Characterization of the biosynthesis, processing and kinetic mechanism of action of the enzyme deficient in mucopolysaccharidosis IIIC

Abstract

Heparin acetyl-CoA:alpha-glucosaminide N-acetyltransferase (N-acetyltransferase, EC 2.3.1.78) is an integral lysosomal membrane protein containing 11 transmembrane domains, encoded by the HGSNAT gene. Deficiencies of N-acetyltransferase lead to mucopolysaccharidosis IIIC. We demonstrate that contrary to a previous report, the N-acetyltransferase signal peptide is co-translationally cleaved and that this event is required for its intracellular transport to the lysosome. While we confirm that the N-acetyltransferase precursor polypeptide is processed in the lysosome into a small amino-terminal alpha- and a larger ß- chain, we further characterize this event by identifying the mature amino-terminus of each chain. We also demonstrate this processing step(s) is not, as previously reported, needed to produce a functional transferase, i.e., the precursor is active. We next optimize the biochemical assay procedure so that it remains linear as N-acetyltransferase is purified or protein-extracts containing N-acetyltransferase are diluted, by the inclusion of negatively charged lipids. We then use this assay to demonstrate that the purified single N-acetyltransferase protein is both necessary and sufficient to express transferase activity, and that N-acetyltransferase functions as a monomer. Finally, the kinetic mechanism of action of purified N-acetyltransferase was evaluated and found to be a random sequential mechanism involving the formation of a ternary complex with its two substrates; i.e., N-acetyltransferase does not operate through a ping-pong mechanism as previously reported. We confirm this conclusion by demonstrating experimentally that no acetylated enzyme intermediate is formed during the reaction.

Figure 1. Scaled diagrams of the full length N-acetyltransferase and the small N-terminal loop fragment of the protein expressed by the cDNA constructs used in this study.

Each construct also encodes a C-terminal His8-Flag tag. Predicted masses (kDa) are shown for each protein segment. TMDs are indicated by hatched rectangles, soluble loops that are predicted to reside in the ER/lysosomal lumen are shown as clear boxes and those predicted to reside in the cytosol as shaded boxes. Putative Asn-linked glycosylation sites (NXS/T) are shown as “*”. The location and ends of the peptide sequence used to generate the “N-terminal” antibody and utilized in this report, are shown between the two panels. (A) Full length N-acetyltransferase with both the long signal peptide (SP-L), found in the genomic sequence , and the short signal peptide (SP-S), used in this report and found in EST data bases are indicated. The positions and sequences of previously published intracellular transport signals are shown, as are the N-termini, identified in this report, of the processed alpha- and ß- chains. (B) The N-terminal fragment containing the short signal peptide and the first luminal loop of N-acetyltransferase are depicted.

Figure 2. Effects of lipids on N-acetyltransferase activity are illustrated.

Each set of data points represents the average of triplicate determinations of N-acetyltransferase activity with their standard deviations (SD) shown as error bars (A) DDM extracts from permanently transfected HeLa cells were serially diluted in CP buffer, pH 5.5 (filled diamonds), CP buffer containing 0.25% HSA (filled squares) or CP buffer containing 1.3 mM 20% PI (filled circles with the error bars representing SD too small to see, connected by the best fit line, R = 1) (X-axis, total extracted protein; Y-axis, transferase activity in nmol/h). (B) Equal amounts of anti-Flag-column purified N-acetyltransferase were assayed in the presence of increasing concentrations of lipids (X-axis, mM) with decreasing mole% of negatively charged PI (40-0% PI+40–80% PC+20% CH); 40% PI, darkly shaded bars; 20% PI, grey hatched bars; or PC (0% PI), white bars. (C) Stability of identical amounts of purified N-acetyltransferase in CP buffer pH 5.5, containing either 0.25% (w/v) HSA (white bars) or 1.3 mM, 20% PI, lipids (shaded bars) (X-axis, hours at 37°C; Y-axis transferase activity in nmol/h).

Figure 3. The effects of detergent extraction procedures on the intensity of bands associated with monomeric N-acetyltransferase (i.e., 62 and 44 kDa), detected by Western blotting with the C-terminal rabbit Flag antibody.

HeLa cells, permanently transfected with N-acetyltransferase-His8Flag, were; lane 1- first extracted with 1% DDM, centrifuged and the extract denatured in SDS-PAGE sample buffer (containing a reducing agent) at room temperature; lane 2- homogenized (in water), sonicated and boiled in lithium dodecyl sulfate reducing sample buffer ; or lane 3- extracted with 2% SDS, diluted with SDS-PAGE reducing sample buffer and heated at 65°C . Bottom panel shows the loading control, the cation-independent mannose-6-phosphate receptor, which contains a single TMD.

Figure 4. Purification of N-acetyltransferase by anti-Flag affinity chromatography.

(A) Lane 1- SDS-PAGE separation of the purified enzyme under reducing conditions detected with fluorescent SYPRO Ruby protein stain. Also shown is a dilution series of BSA used to calculate the protein levels in the transferase sample and the specific activity of the purified enzyme; and (B) Western blotting using either an antibody against the Gln53-Asn156 epitope (N-term), or against the C-terminal Flag tag (Flag). Lane-1 contains the transferase that was bound, eluted with Flag peptides and then denatured, while lane 2 contains enzyme that was directly released and denatured from the anti-Flag column with SDS sample buffer (containing a reducing agent). Aliquots of the initial 1% DDM extract (Ex) were also examined by Western blotting. Bands marked “X” in the Ex lanes are non-specific proteins detected by the antibodies (the ∼30 kDa X band is only seen with the rabbit Flag antibody). (C) The effects of incubating the purified protein at 37°C for 0–2.5 h without added lipids were examined by Western blotting using the N-terminal antibody.

