Molecular identification of NAT8 as the enzyme that acetylates cysteine S-conjugates to mercapturic acids

Abstract

Our goal was to identify the reaction catalyzed by NAT8 (N-acetyltransferase 8), a putative N-acetyltransferase homologous to the enzyme (NAT8L) that produces N-acetylaspartate in brain. The almost exclusive expression of NAT8 in kidney and liver and its predicted association with the endoplasmic reticulum suggested that it was cysteinyl-S-conjugate N-acetyltransferase, the microsomal enzyme that catalyzes the last step of mercapturic acid formation. In agreement, HEK293T extracts of cells overexpressing NAT8 catalyzed the N-acetylation of S-benzyl-L-cysteine and leukotriene E(4), two cysteine conjugates, but were inactive on other physiological amines or amino acids. Confocal microscopy indicated that NAT8 was associated with the endoplasmic reticulum. Neither of the two frequent single nucleotide polymorphisms found in NAT8, E104K nor F143S, changed the enzymatic activity or the expression of the protein by >or=2-fold, whereas a mutation (R149K) replacing an extremely conserved arginine suppressed the activity. Sequencing of genomic DNA and EST clones corresponding to the NAT8B gene, which resulted from duplication of the NAT8 gene in the primate lineage, disclosed the systematic presence of a premature stop codon at codon 16. Furthermore, truncated NAT8B and NAT8 proteins starting from the following methionine (Met-25) showed no cysteinyl-S-conjugate N-acetyltransferase activity when transfected in HEK293T cells. Taken together, these findings indicate that NAT8 is involved in mercapturic acid formation and confirm that NAT8B is an inactive gene in humans. NAT8 homologues are found in all vertebrate genomes, where they are often encoded by multiple, tandemly repeated genes as many other genes encoding xenobiotic metabolism enzymes.

FIGURE 1.

Alignment of human NAT8 and NAT8B with related sequences and with NAT8L (aspartate N-acetyltransferase). The following sequences are shown: Homo sapiens NAT8, NAT8B, and NAT8L (HsapiensNAT8, gi:47115157; HsapiensNAT8B, gi:117168283; HsapiensNAT8L, gi:259016335); Canis familiaris (CfamiliarisCML2, gi:7398098); Mus musculus (MmusculusCML5, gi:19353130); Gallus gallus (GgallusNAT8B, gi:118110777); and D. rerio NAT8L (DrerioNAT8L, gi:139947555). The human NAT8B coding sequence includes a stop codon (X) at codon 16. G23R and Q168X (underlined in Hsapiens NAT8B) are two frequent alleles. E104K and S143F (underlined in HsapiensNAT8) are two common SNPs shown for the human NAT8 coding sequence. Residues strictly conserved among all sequences shown are in boldface. Residues conserved among the NAT8-related sequences are in blue. Residues conserved among the two NAT8L sequences are in red.

FIGURE 2.

Expression of NAT8 in HEK293T cells. Western blot analysis of extracts (20 μg of total protein per lane) from HEK293T cells transfected for 48 h with empty pEF6-HisB (lanes 1–3), pEF6-NAT8 (lanes 4–6), and pEF6/Myc-NAT8 (lanes 7–9). The anti-His antibody used was the penta-His antibody (Qiagen) that detects both N- and C-terminally tagged NAT8s but with different affinities. Triplicates are the result of three independent transfection experiments.

FIGURE 3.

Catalytic activity of overexpressed human NAT8. The acetyltransferase activity of whole cell extracts from HEK293T cells transfected with either pEF6-NAT8 (b) or the corresponding empty plasmid (pEF6-HisB; a) and pEF6/Myc-NAT8 (d) or the corresponding empty plasmid (pEF6/Myc-HisA; c) were assayed as described under “Experimental Procedures” with 0.5 mm of the indicated substrates, 0.2 mm acetyl-CoA and ∼30 μg of total protein. Results are means ± S.E. for three independent measurements performed with the same extract.

FIGURE 4.

Mass spectrometry analysis of the acetylation product made by human NAT8 from S-benzyl-l-cysteine and acetyl CoA. A cell extract from HEK293T cells overexpressing human NAT8 with an N-terminal His₆ tag was incubated for 45 min at 37 °C with 0.25 mm S-benzyl-l-cysteine and 0.2 mm acetyl-CoA. The product of the enzymatic activity was purified from the substrates by reverse phase HPLC and analyzed by mass spectrometry. Tandem mass spectrum of the putative N-acetyl-benzyl-l-cysteine was acquired. The structure of N-acetyl-benzyl-l-cysteine and the assignment of the main fragments are shown.

FIGURE 5.

