Acetylation by blood cells☆

Human N-acetyltransferase 1 (NAT1) is a widely distributed enzyme that catalyses the acetylation of arylamine and hydrazine drugs as well as several known carcinogens, and so its levels in the body may have toxicological importance with regard to drug toxicity and cancer risk. Recently, we showed that p-aminobenzoic acid (PABA) was able to down-regulate human NAT1 in cultured cells, but the exact mechanism by which PABA acts remains unclear. In the present study, we investigated the possibility that PABA-induced down-regulation involves its metabolism to N-OH-PABA, since N-OH-AAF functions as an irreversible inhibitor of hamster and rat NAT1. We show here that N-OH-PABA irreversibly inactivates human NAT1 both in cultured cells and cell cytosols in a time- and concentration-dependent manner. Maximal inactivation in cultured cells occurred within 4 hr of treatment, with a concentration of 30 μM reducing activity by 60 ± 7%. Dialysis studies showed that inactivation was irreversible, and cofactor (acetyl coenzyme A) but not substrate (PABA) completely protected against inactivation, indicating involvement of the cofactor-binding site. In agreement with these data, kinetic studies revealed a 4-fold increase in cofactor K_m, but no change in substrate K_m for N-OH-PABA-treated cytosols compared to control. We conclude that N-OH-PABA decreases NAT1 activity by a direct interaction with the enzyme and appears to be a result of covalent modification at the cofactor-binding site. This is in contrast to our findings for PABA, which appears to reduce NAT1 activity by down-regulating the enzyme, leading to a decrease in NAT1 protein content.

Human arylamine N-acetyltransferase type 1 (NAT1) has been cloned from human genomic DNA, into the vector pET(5a) and expressed in Escherichia coli. The recombinant protein has been purified to apparent homogeneity using anion exchange chromatography. The arylamine acceptor specificity, and the effect of potential NAT1 inhibitors has been investigated using purified recombinant protein. The K_m of the recombinant NAT1 protein for the substrates para-aminobenzoate (p-aba) and 4-aminosalicylate are 14.3 and 11.8 μM, respectively. Folate and amethopterin were found to be potent competitive inhibitors of p-aba acetylation, with K_i values of 13.3 and 9.5 μM, respectively. The pteroate moiety of folate, in contrast is a poor inhibitor, with 100 μM pteroate inhibiting only 40% of NAT1 activity. A catabolite of folate para-aminobenzoly-l-glutamate has also been shown to be a NAT1 substrate with a K_m value of 263 μM.

Human and rat whole blood were shown to metabolize the aromatic amides acetanilide and phenacetin by deacetylation followed by reacetylation in vitro. Derivatives of the parent compounds labelled with deuterium in the N-acetyl group produced non-labelled material after incubation. The reaction was monitored by capillary gas chromatographic-mass spectrometric (GC-MS) analysis. There was no significant difference in the acetyl group exchange of these substrates using blood samples donated by non-diabetic volunteers or Type 2 diabetic patients (respective mean ± SEM values = 4.0 ± 0.2% and 4.2 ± 0.3% for trideuteroacetanilide, 6.2 ± 0.6% and 6.1 ±0.3% for trideuterophenacetin). Increasing the glucose concentration in the incubation medium by 50 mmol/L significantly (P < 0.01) increased deacetylation-reacetylation of trideuteroacetanilide in each group (4.6 ± 0.2% and 4.7 ± 0.2% for non-diabetic and diabetic subjects, respectively). In rat blood the amount of deacetylation-reacetylation was much higher: 7.2 ± 0.6% and 8.3 ± 0.7% for trideuteroacetanilide and trideuterophenacetin, respectively. Induction of experimental diabetes using streptozotocin did not significantly change the extent of deacetylation-reacetylation of either deuterated substrate (10.1 ± 2.1% and 9.5 ± 1.1%). Elevation of the incubation glucose concentration by 50 mmol/L produced an increase in acetyl group exchange (for trideuteroacetanilide) in diabetic (14.3 ± 2.2%) and non-diabetic (10.6 ± 1.0%) rats. The donation of acetyl groups (transacetylation) was observed after incubation of blood samples from both diabetic and non-diabetic human subjects and rats with trideuterophenacetin and a molar excess of aniline. This reaction significantly (P < 0.001) decreased the acetyl group exchange of trideuterophenacetin (these values were 4.5 ± 0.4% and 3.4 ± 0.6% using samples from non-diabetic human subjects and rats, respectively) and demonstrated the ability of whole blood to catalyse transacetylation (acetyl-CoA-independent acetylation). There was correlation between the amount of (unlabelled) acetanilide produced by acetylation with acetyl-CoA and the percentage present as trideuteroacetanilide. The proportion of trideuteroacetanilide was higher using rat blood (e.g. the values for non-diabetic subjects were 25.5 ± 1.7% vs 8.5 ± 0.3%; P < 0.001) although the total amount of acetanilide produced was lower (0.54 ± 0.14 nmol vs 1.82 ± 0- 0.23 nmol; P < 0.05) than that observed using human blood.

Induction of experimental diabetes using streptozotocin significantly reduced the extent of sulphamethazine acetylation by Sprague-Dawley rats. This treatment did not significantly change the total amount of sulphonomide excreted in the urine. The in vitro blood N-acetyltransferase activity of rats treated with streptozotocin was significantly higher than that of untreated animals. Increasing the in vitro glucose concentration of blood samples from both groups significantly increased the amount of acetylsulphamethazine produced.

Acetylation of p-aminobenzoic acid was studied in human blood cell lysates. The rate of acetylation with acetyl-coenzyme A was 3.4 nmoles per min per 0.5 ml lysate (corresponding to 0.5 ml blood with a 50% hematocrit). In the presence of a coenzyme A generating system, the rate was only 0.1 nmole per min per 0.5 ml lysate. N-Acetyltransferase (acetyl-CoA : arylamine N-acetyltransferase, EC 2.3.1.5) activity exhibited a temperature optimum within the range of 34–37° with a Q₁₀of 2 between 24–34°. Heating for 3 min at 50° caused 50 per cent inactivation of enzymatic activity. The pH activity profile showed an optimum at pH 6.0–6.5. p-Chloromercuribenzoate was a potent inhibitor causing 50 per cent inhibition at 9 μM. The apparent K_m value for p-aminobenzoic acid was 0.4 mM and for acetylcoenzyme A, 0.3 mM. The enzyme activity with p-aminosalicyclic acid was about 50 per cent that obtained with p-aminobenzoic acid. Acetylation of sulfamethazine was either very low or in several individuals undetectable. Isonicotinic acid hydrazide at a concentration 10 times that of p-aminobenzoic acid did not interfere with acetylation of the latter, nor did trimethoprim, another compound with an amine moiety. Folic acid and amethopterin competitively inhibited the acetylation of p-aminobenzoic acid, with respective K_i values of 0.01 mM and 0.06 mM. It is concluded that the activity of N-acetyltransferase in human blood may have important clinical implications.