Figure 5. Western blot analysis of the endogenous N-acetyltransferase protein in three human fibroblast cell lines.

Fibroblast from an unaffected individual (WT), a patient with I-Cell disease, and a MPS IIIC patient were extracted with 1% DDM, 20 µg each of extracted protein were separated by SDS-PAGE, and the N-acetyltransferase proteins visualized using the N-terminal antibody. The blot was reprobed with anti GAPDH as a loading control. The specific activity of each extract is shown at the bottom, ND is not detected.

Figure 6. Western blot analysis of the post nuclear supernatant (PNS) and enriched lysosomal (Lys) fractions separated magnetically from extracts of N-acetyltransferase-His8Flag permanently transfected HeLa cells loaded with iron-dextran (FeDex).

After SDS-PAGE the proteins in the PNS and Lys fractions (1 µg total protein from each fraction was loaded) were visualized with the N-terminal N-acetyltransferase antibody (N-term). The nonspecific ∼50 kDa band visualized with the N-terminal antibody is marked as “X”. As control for the enrichment of lysosomes in the “Lys” fraction, the blot was stripped and re-probed with an antibody against human lysosomal ß-hexosaminidase A (Hex A). Furthermore, markers for the ER (Calnexin) and early endosome (EEA1) were visualized after stripping and re-probing the blot with the appropriate antibody (bottom).

Figure 7. Western blot analyses of the N-acetyltransferase-His8Flag protein from the “Lys” fraction of Figure 6 (lane 1) treated with either endo-H (lane 2) or PNGase (lane 3).

Proteins were detected with either, (A) the N-terminal N-acetyltransferase antibody or (B) the Flag M2 antibody.

Figure 8. The protein expressed by a construct encoding myc-N-acetyltransferase-His8Flag was analyzed by Western blotting.

(A) Extracts from HeLa cells permanently expressing the singly-tagged N-acetyltransferase-His8Flag (lane 1) were compared with HeLa cell transiently expressing myc-N-acetyltransferase-His8Flag (lane 2) using the N-terminal N-acetyltransferase antibody. The levels of transferase activity in the extracts are given below each lane. (B) The oligosaccharides present on the doubly-tagged myc-N-acetyltransferase-His8Flag protein (lane 1) were analyzed based on endo-H (lane 2) and/or PNGase (lane 3) sensitivities by Western blotting using a myc antibody.

Figure 9. Indirect immunofluorescence staining and confocal microscopy imaging of HeLa cells transiently expressing various forms of N-acetyltransferase.

(A) N-acetyltransferase-His8Flag, (B & C) myc-N-acetyltransferase-His8Flag, or (D & E) the N-terminal segment (Met1-Asn163-His8Flag) of N-acetyltransferase (Fig. 1). Cells were stained with the mouse M2 Flag antibody (green A, B & D) and a rabbit antibody against either a lysosomal marker, glucocerebrosidase (red, A, B & D), or against the N-terminus of N-acetyltransferase (green) and a mouse antibody against a ER marker, PDI (red, C & E), and the images merged. Scale bars represent 10 µm. Images shown are representative of 10 recorded.

Figure 10. The degree of endo-H versus PNGase sensitivities was used to probe the level of Asn-linked oligosaccharide processing that occurs on the N-terminal fragment (Met1- Asn163-His8Flag) of N-acetyltransferase ( Fig. 1 ) from transiently expressing HeLa cells.

Proteins from soluble cell lysates (i.e., not detergent extracts) were analyzed by Western blotting using the N-terminal antibody.

Figure 11. Proteins extracted from HeLa cells permanently expressing N-acetyltransferase-His8Flag were separated by molecular size exclusion chromatography on a 90×1.5 cm Sephacryl S-400 column eluted at 4°C.

Each 1 mL fraction (X-axis) was analyzed for transferase activity (nmol/mL* h, left Y-axis, solid line) and total protein (µg/mL, right Y-axis, dashed line). Additionally every fifth fraction was analyzed for N-acetyltransferase protein by dot-blot using the N-terminal antibody (shown below the X-axis).

Figure 12. Direct Michaelis-Menten best-fit curves of specific activity (nmol/h*ng) versus [S] for the N-acetyltransferase reaction with graphic insets displaying the corresponding Lineweaver-Burk double reciprocal plots.

Anti-flag affinity column purified N-acetyltransferase was used and each datum point represents the average of two or three assays. (A) The concentration of MU-GlcNH₂ was varied between 0.01 and 1 mM, while AcCoA was fixed at 0.17 mM (▴), 0.33 mM(○) or 1.0 mM (▪). (B) The concentration of AcCoA was varied between 0.083 and 2.0 mM while MU-GlcNH₂ was fixed at 0.05 mM (⧫), 0.10 mM (□) or 0.30 mM (•). The experiment was independently repeated three times with a representative set of graphics shown.