Partial co-localization of NAT8-Myc with the ER marker KDEL and with NAT8L but full segregation from the Golgi complex (GM130) and from mitochondria (MitoTracker). Chinese hamster ovary cells were transfected to express at low levels Myc-tagged NAT8 and immunolabeled for c-Myc (center column, green in merge column at right). a and b, double immunolabeling with either KDEL-bearing proteins (peripheral ER marker) or with GM130 (Golgi) (left, red in merge column). c, pEF6/Myc-NAT8-transfected cells were incubated before fixation with MitoTracker (left, red in merge column). Mitochondrial pattern is identical in nontransfected (N′) and transfected cells (N). d, cells were co-transfected for low expression of nontagged NAT8L, which was immunolabeled using affinity-purified antibodies (left, red in merge column). Notice labeling of NAT8 at the nuclear envelope (filled arrowheads) in continuity (open arrowheads) with a cytoplasmic reticular pattern, showing partial co-localization with the peripheral ER marker KDEL (arrows in a; for intensity profile, see supplemental Fig. S2) and with nontagged NAT8L (arrows in b), contrasting with full segregation from the Golgi (b) and from mitochondria (c). Scale bars, 5 μm. For line intensity profile, see supplemental Fig. S2. For immunolabeling of N-terminal His₆-tagged NAT8, see supplemental Fig. S1.

FIGURE 6.

CCNAT activity of human wild-type (WT) and mutant NAT8s overexpressed in HEK293T cells. a, specific CCNAT activity was assayed in cell extracts of HEK293T cells overexpressing wild-type (Glu-104/Phe-143) or mutant NAT8 proteins with a His₆ tag in the N or C terminus. b, relative expression levels of the mutant NAT8 proteins versus the wild-type NAT8 were evaluated by Western blot analysis using the anti-His antibody from GE Healthcare (to detect N-terminal tagged NAT8s) and the penta-His antibody from Qiagen (to detect C-terminally tagged NAT8s). c, specific CCNAT activity of each extract was corrected taking into account the differences in the expression for all NAT8 proteins.

FIGURE 7.

Cytotoxicity in HEK293T cells caused by overexpression of active and inactive human NAT8s. Media were collected from cultures of HEK293T cells transfected for 24 or 48 h with either an empty pEF6-HisB plasmid (control) or pEF6-HisB expressing active (NAT8-WT and NAT8-E104K) or inactive (NAT8-R149K) forms of NAT8. Cell mortality was determined using CytoTox-96 cytotoxicity assay (Promega) as described under “Experimental Procedures.” Results are means ± S.E. for measurements in the culture media of three independent transfection experiments.

FIGURE 8.

Phylogenetic tree of NAT8L and NAT8-related proteins. Phylogenetic tree was inferred from protein sequences using Bayesian analysis (MrBayes) methods. Posterior probabilities (MrBayes) are indicated at the nodes. The horizontal bar represents 20 substitutions per 100 amino acid residues. The following NAT8L and NAT8-related sequences are shown: H. sapiens (HsapiensNAT8L, gi:259016335; HsapiensNAT8, gi:47115157; HsapiensNAT8B, gi:117168283); Pan troglotydes (PtroglotydesNAT8, gi:114578078; PtroglotydesNAT8B, gi:114578086); Macaca mulatta (MmulattaNAT8, gi:109103385); Equus caballus (EcaballusNAT8B847, gi:149727808; EcaballusNAT8B571, gi:194220604); C. familiaris (CfamiliarisCML2, gi:7398098); Rattus norvegicus (RnorNAT8L, gi:62660669; RnorvegicusCML1, gi:81882032; RnorvegicusCML2, gi:109472305; RnorvegicusCML3, gi:109472294; RnorvegicusCML4, gi:12018332; RnorvegicusCML5, gi:198278521; RnorvegicusLOC681227, gi:109472307; RnorvegicusNAT8B, gi:19424238); M. musculus (MmusculusNAT8L, gi:134288912; MmusculusCML1, gi:14789811; MmusculusCML2, gi:31542409; MmusculusCML3, gi:15928427; MmusculusCML4, gi:12963681; MmusculusCML5, gi:19353130; MmusculusGm4477, gi:149255301; Mmusculus1700019G17, gi:23468249; MmusculusNAT8B, gi:83029409); Monodelphis domestica (MdomesticaNAT8L, gi:126331993; Mdomestica21822, gi:126305502; Mdomestica22265, gi:126305520; Mdomestica22287, gi:126305522; Mdomestica22311, gi:126305524; Mdomestica22341, gi:126305526; Mdomestica22462, gi:126305528); Ornithorhynchus anatinus (Oanatinus83900, gi:149445727; Oanatinus86348, gi:149465132); G. gallus (GgallusNAT8B, gi:118110777); Xenopus silurana (Xsilurana145252, gi:187607346; XsiluranaLOC100124951, gi:156717640; XsiluranaNAT8, gi:56605892); Xenopus laevis (XlaevisCamello, gi:147901566; XlaevisLOC414583, gi:148223311); and D. rerio (DrerioNAT8L, gi:139947555; DrerioCML5a, gi:125833023; DrerioCML5b, gi:189526706).