Synthesis of functionalized sulfonamides as antitubercular agents

Abstract Using up-to-date methods for synthesis and analysis, 51 sulfonamides were prepared for use as tools in antitubercular drug discovery. The synthetic efforts were centered on varying substituents at three key structural units implicated in antimicrobial activity, namely the sulfonyl group, nitrogen N1 and nitrogen N4. Procedures were specific to the sites of functionalization. Preliminary biological assessments are included here on selected compounds. The results suggest that the compounds may be useful in the exploration of the likely interactions of sulfa drugs with enzymes found in tuberculosis (dihydropteroate synthase) or its human host (N-acetyltransferase), interactions that result in drug activity or drug de-activation, respectively. GRAPHICAL ABSTRACT


Introduction
The development of sulfonamides (sulfa drugs, Figure 1) as antibacterial medications was once described as "the most profound therapeutic revolution in the history of medicine." [1] Eighty years ago, their unprecedented success in the treatment of lethal bacterial infections, including the scourges of staphylococcal septicemia and puerperal sepsis, led to remarkable patient recoveries and stimulated widespread fundamental research on this class of compounds. The research was not limited to probing antimicrobial properties, and discovery efforts in other areas were highly productive. For example, sulfonamide structural analogs were subsequently found to be effective as privileged scaffolds in such diverse applications as hypoglycemic agents, [2] diuretics, [3] and serotonin antagonists. [4] When sulfa compounds were first used in medicine as anti-infectives, there were no successful drug regimens against tuberculosis, the disease caused by the micro-organism Mycobacterium tuberculosis (MTB). Since the disease was known as being communicable, patients were isolated in specialized tuberculosis sanatoria and were prescribed rest, fresh air, and a lipid-rich diet to counteract the wasting syndrome associated with the illness. Prognosis was generally poor, and mortality was high. Given their value in ameliorating other bacterial infections, it was logical that sulfa compounds would be explored for the treatment of tuberculosis. Indeed sulfa drugs were investigated as antitubercular agents in monotherapy not long after their introduction, but this line of research was dropped as a result of poor efficacy. [5] Twenty-first century reports from the clinic, however, have suggested the potential usefulness of sulfa drugs against MTB when used in combination therapy [6][7][8][9][10][11][12] with other medications, and recent disclosures of effective antitubercular compounds have included sulfonamide structural moieties. [13][14][15][16][17][18] These reports have encouraged our own interest in preparing sulfa derivatives and probing the relationships between sulfa structure and antimycobacterial activity.
In general, the sulfonamides achieve activity by directly inhibiting dihydropteroate synthase (DHPS) on the folate pathway in bacteria. They act as mimics of the natural substrate p-aminobenzoic acid (pABA). The folate pool is depleted by this drug action, resulting in growth inhibition and ultimately in the death of the pathogen. Folate metabolism has been recognized as an important drug target in MTB, [19] and it has been suggested that the associated binding pocket in the DHPS structure could be exploited for the preparation of novel antitubercular agents. The crystal structure of drug-bound DHPS is now available and can be used to inform the process of drug discovery, an approach which has been previously demonstrated. [20] Ground-breaking research in the field has emphasized the importance of developing sulfa drugs that will only take up the volume in DHPS that would be assumed by its native substrate. Inhibitors that go beyond the volume assumed by the native substrate are less likely to be effective. [21,22] Conventional sulfonamide drug discovery has emphasized substitution at the sulfonyl group, which has historically been a particularly convenient site for chemical manipulation of druggable properties; however, the volume requirement in DHPS also necessitates exploration of substitution at N 1 and N 4 . Drugs with appropriate moieties at N 4 may lodge advantageously within the DHPS binding pocket. Substituents at N 4 could interact with residues within the binding pocket by a combination of effects, including hydrogen bonding, pi stacking, and other polar or non-polar attractions.
