We searched for SNAT homologues in S. pseudintermedius ED99, as a model of SER-producing staphylococci using cSNAT identified in the cyanobacterium Synechocystis sp. PCC 6803 (GenBank accession no. NP_442603) as a reference. cSNAT is a 171 amino acids long protein possessing an N-terminal sequence comprising at least 22 amino acid residues that have no effect on its enzymatic activity (Byeon et al., 2013). The inactive N-terminal domain is typically found in SNATs in higher organisms and usually serves as a translocation signal.

RefSeq database search was performed to identify homologues in ED99 using BLASTP (O’Leary et al., 2016). The three protein hits that were identified and annotated as SPSE_0802, SPSE_0436 and SPSE_1761 are 140, 138 and 142 amino acids long, respectively. SPSE_0802, SPSE0_436 and SPSE_1761 are annotated as GNAT (GCN5-related N-acetyltransferases) family peptides and they share 19, 18, and 10% identity (34, 33, and 16% similarity) with cSNAT, respectively (Table 1). Comparison to the fully functional truncated cSNAT_23-171 increases the similarity to 39, 38, and 19% to each of SPSE_0802, SPSE0_436 and SPSE_1761, respectively.

Secondary structure prediction of the three proteins suggests that they do possess neither a signal peptide nor a transmembrane domain. Therefore, the enzymes were predicted to be localized in the cytoplasm similarly to previously identified SNATs. Tertiary structure prediction shows the presence of a conserved acetyl-CoA binding domain in all of the four aligned proteins (Figure 1A). SPSE_0802 shares notable structural similarity with cSNAT presented by a common fold, comprised of 6–7 antiparallel β -strands (5 present in cSNAT and SPSE_0802) and 4 α -helices in the topology β 1- α 1- α 2- β 2- β 3- β 4- α 3- β 5- α 4- β 6- β 7 (Figure 1B).

To determine whether any of the three identified proteins from ED99 are SNAT enzymes, the corresponding genes were cloned into the bacterial expression vector pET28a and expressed as C-terminal His-tagged fusion proteins in E. coli BL21. His-tagged SPSE_0802, SPSE_0436 and SPSE_1761 proteins were isolated from the cytoplasmic fraction of E. coli BL21 clones and purified by Ni-NTA superflow resin (Figure 2).

The enzymatic activity of the recombinant proteins was assayed by measuring the N-acetylation of 1 mM SER or tryptamine in presence of 1 mM acetyl-CoA as co-factor at 37°C. SPSE_0802 could acetylate SER and TRY into NAS and NAT, respectively as detected by high performance liquid chromatography (HPLC) analysis (Figures 3A,B). At pH 8.0, SPSE_0802 acetylated around 18% of the total SER and 23% of the total tryptamine after 6 h incubation. At pH 6.8, a lower activity was detected, as only 12% of SER and 16% of tryptamine were acetylated after the same incubation period. In both conditions, however, SPSE_0802 had a slightly higher activity when TRY was used as a substrate. These results match the higher activity at pH 8.0–9.0 and the higher specificity using TRY as substrate reported for other SNATs (Byeon et al., 2013; Kang et al., 2013; Lee et al., 2014). HPLC analysis of the enzymatic reactions using recombinant SPSE_1761 and SPSE_0436 under the same conditions did not detect any N-acetylated products, indicating that neither of these two proteins is a functional SNAT enzyme (Figure 3A).

SPSE_0802 could also acetylate other substrates such as dopamine, phenethylamine, and tyramine (Figure 4), which shows the wide spectrum of substrate specificity of the enzyme. There was no acetylation of SER and TRY in absence of SPSE_0802, which rules out the possibility of non-enzymatic acetylation of the compounds. To further confirm the activity of SPSE_0802, the recombinant enzyme was inactivated by incubation at 95°C for 10 min before being added to the reaction mixture. In this case, neither NAS nor NAT were detected by HPLC analysis of the reaction after 6 h.

The N-acetylation reaction catalyzed by SPSE_0802 was acetyl-CoA dependent as no activity was seen in the absence of acetyl-CoA. The reaction was also time-dependent, as more acetylated products accumulated after a longer incubation time. The acetylated products were quantified using the standard curves (Supplementary Figure S1) and confirmed by comparing their retention time to synthetic standards in HPLC.

