Synthesis of typical sulfonamide antibiotics with [14C]- and [13C]-labeling on the phenyl ring for use in environmental studies

Background Due to their widespread use, sulfonamide antibiotics (SAs) have become ubiquitous environmental contaminants and thus a cause of public concern. However, a complete understanding of the behavior of these pollutants in complex environmental systems has been hampered by the unavailability and high cost of isotopically labeled SAs. Results Using commercially available uniformly [14C]- and [13C]-labeled aniline as starting materials, we synthesized [phenyl-ring-14C]- and [phenyl-ring-13C]-labeled sulfamethoxazole (SMX), sulfamonomethoxine (SMM), and sulfadiazine (SDZ) in four-step (via the condensation of labeled N-acetylsulfanilyl chloride and aminoheterocycles) or five-step (via the condensation of labeled N-acetylsulfonamide and chloroheterocycles) reactions, with good yields (5.0–22.5% and 28.1–54.1% for [14C]- and [13C]-labeled SAs, respectively) and high purities (> 98.0%). Conclusion The synthesis of [14C]-labeled SAs in milligram amounts enables the preparation of labeled SAs with high specific radioactivity. The efficient and feasible methods described herein can be applied to the production of a variety of [14C]- or [13C]-labeled SAs for studies on their environmental behavior, including the fate, transformation, and bioaccumulation of these antibiotics in soils and aqueous systems. Supplementary Information The online version contains supplementary material available at 10.1186/s12302-022-00598-z.


Background
Sulfonamide antibiotics (SAs) are widely used in the treatment of human disease and in modern animal husbandry. However, due to their poor biodegradation and insufficient removal by wastewater treatment plants [1,2], high concentrations of sulfadiazine (SDZ), sulfamethoxazole (SMX), and sulfamonomethoxine (SMM) are commonly detected in soils, sediments, rivers and other environmental media [3][4][5]. SAs that enter the environment exert adverse effects on organisms [6][7][8][9], thus raising public concern. A comprehensive understanding of the environmental fate of SAs, including their adsorption, biodegradation, transformation, formation of nonextractable residues (NERs), and transport, is essential to assessing their environment risks.
Studies of the environmental behavior of pollutants often rely on the use of [ 14 C]-radioactive and [ 13 C]-stable isotopes. The advantages of [ 14 C]-tracers include their low detection limit and convenient handling with complex environmental samples. Consequently, they are frequently used to investigate the environmental fate of organic pollutants, especially mineralization and NERs formation. For example, [ 14 C]-tracers have been employed to examine the environmental impacts of pesticides, brominated flame retardants, alkylphenols, and polycyclic aromatic hydrocarbons [10][11][12][13][14]. Stable isotopes (e.g., 13 C, 15 N) have been used in mass spectrometry and nuclear magnetic resonance (NMR) spectroscopy studies to quantify and identify metabolites of pollutants in complex matrices [15][16][17][18][19][20]. [ 13 C]-tracers provide powerful tools for investigations of microbial biomass and community composition and have thus been used in phospholipid fatty acid analyses and as DNA stable-isotope probes [21,22]. This wide range of applications has increased the demand for [ 14 C]-and [ 13 C]-labeled SAs, but these isotopes are either commercially unavailable or too expensive. An efficient, simple method allowing the ''in house'' synthesis of [ 14 C]-and [ 13 C]-SAs, especially on micro-scales with good yields is therefore needed.
The successful synthesis of [ 14 C]-SAs on a micro-scale requires stable solvents, suitable reaction conditions, and simple purification methods [23]. The conventional method of synthesizing unlabeled SAs consists of four steps: acetylation of aniline using acetic anhydride, chlorosulfonation of N-acetylaniline with ClSO 3 H, condensation of sulfonyl chloride with nucleophiles such as amines, and alkaline hydrolysis of the acetyl-protecting group [24][25][26][27]. However, the synthetical conditions are suitable for obtaining SAs in gram amounts and cannot be down-scaled to synthesize SAs in milligram quantities, due to the difficulty of mixing under solvent-free conditions and the crystallization of the products. In a previous study, [ 14 C]-SDZ labeled on the heterocyclic ring was prepared via the reaction of N-acetylsulfanilyl chloride with [ 14 C]-2-aminopyrimidine [28]. By contrast, [ 14 C]-labeling of the phenyl ring of SDZ and other common SAs (such as SMM and SMX), required to trace the transformation of phenyl ring of SAs, has yet to be accomplished.
In this study, we report methods for preparation of typical SAs with [ 14 C]-or [ 13 C]-labeling of the phenyl ring with good yields, especially the synthesis of [ 14 C]-labeled SAs on a micro-scale (milligram-level). These methods can be employed to prepare a variety of [ 14 C]-or [ 13 C]-labeled SAs.