Crucially, sulfa drugs are susceptible to deactivation by arylamine N-acetyltransferases (NATs), enzymes endogenous in both host and MTB. [23,24] In human beings, NATs form part of the normal processes of xenobiotic transformation, whereby compounds foreign to the body's normal biochemistry, including drugs, are converted to products that may be more readily excreted. Hepatocytes are important foci for enzymatic deactivation of drugs by NATs, and compounds can be metabolized prior to reaching effective levels in systemic circulation. The metabolites are the principal forms of sulfa drugs in systemic circulation after NAT action. In mycobacteria, NATs may also form part of the pathogen's defense system against antitubercular drugs. [25] The metabolic products have far lower antibacterial activities than their un-acetylated counterparts, compromising their role in effective chemotherapy. [26] A potential strategy for the improvement of antimicrobial character emphasizes blocking the deactivating effects of NATs through changes in drug structure. [27][28][29] We wondered if design characteristics intended to thwart the deactivating effects of NATs might, at the same time, change the binding of sulfa drugs to DHPS and provide information about drug action. [30,31] In the work presented here, our goal was to provide reliable methods for the preparation of substituted sulfa compounds on gram scale for use as tools in antitubercular drug discovery. We also report preliminary assessments of representative materials to illustrate how they can be used as probes of the biological topics discussed earlier.

Results and discussion
All of the compounds prepared in this study are shown in Table 1. In beginning our work, we examined the chemical synthesis of sulfonamide NAT metabolites. These are important reference compounds for any work on the antitubercular activities of sulfonamides. Although known in the older literature, their chemical synthesis and description is often inadequate by present-day standards. In light of this, we developed a general protocol for their preparation (see the Experimental section) and fully characterized them. In a typical reaction, sulfa drug 1 [ Figure 1, R 1 ¼ C 6 H 5 CO, R 2 ¼ R 3 ¼ H, "sulfabenzamide" (2)] was smoothly transformed by acetic anhydride in acetic acid to its N 4 -acetyl derivative 3 (Table 1, ). In our biological assays, the activity of metabolite 3 in vitro against MTB was, as expected, poor ( Table 2, Entry 1). Since it represents the form of sulfabenzamide actually in circulation in the human system, this N 4acetyl metabolite serves as a realistic reference point for gauging the antitubercular activities of sulfa compounds. In a similar way, we prepared the N 4 -acetyl metabolites of several other sulfa drugs (Table 1, compounds 4-9).
We then formed an exploratory library (Table 1, compounds 10-51) of sulfa compounds and fully characterized them. Representative examples for introducing groups into the sulfa drug framework are shown in Figure 2. The methods allowed for substitution at the sulfonyl group ( Figure  2A), acylation at nitrogen N 4 ( Figure 2B) and disubstitution at nitrogens N 1 and N 4 ( Figure 2C). At the bench, conversions were readily monitored by spectrometric analysis, including the appearance, or loss, of distinctive 13 C-NMR signals near d 170 ppm for carbonyl groups and infrared bands near 1670 and 1160 cm À1 for amide and sulfonamide functions, respectively.
Introduction of substituents at the sulfonyl unit ( Figure  2A) was performed by treatment of the appropriate sulfonyl chloride with substituted amines, hydrazines, or hydrazides in pyridine to produce compounds 10-19. As a typical example of sulfonyl substitution, 4-(acetylamino)benzenesulfonyl chloride reacted in pyridine with methyl 4-aminosalicylate to yield the sulfonamide 10 (Table 1) (70%), with two characteristic peaks in the 13 C-NMR spectrum near d 170 ppm (C ¼ O) and infrared absorptions at 1675 (CONH) and 1174 cm À1 (SO 2 NH).
