290 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS lower toxicities, it has been demonstrated that the presence of a sugar group decreases the toxicity of some antibacterial agents (17,18) and some anticancer agents (19,20). N-glucosylthioureas and N-glucosylrhodanine derivatives have therefore been synthe- sized, and their antimicrobial activities were determined by the agar plate streak method and by the broth dilution method for determining minimal inhibitory concentrations (MIC). SYNTHESIS As early as 1914, Fischer (21) synthesized tetra-0-acetyl-glucopyranosylthiourea from tetra-0-acetylglucopyranosylisothiocyanate and ammonia, but he was unable to get the deacetylated product. Other investigators (22,23) have prepared glycosylthioureas by using acylated glycosylisothiocyanates and amines, but have generally been unsuccess- ful in hydrolyzing the acyl groups. Foye and Tovivich (24) synthesized N-gluco- sylthiourea from the condensation product of o•-acetobromo-D-glucose and 5- aralkylidenerhodanines followed by a successful hydrolysis using a methanol-ammonia system to remove acetyls and cleave the rhodanine ring. The N-glucosylthiourea derivatives were synthesized by hydrolyzing N-(2,3,4,6-tetra- O-acetylglucopyranosyl)o5-benzylidenerhodanine in methanolic amine solutions. N-/5-D-Glucopyranosyl-5-aralkylidenerhodanine derivatives were prepared by hydro- lyzing their tetraacetyl derivatives with methanolic hydrogen chloride without cleavage of the rhodanine ring structure. Representative syntheses for the N-glucosylthioureas and N-glucosylrhodanines are described in the Experimental section. ANTIMICROBIAL EVALUATION The term "antimicrobial agent" (25) may be used to designate any substance of natural, semi-synthetic, or synthetic origin that inhibits or kills free living, commensal, or pathogenic microorganisms while causing little or no injury to the host. There are several currently used methods for testing antimicrobial susceptibility (26). The agar diffusion test and antimicrobial dilution test have been the most widely employed. For the quantitative estimate of antimicrobial activity, the minimal inhibitory concen- tration (MIC) is determined as the lowest concentration that will inhibit growth of the microorganism. In our experiments, the 10-fold dilution method was used instead of the doubling dilution method, and the MIC was expressed in/xg/ml. In our antimicrobial test, five different types of organisms were selected. The inclusion of Aspergillus niger and Pseudomonas aeruginosa in the inocula is recommended (27) to ensure a wide range of protection, because these ubiquitous organisms grow in a large variety of formulations and are relatively resistant to antimicrobial agents. Staphylococ- cus aureus and Escherichia coli were also included as representative of Gram-positive and Gram-negative organisms, and Candida albicans was included to provide a yeast. The incidence of contaminated cosmetics has been the subject of several investiga- tions (28,29) Pseudomonas was the predominant organism found. Microbial spoilage of cosmetic preparations was a major concern of manufacturers in former years (30) as well. Spoilage in this context was defined as microbial growth that results in various deleterious effects such as noxious odors and gases, changes in pH, viscosity, and

ANTIMICROBIAL ACTIVITY OF N-GLUCOSYLRHODANINES 291 color, and destruction of emulsions. The presence of a visible means of growth or bacterial slime on the surface of a product obviously renders it unsuitable for marketing. EXPERIMENTAL 2,3,4,6-TETRA-O-ACETYL-fl-D-GLUCOPYRANOSYL BROMIDE This was prepared by the method of Lemieux (31) in 89% yield mp 88-89 ø (lit. mp 88-89ø)(31). 5-BENZYLIDENERHODANINE Rhodanine (13.32 g, 0.1 mole) was dissolved in 100 ml of glacial acetic acid, and to this was added 24.6 g (0.3 mole) of fused sodium acetate. After the solution became clear, 10.61 g (0.1 mole) of benzaldehyde was added and the reaction mixture was refluxed for 45 min. The whole mixture was poured into 500 ml of water, and the crude product was separated by filtration to give 19.88 g (90% yield) of yellowish brown crystals with mp 204-208 ø. Recrystallization from toluene improved the mp to 208-209 ø (lit. mp 204-205ø)(32). N-(2,3,4,6-TETRA-O-ACETYL-•-D-GLUCOPYRANOSYL)-5-BENZYLIDENERHODANINE To a solution of 22.13 g (0.1 mole) of 5-benzylidenerhodanine in 700 ml of acetone, 40 ml of 10% sodium hydroxide solution was added, followed by 44.12 g (0.1 mole) of acetobromoglucose. The mixture was stirred at room temperature for 22 hr. The reaction mixture was evaporated to dryness in vacuo. The crude product was recrystallized from hot methanol, yielding 42.3 g (82%) mp 192-193 ø, [Ot]D 25 = --168 ø (C = 3.1 pyridine)(lit. mp 193-194 ø) (33). N-fi-D-GLUCOPYRANOSYLTHIOUREAS N-(2,3,4,6-Tetra-0-acetyl-/5-D-glucopyranosyl)-5-benzylidenerhodanine (III)(5.52 g, 0.01 mole) was added to a solution of diethylamine (5.85 g, 0.08 mole) in 70 ml of methanol in a pressure bottle. The reaction mixture was stirred at room temperature for 20 hr with the stopper well closed. The reaction mixture was stirred for 1 hr with 70 ml of DOWEX-50WX8 (cation exchange resin) to remove the excess amine and ot-mercaptocinnamyl amide, and filtered. The flitrate was dehydrated with anhydrous sodium sulfate and evaporated to a syrupy material. The residue was crystallized with acetone-methanol (2:1) mixture, yielding 2.41 g (81.5%) mp 106-109 ø dec., [OI]D 25 = - 11.0 ø (c = 1.1). IR(KBr): 3520 (OH), 3100 (NH), 1080 (C = S), 890 (/5-form)cm -t. NMR (dimethylsul- foxide-d6): b 5.40 (d, 1H, NH), 3.21 (m, 4H, N(CH2)2), 1.17 (t, 6H, 2CH,) ppm. Analysis-Calculated for CnH22N2OsS. H20: C, 42.29 H, 7.74 N, 8.96 S, 10.26. Found: C, 42.57 H, 7.66 N, 8.87 S, 10.34. Other glucosylthioureas were prepared by the same method.