To investigate the role of SPSE_0802 in N-acetylation of monoamines in S. pseudintermedius ED99, we constructed the deletion mutant ED99ΔSPSE_0802 and compared both strains’ ability to synthesize NAS. Due to its hydrophilic properties, exogenous SER needs to be actively transported into the cell cytoplasm in order to be processed by SPSE_0802. Since ED99 cannot uptake SER from the extracellular medium, we used 5-HTP, the substrate of SadA decarboxylation and precursor of SER. We have previously shown that the decarboxylation reaction by SadA takes place in the cytoplasm where SER is produced (Luqman et al., 2018). Accordingly, we expected 5-HTP to be processed into NAS in a two-step process: decarboxylation to SER by SadA, followed by N-acetylation to NAS by SPSE_0802.

S. pseudintermedius ED99 WT and ΔSPSE_0802 cells were fed with 5 mM 5-HTP and incubated overnight at 37°C. HPLC analysis of the culture supernatant of cells fed with 5-HTP shows that ED99 WT produced around 24 nM NAS whereas ΔSPSE_0802 mutant produced around 4 nM NAS (Figure 5). These results verify that ED99 can produce NAS and that SPSE_0802 plays an important role in this production. The residual amounts of NAS in the cultures of the ED99 ΔSPSE_0802 mutant suggest the presence of another mechanism or enzyme that could produce NAS with a much lower substrate specificity than SPSE_0802.

SNAT enzymes in plants were recently found to be implicated in an alternative pathway in which they catalyze the conversion of 5-methoxytryptamine (5-MT) directly to melatonin (Back et al., 2016; Tan et al., 2016). To test whether the identified SPSE_0802 could also catalyze this reaction in staphylococci, we searched for melatonin production by S. pseudintermedius ED99 and its congenic deletion mutant ED99ΔSPSE_0802. No melatonin was detected by HPLC analysis of the cells from overnight cultures fed with 5-MT which indicates that SPSE_0802, unlike plant SNAT, cannot use 5-MT as a substrate and that ED99 does not harbor the enzyme involved in the last step of melatonin biosynthesis from N-acetylserotonin (namely, the acetylserotonin o-methyltransferase ASMT) (results not shown).

To verify whether other staphylococci possess a SNAT activity like S. pseudintermedius ED99, we tested the ability of 40 different staphylococcal strains belonging to 15 cluster groups to synthesize NAT from TRY as a substrate. HPLC analysis of the bacterial culture supernatants revealed that 22 of the 40 tested strains were able to produce NAT (Table 1). This indicates that SNAT is widespread in the genus Staphylococcus.

As many of the tested staphylococcal strains could produce NAT, we searched for SPSE_0802 homologues in the genus Staphylococcus and found 13 hits based on amino acid sequence availability on NCBI. Among the 13 hits, 9 were tested as NAT producers in our experiment. The homologous proteins and accession numbers are listed in Table 2. The predicted 3D structures of the homologues found in NAT-producing strains are presented with that of cSNAT in Figure 6. We note that cSNAT and the homologues in staphylococci share the previously described organization of β -helices and α -sheets. The protein alignment of the SPSE_0802 homologues and the organization of their corresponding genes are presented in Supplementary Figures S2, S3, respectively.

Bacterial strains and plasmids used in this study are listed in Supplementary Table S1. For cloning procedures in S. pseudintermedius ED99, the genome sequence with GenBank accession number NZ_CP065921.1 was used as a reference. S. pseudintermedius strains were grown in tryptic soy broth (TSB) and E. coli strains were grown in Luria Bertani (LB) medium. Bacteria were cultivated aerobically (200 rpm) at 37°C. Each experiment was started from an overnight preculture by adjusting the OD_578nm to 0.05–0.1 in the corresponding medium. The medium was supplemented with the following antibiotics, where applicable, at the indicated final concentrations: chloramphenicol at 10 μgml−1 for staphylococcal strains and 30 μgml−1 kanamycin for E. coli strains.

The protein sequence of serotonin N-acetyltransferase of Synechocystis sp. (Byeon et al., 2013) (GenBank accession number WP_010873901.1) was used to search the homologous protein in S. pseudintermedius ED99 using BLASTP. The protein hits were identified as putative SNAT enzymes.