In addition to considering the effects of NAT acetylation, our design choices were informed by current information on the binding pocket of DHPS. [20] Substituents of different sizes and affinities at N 1 and N 4 could potentially modify the fit of the sulfa compound within the binding pocket. It   Figure 1).  seemed to us that non-polar interactions with phenylalanine 190 of the binding pocket, as previously noted, [19] represented an opportunity that could be probed with the variations at N 1 and N 4 as described earlier (Figure 3). Preliminary biological assessments were made using methods for which detailed accounts have previously appeared. [28,50,52] Given the costs of biological assays, some selection of compounds was necessary. Compounds were tested for efficacy from each of the representative types prepared earlier, using C logP as a selection guide, and as noted in the discussion later. In brief, we used whole cell assays with the virulent strain M. tuberculosis Erdman to measure minimum inhibitory concentration (MIC) values ( Table 2, see Experimental section). Representative MIC values of known antitubercular agents range from 200 mg/mL (ethambutol) to 0.06 mg/mL (isoniazid). [50,51] Acylation at N 4 with a lipophilic group led to improved activity and a more effective MTB MIC value compared to the N 4 -acetyl metabolite, as shown in Table 2. Thus, sulfabenzamide derivatives 43 (C logP 3.58), 44 (C logP 7.28), 50 (C logP 4.26), and 51 (C logP 4.47) ( Table 2, Entries 8-11) all had improved activities versus 3 (the N 4 -acetyl metabolite, C logP 1.99). Lipophilic modifications of sulfathiazole 23 and sulfacetamide 31 ( Table 2, Entries 5 and 6) also displayed useful activities in the range of 16-64 mg/mL. Our results thus appear to show that antimycobacterial activity may be improved by acylation at N 4 , provided that the acyl group is not acetyl. These activities are consistent with enhanced non-polar interactions with residues inside the DHPS binding pocket, as suggested earlier. They may also be due to escaping the deactivating action of NAT or to the better penetration of the waxy mycobacterial cell wall expected from enhanced lipophilicity. It is likely that the effect is multifactorial in its origin.
Interestingly, N 4 -valeroyl compounds 20, 21, and 22 tended toward decreasing activity with successive removal of methyl groups from the sulfamethazine core (Table 2, Entries 2-4). In the model of DHPS binding, [20] the heterocycle extends slightly from the pocket, into the solvent front, remote from the putative site of chemical action ( Figure 3). Subtle interactions at this interface may indeed affect the positioning of the rest of the molecule deeper within the pocket; however, the trend toward lower activity is also in step with reduced lipophilicity.
Compound 38 is representative of materials acylated at both N 4 and N 1 . Although its calculated lipophilicity is considerable (C logP 3.94), the compound is inactive (Table 2, Entry 7). Since lipophilic N 4 monoacylated compounds, such as 20, did show activity, this suggests that the second acylation at N 1 eliminates the antitubercular action. The calculated Connolly parameters of 38 are considerably larger than those for 20 (see Experimental section), consistent with the idea that inhibitors larger than the volume assumed by the native substrate in DHPS are less likely to be effective. This adds to the evidence that sulfa drugs fit into the DHPS binding pocket precisely. The introduction of a third substituent at N 1 may thus alter the shape and volume such that the drugs no longer mimic pABA.
In a recent study, [31] sulfabenzamide was verified for its ability to act synergistically with the antibiotic trimethoprim (TMP) to inhibit the growth and virulence of certain pathogenic bacteria. With an application of this information to MTB in mind, we prepared and characterized ten N 4 -substituted derivatives of sulfabenzamide (Table 1, [42][43][44][45][46][47][48][49][50][51]. Compounds 43,44,50, and 51 are representative and had moderate anti-MTB activities as shown in Table 2. We also examined 43 and 50 in Kirby-Bauer disk diffusion testing ( Table 3, see Experimental section), alone and in combination with TMP. TMP itself is inactive in this assay.
Although our combination studies indicated reduced growth of mycobacteria on the Kirby-Bauer test plates, further microbiological experiments will be needed to quantify this activity.

Conclusions
Sulfonamide derivatives with substituents at sulfonyl, N 1 and N 4 were formed in good yields and purities at gram scale using up-to-date methods for synthesis and analysis, with procedures specific to the sites of substitution. Such compounds serve in the exploration of factors influencing antitubercular properties. Preliminary biological assessments affirm that the N 4 -acetylated metabolites formed by NAT de-activation of sulfa compounds are not effective against MTB, as expected. However, acylation at N 4 with larger and more lipophilic structural units produces moderately active compounds. This may be due to a combination of escaping the de-activating effects of NAT, a better fit within the pABA binding pocket of DHPS, and better penetration of the drug through the lipid-rich mycobacterial cell wall. Further acylation at N 1 leads to N 1 ,N 4 di-acyl compounds, which have substantial lipophilicity values but are inactive. This suggests that the fit between the DHPS binding pocket and sulfa drugs may indeed be driven by volume requirements.