Analyses
The reaction products were purified by flash column chromatography (CHEETAH TMMP100; Agela, Tianjin, China) or preparative thin-layer chromatography (TLC) on preparative silica gel plates (GF254, 1 mm, 20 × 20 cm; Huanghai, Shandong, China). The purity of the products was analyzed on analytical silica gel plates (GF254, 0.25 mm, 3 cm × 10 cm, Huanghai, Shandong, China) by analytical TLC coupled to an imaging scanner (Typhoon Trio + ; GE Healthcare, U.S.), or by high-performance liquid chromatography (HPLC, 1100 system; Agilent Technology, USA). The synthesized products were identified on an HPLC system (1260; Agilent Technology, USA) coupled to a Q-TOF tandem mass spectrometer (HPLC-Q-TOF-MS/MS, triple TOF 5600 system; AB SCIEX, USA) and by NMR spectroscopy (AVANCE III HD-500; Bruker, Germany). Radioactivity was determined by liquid scintillation counting (LSC, LS6500; Beckman Counter, USA). The specific activities of 5a, 7a, and 10a were calculated from the chemical masses determined by HPLC and the amounts of radioactivity determined by LSC. Details of the instruments used in the purifications and analyses are provided in the Additional file 1.
Next, 2a (2.48 × 10 8 Bq, 0.22 mmol) in CCl 4 (0.5 mL) was diluted with unlabeled N-acetylaniline (23.5 mg, 0.17 mmol). ClSO 3 H (170 μL, 2.52 mmol) was added dropwise with stirring in an ice bath. After the mixture had been stirred at 58 °C for 2 h, SOCl 2 (25 μL, 0.34 mmol) was added. The resulting solution was heated for another 2 h at 58 °C (Fig. 1, Method II), cooled to room temperature, and extracted twice with ethyl acetate (35 mL each). The extracts were dried with anhydrous Na 2 SO 4 and evaporated to ~ 1 mL. The purity of the product, 3a (1.98 × 10 8 Bq), in the mixture was 93.0%, as determined by TLC, using petroleum ether:ethyl acetate (1:4 / v: v) containing 0.2% CH 3 COOH as the eluent (R f = 0.35), coupled to autoradiography. The mixture without purification was directly used for the synthesis of 8a. The yield of 3a according to its purity in the mixture was 74.3%.
Uniformly [phenyl-ring-14 C]-labeled SDZ (10a) Crude 9a (9.25 × 10 7 Bq, 6.29 × 10 8 Bq/mmol, 57.0% radiochemical purity) was reacted with NaOH solution (10%, 5 mL) for 3 h at 100 °C and neutralized with 6 M HCl to pH 6. The product was extracted eight times with ethyl acetate (15 mL each). The extract was dried with anhydrous Na 2 SO 4 , evaporated to ~ 0.5 mL, and the crude product then recrystallized from boiling methanol. The precipitate was centrifuged and washed three times with methanol, resulting in 10a (3.11 × 10 7 Bq, 6.29 × 10 8 Bq/mmol). The purity was 98.3% as determined by HPLC (t R = 5.73 min. For details, see Additional file 1). The supernatant was further extracted five times with ethyl acetate (15 mL each). The extract was dried with anhydrous Na 2 SO 4 and evaporated to dryness, resulting in solids containing 10a. These were mixed with unlabeled SDZ (54 mg) and then recrystallized from boiling methanol. The precipitate was washed three times with methanol, resulting in another portion of 10a with a low specific activity (1.10 × 10 7 Bq, 7.40 × 10 7 Bq/mmol) and a radiochemical purity of 98.3%. The total amount of 10a was 4.21 × 10 7 Bq, with a total yield of 79.9%. The chemical structure of 10a was elucidated by 1 H-NMR, 13 C-NMR (Additional file 1: SI.5), and LC-Q-TOF-MS/MS (Additional file 1: SI.6) using the corresponding unlabeled compounds synthesized according to the same procedures.
Synthesis of 8b: Ammonium hydroxide (5 mL, 28% NH 3 in water) was mixed vigorously with 3b (1.0 g, 4.3 mmol, 99% of 13 C atom) in acetone (10 mL) in an ice bath. After 1 h of stirring at 25 °C, the acetone was removed by evaporation, ice-cold water was added, and the pH was adjusted to 6 with 6 M HCl. Filtration and washing of the mixture with ice-cold water resulted in 8b (672 mg, 99% of 13 C atom, 98.0% purity) (Additional file 1: SI.4) with a yield of 73.0%.