Expression plasmids were constructed using pET28a (Novagen). pET28a was linearized, using the restriction enzymes NotI and NcoI to add a histidine tag to the C-terminal end, and ligated with either of the 3 genes encoding SPSE_1761, SPSE_0802 or SPSE_0436 amplified from S. pseudintermedius ED99 genome (oligonucleotides listed in Supplementary Table S2). The ligations were performed using Hi-Fi DNA Assembly Master Mix (New England Biolabs), then transformed into E. coli DC10B by heat shock method. The colonies grown on selective agar media containing kanamycin were confirmed by plasmid isolation and sequencing. The plasmids were then transformed into the expression host E. coli BL21 (DE3). The clones containing the correct plasmid were cultured overnight at 25°C in LB supplemented with kanamycin. The cells were harvested and lysed using 0.1 mm glass beads in FastPrep instrument (MP Biomedicals). The cell lysate was centrifuged (15,000 g, 30 min, 4°C) and the proteins in the supernatant were subjected to protein purification using Ni-NTA superflow resin (IBA). The expression and purification steps were verified by 14% SDS-PAGE. The protein concentration was determined by the Bradford method using a protein assay dye (Bio-Rad, Hercules, CA, United States).

The purified proteins were subjected to in vitro enzymatic assay. The purified recombinant proteins (200 μg) were incubated in a total volume of 1 mL containing 1 mM substrate and 1 mM acetyl-CoA in 50 mM potassium phosphate (pH 6.8 or 8) at 37°C. The assay was performed using serotonin as substrate as well as dopamine, tryptamine, phenethylamine, and tyramine. At the indicated time points, the reaction was stopped by adding 250 μL methanol, and the samples were stored at −20°C before being analyzed by HPLC. For the heat inactivation of SPSE_0802, the recombinant protein was incubated at 95°C for 10 min before being added to the reaction mixture.

The in vitro enzymatic assays and the bacterial supernatants were analyzed using reversed-phase HPLC (RP-HPLC) as previously described (Luqman et al., 2020a; Luqman et al., 2020). Briefly, the HPLC analysis was performed at room temperature with an Eclipse XDB-C18 column (4.6150 mm; 5 m) (Agilent) and an analytical guard column for Eclipse XDB-C-18 (4.6 12.5 mm; 5 m) (Agilent) with a 15 min linear gradient of 0.1 percent phosphoric acid to acetonitrile and 5 min post time washing with an injection flow rate of 1.5 mL/min, and a sample volume of 10 μL. As a reference, diode array detectors (DAD) were employed at 210 and 360 nm.

The null mutant lacking the putative serotonin N-acetyltransferase-encoding gene in S. pseudintermedius ED99 was constructed using the plasmid pBASE6 (Geiger et al., 2012). Briefly, pBASE6 was linearized using EcoRV and ~ 1,000 bp upstream and ~ 1,000 bp downstream fragments of the gene of interest were amplified from the genomic DNA of S. pseudintermedius ED99. The fragments were fused by Gibson Assembly (Gibson et al., 2009) using Hi-Fi DNA Assembly Master Mix (New England Biolabs). The resulting plasmid was first introduced into E. coli DC10B (Monk et al., 2012) and then into ED99. The deletion mutant construction was performed as previously described (Bae and Schneewind, 2006). Deletion of the genes were confirmed by PCR and sequence analysis. The mutants were named as described in the Supplementary Table S1. Oligonucleotides used are listed in the Supplementary Table S2.

HPLC analysis of bacterial cell lysates was performed to determine the production of NAS by S. pseudintermedius ED99. The strains were cultured in 10 mL TSB supplemented with 5 mM 5-hydroxytryptophan (5-HTP) and incubated overnight at 37°C with shaking at 150 rpm. Then, the cells were harvested, resuspended in 2 mL potassium phosphate buffer (pH 7.2) and lysed using 0.1 mm glass beads in FastPrep instrument (MP Biomedicals). The cell lysate was centrifuged for 30 min, 5,000 × g, 4°C. The supernatant was collected and stored at −20°C until analyzed by HPLC.

HPLC analysis of bacterial cultures supernatants was performed to determine the production of NAT by different Staphylococcus strains. The strains were cultured in TSB overnight. Then, the cells were harvested, washed with PBS (pH 7.2), and resuspended in PBS (pH 7.2) supplemented with 1% glucose and 5 mM TRY and incubated overnight at 37°C with shaking at 150 rpm; the cell density was kept relatively high (OD₅₇₈ = 50). The cultures were then analyzed using HPLC to determine the presence of NAT.

For each experiment, the results were expressed as the mean value ± SEM of data from 3 to 5 replicates. All the statistical analyses were performed using GraphPad Prism software, and a p-value of <0.05 was considered statistically significant. Statistical test choice and significance are indicated in the figure legends.