General
Elemental analyses were carried out by Robertson Microlit Laboratories, Ledgewood, New Jersey, USA. Melting points (mp, C) were taken in open capillary tubes using a MelTemp apparatus (Laboratory Devices, Cambridge, Massachusetts, USA), and are uncorrected unless otherwise noted. Due to recently-implemented institutional safety requirements on the use of mercury, some melting points were taken using alcohol thermometers and thus recorded only as high as 260 C. From the earliest days of research on their preparation, it has been well-known that sulfa compounds display polymorphisms and often form solvates of crystallization; these phenomena may influence the apparent physical properties, particularly melting points, of compounds prepared in different laboratories. [53] Reactants, reagents, and solvents were obtained from Alfa Aesar, Ward Hill, Massachusetts, and were used as received. Infrared (FTIR) spectra were recorded on a Perkin-Elmer Spectrum One Fourier transform spectrophotometer fitted with a universal attenuated total reflectance sampling accessory, reported in wavenumbers (, cm À1 ). Nuclear magnetic resonance (NMR) spectra were taken on a Bruker Avance 500 Fourier transform instrument as dilute solutions in dimethyl sulfoxide-d 6 (DMSO-d 6 ) or chloroform-d, recorded at 500 MHz ( 1 H NMR) or 125 MHz ( 13 C NMR) and are reported in parts per million delta (d) downfield from internal tetramethylsilane (TMS) as reference. Coupling constants for common multiplets were in the expected range of

Biological assessments
For some screening, Kirby-Bauer disc diffusion testing was used. [54] The compounds of interest (23, 43, 50, 20 mg) were dissolved in enough DMSO to prepare solutions that had a concentration of 20 mg/mL. The solutions were then applied to 6-mm filter paper discs such that the total weight of test compound was 200 lg. The discs were laid on 7H10 agar plates having a cell density of the test organism M. bovis BCG Tice of three McFarland units. M. bovis BCG Tice was obtained from the American Type Culture Collection (ATCC, Manassas, Virginia, USA). The plates were incubated at 37 C for 18-21 days and then read using transmitted light. The antimicrobial activity of the compound was measured by the dimensions of the circular clear zone surrounding the disc in which no growth occurred, while the remainder of the plate showed a luxuriant bacterial lawn. Trimethoprim (TMP) was inactive in this assay. For testing against the virulent strain M. tuberculosis Erdman, we determined minimum inhibitory concentrations in mg/mL by a method that has been thoroughly documented. [55] In brief, M. tuberculosis ATCC 35801 (strain Erdman) was obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). The standard of reference isonicotinic acid hydrazide (INH) was purchased from Sigma Chemical Company (St. Louis, MO, USA). For testing, a given compound was dissolved in dimethyl sulfoxide and subsequently diluted in distilled water. The reference standard was dissolved in distilled water. Stock solutions were filter-sterilized by passage through a 0.22-lm pore-size membrane filter and stored at -20 C until use. The drugs were prepared each morning, before experimentation. With respect to testing against this isolate, the MICs of all antimicrobial agents were determined in modified 7H10 broth (7H10 agar formulation with agar and malachite green omitted; pH 6.6) supplemented with 10% Middlebrook oleic acid-albumin-dextrose-catalase (OADC) enrichment (Difco Laboratories, Detroit, MI, USA) and 0.05% Tween 80. [56] The