Results and discussion
SMX, SMM, and SDZ with uniform 13 C and 14 C labeling on the phenyl ring were prepared from commercially available, labeled aniline in a four-step or five-step synthesis (Fig. 1). The yields and radiochemical or chemical purities of the products are reported in Fig. 1. Three unlabeled SAs and the respective intermediates were similarly synthesized and characterized using HPLC-Q-TOF-MS/ MS, and NMR (Additional file 1: Table S2).

Synthesis of [ 14 C]-or [ 13 C]-SMX, [ 14 C]-or [ 13 C]-SMM, and [ 14 C]-or [ 13 C]-SDZ
Chlorosulfonation of aniline on the para-position of the amino group using ClSO 3 H was the key step in the synthesis of SAs. Prior to this step, the aniline was acetylated to prevent possible oxidation of the amino group and bis-sulfonation on the ring during chlorosulfonation. The acetylation was performed in aqueous solution, and K 2 CO 3 was added to improve nucleophilic activity of aniline (1a), resulting in acetylaniline (2a), with a good yield of 90.0%. Our preparation of 2a with fewer procedures was more convenient than the previously reported method [29].
Chlorosulfonation of 2a with ClSO 3 H generated the key intermediate 3a, a precursor in the synthesis of a variety of [ 14 C]-SAs labeled on the phenyl ring via reactions with different amino heterocycles and subsequent alkaline hydrolysis. In a previous study, 1.1 g of 2a at a high molar ratio of ClSO 3 H to 2a (18:1) was used to obtain 3a, which formed a white solid after crystallization in water [29]. However, this method of using a high volume of ClSO 3 H cannot be applied to synthesize 3a at milligram scale (12 mg of 2a), because the hot H 2 SO 4 , derived from the hydrolysis of excess ClSO 3 H in water, causes the decomposition of 3a, resulting in a very low yield. HPLC-Q-TOF-MS/MS also showed the conversion of a large amount of 3a to N-acetylsulfanilic acid (data not shown). In addition, the solvent-free condition used in previous studies may result in an inhomogeneous mixture of the reactants at a micro-scale. Nguyen-Hoang-Nam et al. [28] also found that it was difficult to synthesize sulfonyl chloride in a small amount and thus failed to obtain micro-quantities (100 mg) of N,N-di(2chloro-n-propyl)aminobenzenesulfonyl chloride labeled on the phenyl ring by chlorosulfonation with ClSO 3 H and the corresponding [ 14 C]-sulfonamide derivatives. Therefore, we used a low molar ratio of 1:7.4 in solvent CCl 4 , and added NaCl to the reaction mixture to consume the by-product H 2 SO 4 . These modifications not only completely converted 2a, they also reduced the decomposition of 3a to N-acetylsulfanilic acid by hot H 2 SO 4 , such that 3a was produced in a good yield (53.9% after purification; Fig. 1).
Water inhibits the condensation of 3a with amino heterocyclic compounds (e.g., 11 and 12). In our method, water interference was avoided by the inclusion of molecular sieves to adsorb the water during the condensation. With this method, 4a and 6a were obtained in yields of 51.0% and 15.6%, respectively (Fig. 1).
The condensation of 3 with amino heterocycles involved a nucleophilic substitution. Compound 11 had a higher nucleophilic activity than compound 12, according to their electron cloud density, which was in agreement with the higher yield of 4a than 6a (51.0% vs. 15.6%) and of 4b than 6b (73.8% vs. 42.3%) (Fig. 1). The condensation of 3 with other heterocyclic compounds can be used to synthesize other [ 14 C]-or [ 13 C]-labeled sulfonamides, such as the synthesis of SDZ from 2-aminopyrimidine [28]. However, owing to the low nucleophilic activity of 2-aminopyrimidine, the yield of 10a at micro-scale was very low (7.4%; overall yield of 10a from 1a: 2.4%) and the yield of 9b (21.0%) was lower than that of either 4b (73.8%) or 6b (42.3%) (Fig. 1). Therefore, 10a and 10b were synthesized in a five-step synthetic pathway (Fig. 1), in which two steps were used to synthesize 9 instead of one step. First, 8 was synthesized by the condensation of 3 with ammonium hydroxide, which has high nucleophilic activity and is a base capable of neutralizing the by-product H 2 SO 4 . Good yields were achieved for both 8a (98.3%) and 8b (73.0%). The coupling of 8 to 13 produced both 9a and 9b in good yields of 36.0% and 74.8%, respectively. The synthesis of 9 from 3 via this two-step pathway not only completely converted 3 to 8 with a high stability, thus avoiding the decomposition of 3, it also resulted in a much higher overall yield than the one-step reaction (35.4% vs. 7.4% for 9a, 54.6% vs. 21.0% for 9b).

