Human ZMPSTE24 disease mutations: residual proteolytic activity correlates with disease severity

Abstract

The zinc metalloprotease ZMPSTE24 plays a critical role in nuclear lamin biology by cleaving the prenylated and carboxylmethylated 15-amino acid tail from the C-terminus of prelamin A to yield mature lamin A. A defect in this proteolytic event, caused by a mutation in the lamin A gene (LMNA) that eliminates the ZMPSTE24 cleavage site, underlies the premature aging disease Hutchinson-Gilford Progeria Syndrome (HGPS). Likewise, mutations in the ZMPSTE24 gene that result in decreased enzyme function cause a spectrum of diseases that share certain features of premature aging. Twenty human ZMPSTE24 alleles have been identified that are associated with three disease categories of increasing severity: mandibuloacral dysplasia type B (MAD-B), severe progeria (atypical 'HGPS') and restrictive dermopathy (RD). To determine whether a correlation exists between decreasing ZMPSTE24 protease activity and increasing disease severity, we expressed mutant alleles of ZMPSTE24 in yeast and optimized in vivo yeast mating assays to directly compare the activity of alleles associated with each disease category. We also measured the activity of yeast crude membranes containing the ZMPSTE24 mutant proteins in vitro. We determined that, in general, the residual activity of ZMPSTE24 patient alleles correlates with disease severity. Complete loss-of-function alleles are associated with RD, whereas retention of partial, measureable activity results in MAD-B or severe progeria. Importantly, our assays can discriminate small differences in activity among the mutants, confirming that the methods presented here will be useful for characterizing any new ZMPSTE24 mutations that are discovered.

Figure 1.

Summary of the ZMPSTE24 disease alleles reported in cases of MAD, atypical HGPS and RD. Diseases are grouped in order of increasing severity, with MAD-B as the least severe and RD as most severe. The pairwise combination of specific mutations that have been attributed to each patient case is provided. Grey shading indicates mutants that do not retain the domain containing the HEXXH motif, which lies in positions 335–339 in ZMPSTE24. Mutants tested in this study are highlighted in yellow.

Figure 2.

Schematic of the patient mutations analyzed in this study by in vivo and in vitro activity assays. ZMPSTE24 is predicted to contain seven transmembrane spans, with three cytosolic loops and a C-terminal cytosolic tail containing a dilysine ER retrieval motif (green). The loops are drawn roughly to scale. Representative schematics show the approximate location of the molecular lesions in the ZMPSTE24 mutants examined in this study. The HEXXH catalytic motif (HELGH) in human ZMPSTE24 is at positions 335–339 (yellow). The correlation to the disease/patient case as shown in Figure 1 is noted in parentheses.

Figure 3.

The halo and mating assays reveal measurable differences in activity of ZMPSTE24 mutants. Results of the qualitative halo assay (A) and the plate mating assay (B) are shown. Results of a quantitative filter mating assay are graphically represented below (C). The details of how these tests were performed are discussed in the text. The mating percentages graphed in (C) are normalized to WT (100%). The values for each mutant tested were: H335A (1%), W340R (47%), P248L (25%), N265S (11%), L438F (6%), L94P (2%), W450X (1%), L362F(fsX18) (1%) and T159_L209del (1%).

Figure 4.

A coupled in vitro proteolysis activity assay reveals differences in specific activity for the ZMPSTE24 disease mutants. The specific activity of membrane preparations containing wild-type ZMPSTE24, the H335A catalytic site mutant and disease mutants were determined by the coupled proteolysis and methylation assay described in the text. Experiments were performed three times in triplicate for each ZMPSTE24 protein (for a total of nine trials).

Figure 5.

Expression of ZMPSTE24 mutant proteins when compared with the wild-type. Crude membranes prepared from yeast strains expressing WT and mutant His₁₀HA₃ ZMPSTE24, or empty vector (EV) were subjected to 10% SDS–PAGE, transferred to nitrocellulose membrane, probed with α-HA primary antibody and a goat α-mouse horseradish peroxidase-conjugated secondary antibody and detected by ECL. Varying amounts of crude membranes were added to each lane to optimize visualization of the signal. When compared with the WT, H335A catalytic site mutant, and most point mutants (left), the frameshift, deletion, nonsense and L94P mutations (right) required significantly more membranes for visualization. Amounts added to each lane are as follows: left panel: WT (0.5 µg), N265S (0.5 µg), H335A (0.5 µg), L438F (0.5 µg), P248L (1.0 µg), W340R (3.0 µg), EV (3.0 µg); right panel: WT (0.1 µg), L362F(fsX18) (2.0 µg), W450X (3.0 µg), L94P (12 µg), T159_L209del (12 µg) and EV (12 µg).

Figure 6.

Schematic of the ZMPSTE24-associated progeroid disease severity spectrum. The results presented here and in other studies support the model that a spectrum of progeroid diseases of increasing severity, ranging from MAD-B to atypical progeria to RD, appears to correlate with an increasing amount of uncleaved prelamin A resulting from a failure of ZMPSTE24 cleavage. It has yet to be established whether metabolic syndrome is also directly connected to defects in prelamin A maturation, hence the question mark. The L438F studied here has been reported to be associated with metabolic syndrome (47), but the second patient allele is not known, and whether this mutation is truly causative of metabolic syndrome remains to be established.