MICs of the antimicrobial agents were determined by a broth dilution method. [57] The organism was grown in the modified 7H10 broth with 10% OADC enrichment and 0.05% Tween 80 on a rotary shaker at 37 C for 5 days. The culture suspension was diluted in modified 7H10 broth to yield 100 Klett units/mL (Photoelectric Colorimeter, Manostat Corporation, New York, NY, USA), or approximately 5 Â 10 7 cfu/mL. The size of the inoculum was determined by titration and counting from triplicate 7H10 agar plates (BBL Microbiology Systems, Cockeysville, MD, USA) supplemented with 10% OADC enrichment. The plates were incubated at 37 C in ambient air for 4 weeks before counting of the colonies. The use of M. tuberculosis Erdman for determinations of MIC values is regarded as a rigorous challenge of antitubercular behavior. [58] N 4 -Acetyl sulfa drugs (NAT metabolites, compounds 3-9) In a procedure representative of the preparation of sulfa drug metabolites, glacial acetic acid (5 mL) was placed in a 100-mL round bottom flask fitted with a magnetic stirrer and heating mantle. The liquid was warmed to 85 C. Sulfabenzamide (1.38 g, 5.00 mmol) was added to the warm acid with rapid stirring to form a clear colorless solution (1.0 M). Slow dropwise addition of acetic anhydride (0.56 g, 5.50 mmol, 11 equiv.) in acetic acid (5 mL) over 20 min was followed by warming at 85 C and stirring for 1 h. Distilled water (40 mL) was added, and the mixture was cooled to room temperature to give an abundant white crystalline solid. Filtration and drying led to the title compound, 1.52 g (96%). The analytical sample was readily obtained by recrystallization from ethanol, m.p. 245 C, lit mp 245-248 C. [    Yield: 0.62 g (59%), mp 214 C, lit mp 211-212 C, [39] lit mp 215-216 C. [

4-[[[4-(Acetylamino)phenyl]sulfonyl]amino]-2-hydroxybenzoic acid methyl ester (10)
In a procedure representative of the introduction of substituents at the sulfonyl group, methyl 4-aminosalicylate (3.36 g, 20.12 mmol) was weighed into a 100-mL round bottom flask fitted with a heating mantle and magnetic stirrer. The addition of 25 mL of pyridine with stirring produced a clear brown solution. The portionwise addition of 4-acetamideobenzenesulfonyl chloride as the dry solid gave a mild exotherm. After stirring for several min, the reaction mixture was clear and wine red in color. The mixture was warmed to 75 C for 30 min and became brown. The temperature was gradually increased to 90 C over 1.5 h and held at this temperature for 1.75 h. The mixture was poured into crushed ice (100 g) and allowed to stand for 1.5 h, producing a voluminous solid. The solid was filtered off, allowed to dry overnight on the filter and recrystallized from ethanol. Yield: 5.12 g (70%), mp 240-242 C, lit mp 238 C. [ (11) Compound 10 (1.60 g, 4.39 mmol) was weighed into a 100-mL round bottom flask fitted for reflux and magnetic stirring. A mixture of ethanol and water (7.5:2.5, 5 mL) was added, followed by aqueous HCl (1 mL conc HCl:4 mL H 2 O, 5 mL). The mixture was brought to reflux with stirring and the reaction allowed to proceed for 2.25 h. While still hot, the mixture was filtered free of a small amount of solid. The filtered mixture was evaporated in a gentle stream of air, and the resulting solid was taken up in 45 mL of hot ethanol. The mixture was concentrated to 12 mL, then placed on an ice bath, producing a solid. The mixture was allowed to stand on ice for 1 h and the solid was filtered off and washed on the filter with ethanol (3 Â 5 mL). Yield: 1.12 g (79%), mp 201-202 C, lit mp 210 C. [