C-NMR of [ 13 C]-SMX, [ 13 C]-SMM, and [ 13 C]-SDZ
The 13 C-NMR spectra of three [ 13 C]-SAs and their corresponding unlabeled compounds are shown in Fig. 2. The significant triplet signals allowed the assignment of the signals at 112.53-112.98 ppm, 124.46-125.08 ppm, 129.16-130.16 ppm, and 153.25-153.51 ppm to 13 C-atoms of the benzene ring. 13 C-tracers can provide more structural information about the fate and behaviors of labeled C-atoms in environmental matrixes than obtained with radioactive [ 14 C]-tracers [30]. The peaks of the C-atoms in 13 C-labeled compounds are split into triplets due to 13 C- 13 C coupling and they have a much higher intensity than those in non-labeled compounds containing 13 C-atoms in natural abundance (1.1%). Accordingly, the triplet signals can be used to identify the chemical nature of labeled carbon atoms, as demonstrated for the residues of pesticides (e.g., cyprodinil), humus monomers (e.g., catechol), and emerging pollutants (e.g., tetrabromobisphenol A) bound to soil humic substances [31][32][33], thus providing unambiguous information about the incorporation of pollutants (e.g., SDZ, nonylphenol, and chlorophenol) into humic substances [34][35][36].

Advantages of the synthetic methods
The main advantage of our synthetic methods over those previously reported is the ability to synthesize [ 14 C]-SAs at micro-scale using commercially available, relatively inexpensive [ 14 C]-labeled 1 (~ 16.2 mg). Unlike the classic synthetic pathway, which proceeds via the condensation of 3 with aminoheterocycles, in our reaction the pathway that includes the condensation of 8 with chloroheterocylces has been optimized for the synthesis of [ 14 C]-labeled SAs using an aminoheterocycle of low nucleophilic activity or of high steric hindrance, either of which results in [ 13 C]-labeled SAs in good yields.
Purification is important for product quality. The methods for purification described herein are appropriate for [ 14 C]-compounds produced in small amounts as they result in a high purity. Both crystallization in water, as a purification procedure, and the direct use of the reaction mixture without further purification are applicable to the synthesis of unlabeled SAs at a gram scale [26,27], but not to the synthesis of [ 14 C]-labeled SAs at a milligram scale, because impurities may affect the next reaction in the absence of purification, but recrystallization may result in the recovery of smaller amounts of product. In this study, we used classic chromatographic separation methods, such as flash column chromatography and preparative TLC, to purify small amounts of [ 14 C]-products.

Conclusions
This study describes optimized methods for the synthesis of SAs labeled with 14 C or 13 C on the phenyl ring using commercially available [ 14 C]-or [ 13 C]-aniline, especially the synthesis of [ 14 C]-labeled SAs on a micro-scale (milligram amounts). Three typical sulfonamide antibiotics, SMX, SMM, and SDZ, with [ 14 C]-or [ 13 C]-labeling were prepared in good yields (5.0-22.5% for 14 C, 28.1-54.1% for 13 C relative to aniline). Both four-step (via the condensation of 3 and aminoheterocycles) and five-step (via the condensation of 8 and chloroheterocycles) reactions were examined. The four-step pathway is suitable for the synthesis of large amount of SAs (e.g., grams) or SAs containing aminoheterocyles of high nucleophilic activity, and the five-step pathway for the synthesis of SAs (e.g., SDZ) in milligram amounts and containing an aminoheterocycle of low nucleophilic activity. Both can be employed to prepare commercially unavailable labeled SAs for use in studies on
Additional file 1. The details of the instruments and analytical methods are provided. Table S1. Purification of [ 14 C]-labeled intermediates and SAs. Table S2. 1 H-NMR, 13 C-NMR, and HPLC−MS/MS analyses of synthesized unlabeled intermediates and SAs.