methylated cysteine, releasing mature lamin A into the nucleoplasm. Mutations in the gene encoding ZMPSTE24 that impair proteolytic activity cause a set of progeroid diseases, including B-type mandibuloacral dysplasia (MAD-B) and restrictive dermopathy (RD). Recently, ZMPSTE24 mutations were also detected in patients with metabolic syndrome (MS) and nonalcoholic fatty liver disease (NAFLD). Patients with these diseases have shown defective prelamin A processing, leading to the accumulation of persistently farnesylated prelamin A in the nucleus. However, how this accumulation causes disease remains unclear. We demonstrated that both ZMPSTE24 and another known CAAX protease, RCE1 are both capable of mediating the C-terminal cleavage of prelamin A. RCE1 retains the capacity to cleave the AAX residues of prelamin A in progeroid diseases induced by inactive ZMPSTE24 mutants, therefore the disease molecule will be most likely farnesylated and methylated prelamin A. These factors suggest that the ability of ZMPSTE24 to perform the upstream cleavage determines the accumulation level of uncleaved prelamin A in progeroid diseases. However, there was no available assay that could quantitatively demonstrate the in vitro upstream cleavage activity of ZMPSTE24. Therefore, we first developed a FRET-based assay that was able to precisely quantify the upstream cleavage activity of Ste24, the yeast homolog of ZMPSTE24. The 33-mer analog of a-factor, Ste24 natural substrate, has a 2-aminobenzoic acid (Abz) fluorophore at the N-terminus and a dinitophenol (Dnp) quencher located on the either side of the proposed cleavage site. After cleavage, quantification of the fluorescence from the dequenced peptide enabled us to continuously monitor the upstream cleavage activity of Ste24. We then utilized this FRET-based assay to examine the upstream cleavage activity of wild-type ZMPSTE24 and its disease mutants. We demonstrated that the a-factor analog FRET substrate could be recognized and cleaved at the predicted position by ZMPSTE24. Disease variants were examined using this assay and the results revealed reduced upstream cleavage activity. Moreover, blocking ubiquitylation restored catalytic activity of some ZMPSTE24 disease variants, suggesting that diminished activity of these mutants is due to protein instability. Limited trypsin digestion results also indicated that some variants may not be properly folded, as compared to wild-type. The crystal structure of ZMPSTE24 revealed that seven transmembrane helices of ZMPSTE24 surround a large intramembrane chamber. The HEXXH zinc metalloprotease motif faces inward to cap the top of the chamber inside which proteolysis is proposed to occur. The four side portals apparent in the structure may provide substrate entry and exit routes. Based on structural considerations, besides interfering with structural integrity, these disease mutations may decrease ZMPSTE24 activity by preventing substrate binding in the active site or occluding substrate entry into or exit from the enzyme chamber. Using a yeast a-factor sequence-based photoactive analog containing benzophenone in the farnesyl portion, which can be well processed by ZMPSTE24, we have shown that certain disease mutants may affect farnesyl binding of the C-terminal cleavage substrate. We also designed several double cysteine mutants near portal 1 in Cys-less background of Ste24. We then utilized bismaleimide sulfhydryl-to-sulfhydryl crosslinkers to block the portal opening. Some of the crosslinked double mutants showed reduced AAX cleavage activity, suggesting the portal 1 may be important for the C-terminal cleavage. Together, these data will clarify how the enzyme functions and also provide further insights into progeroid diseases.