4-(Acetylamino)-2-(3-nitrophenyl)benzenesulfonic acid hydrazide (14)
3-Nitrophenylhydrazine hydrochloride (1.90 g, 10.0 mmol) was weighed into a 100-mL round-bottom flask and mixed with pyridine (10 mL). Triethylamine (1.01 g, 10.0 mmol) was mixed with pyridine (2 mL) and added to the reaction flask, then washed in with pyridine (3 mL). The resulting slurry was warmed to 80 C and more pyridine was added (5 mL). 4-Acetylbenzenesulfonyl chloride (2.34 g, 10 mmol) was added in several small portions, producing a mild exotherm, and more pyridine (5 mL) was added. The reaction mixture clarified and became wine red. Warming at 80 C was continued for 1.25 h. The mixture was poured into ice (100 g) and allowed to stand for 2 h. The resulting solid was filtered off and recrystallized from ethanol to give compound 14

4-(Acetylamino)-2-(2-methylphenyl)benzenesulfonic acid hydrazide (16)
2-Methylphenylhydrazine hydrochloride (1.59 g, 10.0 mmol) was weighed into a 100-mL round-bottom flask fitted for reflux and magnetic stirring. Chloroform (10 mL) was added and stirring begun. Separately, triethylamine (2.02 g, 20.0 mmol, 2.0 equiv) was weighed into a vial, and chloroform (5 mL) was added. The triethylamine solution was added to the slurry of 2-methylphenylhydrazine hydrochloride in several portions, and after several min a homogeneous solution was obtained. 4-Acetamidobenzenesulfonyl chloride (2.34 g, 10.0 mmol) was added as the dry solid in 30 small portions, giving a sufficient exotherm to cause the reaction mixture to boil. Another portion of chloroform (5 mL) was used to wash in the 4-acetamidobenzenesulfonyl chloride. Reflux was continued for 1.25 h. The mixture was cooled to room temperature. Ether (35 mL) and distilled water (10 mL) were added, producing an oil. The liquid portion was carefully decanted away from the oil. The oil was washed with distilled water (2 Â 10 mL), and the water decanted. Ethanol was added (105 mL) causing the oil to become a granular solid. The solid was recrystallized from the ethanol to give the 4-(acetylamino)-2-(2-methylphenyl)benzenesulfonic acid hydrazide. Yield: 2.31 g (72%), mp 143 C.  [59] Substitution at N 4 (compounds 20-33)

N 4 -Valeroylsulfamethazine (N-[4-[[(4,6-dimethyl-2-pyrimidinyl)amino]sulfonyl]phenyl]pentanamide) (20)
In a procedure representative of the preparation of the lipophilic N 4 -acylated sulfa compounds, valeric acid (10 mL) was placed in a 100-mL round bottom-flask fitted with a magnetic stirrer and heating mantle. The liquid was warmed to 80 C and rapid stirring was begun. To the warm stirred liquid was added sulfamethazine (1.39 g, 5.00 mmol). The sulfamethazine did not all dissolve, but it did form a tractable slurry. Valeric anhydride (1.02 g, 5.50 mmol, 1.1 equiv) was added dropwise at a rapid rate. Within less than 2 min after the completion of the addition of the anhydride, all of the white slurry had dissolved, and there was a homogeneous yellow solution. During the addition of the anhydride, the temperature had risen; it was now maintained at 105 C over the course of 1 h. Heating was stopped, and the reaction mixture was cooled to near room temperature using a cool water bath. The mixture was poured onto 50 g of chipped ice in a large beaker and allowed to stand overnight. The resulting white solid was filtered off, and a second crop was obtained from the mother liquor, total 1.41 g (78%  [44] The Connolly parameters were calculated as noted above in Section "General" and were: accessible area 566 Å 2 , molecular area 322 Å 2 , excluded volume 308 Å 3 .

N 4 -Valeroylsulfadiazine (N-[4-[(2-pyrimidinylamino)sulfonyl]-phenyl]pentanamide) (22)
The compound was prepared using a procedure similar to the one in section "N4-Valeroylsulfamethazine (N- [4-[[(4,6-dimethyl-2-pyrimidinyl)amino]sulfonyl]phenyl]pentanamide) (20)." Yield: 3.17 g (95%, 10.0 mmol scale), mp 218-219 C, lit mp 222-223 C. [44]   . [44] FTIR    Pyridine (10 mL) was placed in a 100-mL round-bottom flask fitted with a magnetic stirrer and voltage-regulated heating mantle. The liquid was warmed to 70 C and rapid stirring was begun. To the warm stirred liquid was added sulfamethazine (1.39 g, 5.00 mmol) in several portions, creating a homogeneous yellow solution. The warm solution was stirred vigorously as benzoyl chloride (0.70 g, 5.00 mmol) was added dropwise in a rapid manner. There was a distinct exotherm. The mixture was stirred for 30 min, at the end of which time the temperature was 70 C. The mixture was cooled to near ambient temperature in a cool water bath, then poured onto 75 g of chipped ice. This produced a white precipitate. The mixture was allowed to stand several hours. The solid was filtered off and allowed to dry for several days. The solid was recrystallized from ethanol (105 mL), dried on the filter and then dried in vacuo. Yield: 1.77 g (93%), mp 237 C, lit mp 233-235 C. [54] The 1 H-NMR spectrum was identical to the spectrum obtained online on Chemical Abstracts Service SciFinder under the entry for the title compound (CAS Registry Number 102017-64-1  Sulfamethazine (1.39 g, 5.00 mmol) was weighed into a 50-mL pear-shaped flask fitted with a heating mantle and mixed with pyridine (4 mL). The mixture was brought to 80 C to produce a clear slightly yellow solution. To this warm mixture was added 4-phenylbenzoyl chloride (1.09 g, 5.00 mmol) in several portions with swirling after each addition. Finally, the last of the acid chloride was washed in with pyridine (2 mL). The mixture was maintained at 80 C for 45 min. Heating was stopped, and the mixture was poured onto 150 mL of chipped ice. The material was allowed to stand overnight to produce a voluminous white crystalline solid. The solid was filtered off and washed with ether (10 mL, then 3 Â 5 mL) and allowed to dry to a white free-flowing solid. The analytical sample was easily recrystallized from ethanol (45 mL   Pyridine (20 mL) was placed in a 100-mL round-bottom flask fitted with a magnetic stirrer and voltage-regulated heating mantle. The liquid was warmed to 70 C and rapid stirring was begun. To the warm stirred liquid was added sulfathiazole (2.55 g, 10.0 mmol) in several portions, creating a homogeneous yellow solution. The warm solution was stirred vigorously as benzoyl chloride (1.40 g, 10.0 mmol) was added dropwise in a rapid manner. There was a distinct exotherm. The mixture was stirred and warmed for 30 min, at the end of which time the temperature was 70 C. The mixture was cooled to near ambient temperature in a cool water bath, then poured onto 150 g of chipped ice. This produced a white precipitate. The mixture was allowed to stand overnight. The solid was filtered off and gave 3.72 g of a slightly damp mass. The solid was then washed with 150 mL of boiling ethanol for 10 min, then filtered and dried. Yield: 3.28 g (91%), mp >260 C.   (29) The preparation of this compound was done from the reaction of sulfathiazole with lauroyl chloride and was carried out in a manner similar to that in section "N4-Benzoylsulfathiazole (N- [4-[(2-thiazolylamino)sulfonyl]phenyl]benzamide) (28)." Yield: 3.18 g (73%, 10.0 mmol scale), mp 167-168 C, lit mp 165-167 C, [44] lit mp 166 C. [ 47, 36.92, 31.76, 29.49, 29.46, 29.39, 29.26, 29.18, 29.09, 25.43, 22.56, 14.38. (30) This preparation was carried out in a manner similar to that in section "N 4 -Benzoylsulfathiazole (N- [4-[(2-thiazolylamino) 31, 133.11, 129.29, 118.83, 36.94, 31.77, 29.49, 29.47, 29.40, 29.26, 29.19, 29.  Sulfamethazine (1.00 g, 3.6 mmoles) was weighed into a 50-mL pear-shaped flask fitted with a heating mantle. Pyridine (4 mL) was added and the mixture was warmed until a homogeneous slightly yellow solution was obtained. Acetic anhydride (1.84 g, 18.0 mmol, 5.00 equiv) was added to the warm solution in several portions and washed in with further pyridine (1 mL). The mixture was brought to the boil. Within several minutes, the mixture became milky in appearance and a solid was suspended within the liquid. After 10 min, a white solid had deposited. Refluxing was continued for 1 h. Heating was stopped and the mixture allowed to cool and stand overnight. The solvent was evaporated. The resulting beige mass was washed with portions of ether (20 mL, then 5 mL) and dried. The analytical sample was obtained by recrystallization from ethanol. The following characterization data made it clear that N1,N 4 -diacetylation had indeed occurred and that the compound was not merely N 1 -monoacetylated (compare data for compound 5). Yield: 1.08 g (83%), mp 239-240 C, lit mp 237-238 C (dimethylacetamide, water). [  [N 4 -valeroylsulfamethazine (0.58 g, 1.60 mmol)] was weighed into a 50-mL pear-shaped flask fitted with a heating mantle. Pyridine (4 mL) was added and the mixture was warmed until a homogeneous slightly yellow solution was obtained. Acetic anhydride (1.84 g, 18.0 mmol, 11.25 equiv) was added to the warm solution in several portions and washed in with further pyridine (1 mL). The mixture was brought to the boil. Refluxing was continued for 1 h. Heating was stopped and the mixture allowed to cool and stand overnight. The solvent was evaporated. The resulting beige mass was washed with portions of ether (3 Â 10 mL) and dried.
Yield  (N 4 -valeroylsulfamethazine, 0.58 g, 1.60 mmol) was weighed into a 50-mL pear-shaped flask fitted with a heating mantle. Pyridine (4 mL) was added and the mixture was warmed until a homogeneous slightly yellow solution was obtained. Propionic anhydride (1.04 g, 8.00 mmol, 5.00 equiv) was added to the warm solution in several portions in a rapid dropwise manner and washed in with further pyridine (1 mL). The mixture was brought to the boil. Refluxing was continued for 1 h. Heating was stopped and the mixture allowed to cool and stand overnight. The solvent was evaporated. The resulting tan mass was washed with portions of ether (3 Â 10 mL) and dried.        ). Each of the three high-field multiplets appeared to be the very close overlap of two equal-sized signals, possibly indicating hindered rotation. 13    Toluene (10 mL) was added, and the mixture brought to the boil. This gave a tractable slurry. Butyric anhydride (0.87 g, 5.5 mmol, 1.1 equiv) was added by dropper pipet to the rapidly stirred slurry in one continuous portion. Refluxing was continued for 1 h. Heating was ended, and the mixture was allowed to cool slightly. The resulting white solid was filtered off and allowed to dry. The white solid was readily recrystallized from ethanol (85 mL  Sulfabenzamide (1.38 g, 5.0 mmol) was weighed into a 25-mL round-bottom flask fitted with a heating mantle and magnetic stirrer. Valeric acid (10 mL) was added, and the mixture was warmed to 140 C, giving a homogeneous solution. Valeric anhydride (1.02 g, 5.5 mmol, 1.1 equiv) was mixed with valeric acid (5 mL) and added dropwise over 15 min to the hot solution of the sulfabenzamide. After the addition was complete, the heating was continued for 20 min, then stopped. A voluminous solid was now present. The mixture was allowed to cool to room temperature, then filtered off and washed on the filter with ethanol (3 Â 5 mL), then allowed to air dry. The solid was readily recrystallized from ethanol (35 mL In a 100-mL round-bottom flask fitted with a magnetic stirrer and heating mantle, pyridine (15 mL) was warmed to 70 C. Sulfabenzamide (2.76 g, 10.0 mmole) was added in several portions and washed in with pyridine (5 mL). With rapid stirring, lauroyl chloride (2.19 g, 10.0 mmol, 1.0 equiv) was added rapidly in three portions using a dropper pipet. Stirring and warming were continued for 2 h. The blond reaction mixture was poured into 100 g of ice and allowed to stand overnight. The resulting white solid was filtered off and washed on the filter with ethanol (3 Â 5 mL), then allowed to dry. The solid was readily recrystallized from ethanol (70 mL). Yield: 3.60 g (79%), mp 208 C. FTIR: max