外文原文-1β-Lactams.3.Asymmetric Total Synthesis of New Non-Natural 1β-Methylcarbapenems Exhibiting Strong Antimicrobial Activities and Stability against Human Renal Dehydropeptidase-IYoshimitsu Nagao,*1aYunosuke Nagase,1bToshio Kumagai,1bHiroshi Matsunaga,1bTakao Abe,1bOsamu Shimada,1bTakaaki Hayashi,1band Yoshinori Inoue1bFaculty of Pharmaceutical Sciences,The University of Tokushima,Shomachi,Tokushima 770,Japan,and TheChemical and Formulation Laboratories,Lederle(Japan)Ltd.,Kashiwacho,Shiki,Saitama 353,Japan Asymmetric synthesis of 11,the precursor to chiral(3R,4R)-3-[(1R)-1-[(tert-butyldimethy-leilyl)oxy]ethyl]-4-acetoxyazetidin-2-one(3)was achieved by utilizinga highlydiastereoselectivealdoltypereaction of acetaldehydeand the chiral tin(Ⅱ)enolate of 5.Similar diastereoselectivealkylations of chiral and achiral tin(Ⅱ)enolates13a-d with chiral3 were also performed to obtain the desiredalkylated azetidin-2-ones(17a-d).Compounds17a,bwere successfullyconvertedto new,non-natural lβ-methylcarbapenemla and lb,which exhibited strongand wide-rangingantimicrobial activities and excellent stabilityagainst human renal dehydropeptidase-I. In 1980,a Merck Sharp&Dohme research group pub-lished a stereoselectivetotal synthesisof(+)-thienamycin,a fascinating natural&lactam antibiotic.2 This synthesis established an excellent synthetic methodology for carbapenems.Since then,there have been numerous reports related to the synthesis of thienamycin and modified thienamycins.3,4 1β-Methylcarbapenems in particular attracted attention in the developmentof new,non-natural carbapenemsbecause they possess strong stability against renal dehydropeptidase-I maintaining the superior antibacterial activity of(+)-thienamycin.5The Nagao and Lederle(Japan)groups,6 the Fuentes group,’and other groups4,8have each reported a highly diastereoselective alkylation method useful for 16-methylcarbapenemsynthesis. In the preceding papers,we reported on highly diastereoselectivealkylations of 4acetoxyazetidin-2-0ne3a,6 and racemic(3R*,m*)-3-[(1R*)-l-[(tert-butyldimethylsily1)oxy]ethyl]-4-acetoxyazetidin-2-0ne3b,6 and the utilization of the reaction products for the preparation of chiral key intermediates for the synthesis of cabapenems.3,6,9 Continuingour series of studies on β-lactamsyntheses,we now describe in detail a practical method useful for carbapenem synthesish and ita applicationto the asymmetric total synthesisof new,non-natural 10-methylcarbapenems la and lb.These particular 1S-methylcarbapenems,bearing a heterocyclic quartenary ammonium as the RSgroup at C-2,are expected to exhibit excellent antimicrobial activity.Especially,the bicyclic triazoliummoiety of la can be regarded as a prochiral a-symmetric heterocycle by delocalization of the π-electron system.A synthetic strategyfor la and lb using C-6chiral thiazolidines.Was designed as shown in eq 1.In the synthesis of 1β-methylcarbapenems,the constructionof four consecutive asymmetric carbon atoms(Le.,C1,C5,C6,and C8)is intriguing.We adopted an asymmetric,aldol-type reactionlo of acetaldehyde with a chiral tin(I1)enolate for C6-C8 bond formation,which leads to optically active (3R,4R)-3-[(1R)-1-[(tert-butyldimethylsilyl)oxy]ethyl]-4-acetoxyazetidin-2-one(3).An efficient,diastereoselectiveimine alkylation6,9-11between another chiral tin(Ⅱ)enolate and the chiral cyclic acylimineobtained in situ from 3 was adopted for C1-C5 bond formation.Utilization of our C-4-chiral thiazolidine reagents for construction of all fourasymmetric centers in la,b is the remarkable feature in this 10-methylcarbapenemsynthesis. 3-[3-(Benzyloxy)propionyl]-(4R)-isopropyl-l,3-thiazolidine-2-thione(5),obtained by the reactio-n of 3-(benzyl-0xy)propionic acid(4)and(4R)-isopropyl-l,3-thiazolidine-Zthione,was treated with a suspension of tin(Ⅱ)trifluoromethanesulfonate12and N-ethylpiperidine12in CH2C12at-78℃ for 2 h.Excess acetaldehyde was added, and the mixture was stirred at-78"C for 1h to afford alcohol 7a in 84%yield and in 94%diastereomericexcess by HPLC analysis.The stereochemistry of 7a can be rationalized in terms of transition state 6,in which acetaldehyde approaches the chiral tin(Ⅱ)enolate from the less-hinderedβ-side,opposite the a-isopropyl group of the thiazolidine moiety,to form a chairlike six-membered ring.1°In the chairlike six-membered ring,the methylgroup of acetaldehyde is equatorial.After protection of the hydroxy group of 7a with the TBDMS group,compound 7b was submitted to aminolysis with p-anisidine to give amide 8a in 91%yield from 7a.13 Hydrogenolysis of the benzyloxy group of 8a followed by mesylation gave compound 9 in high yield.Cyclization of amide 9 with NaH in 4 :1 CH2C12-DMF’proceeded well to giveβ-lactam 10 quantitative1y.14Oxidative deprotection of the pmethoxyphenylgroup of 10 with ceric ammoniumnitrate15 afforded known compound 1116in 67%yield(Scheme I).Efficient conversion of 11 to chiral(3R,4R)-3-[(1R)-l-[(tert-butyldimethylsilyl)oxy]ethyl]-4acetoxyazetidin-2-one(3)with RuC13.nH20and peracetic acid has alreadybeen achieved by the Murahashi group.17 We have demonstrated that 3-acetyl-(4S)-ethyl-1,3-thiezolidine2-thioneis a matched partner in the alkylation reactions of chiral(3R,4R)-3-[(1R)-l-([tert-butyldimethy~ilyl)oxy]ethyl]-4acetoxyctidin-2-one(3).3bThus,the chiral tin(I1)enolates generated in situ by enolization of 3-propionyl-(4s)-ethyl(and isopropyl)-1,3-thiazolidine-2-thiones(12a,b)with tin(Ⅱ)trifluoromeThan esulfonate and N-ethylpiperidine were treated with chiral 3 in THF at 0℃ for 1h.The reaction of 12a and 3 followed by the usual workup afforded C-4-alkylated azetidin-2-ones 17a and 18a in a 90:10 ratio by HPLC analysis(Scheme11and Table I).Alkylation of 3 with 12b gave a similar mixture of 17b and 18b(91:9).The alkylation of chiral 3 with 3-propionyl derivatives 17c,d of achiral l,&thiazolidine-2-thiones has also been carried out.la Although the alkylation proceeded smoothly,the diastereoselectivities were poorer than those of the alkylations with C-4chiral thiazolidines(see Table I). The absolute configurations of the major products(17a,b)were determined by their chemical conversion toknown compound 22,5a a key intermediate for 1β-methylcarbapenem synthesis(see Scheme Ⅲ).Compounds 17a and 17b,which both have an active amide structure,were treated with imidazole in MeCN at room temperature to form imidazolide 19.9~~C9ompound 19 wassubmitted in situ to the decarboxylative Claisen-type condensation2to affordβ-keto PNB ester 21 in 80%yield from 17a and in 86%yield from 17b.Elimination of the TBDMS group of 21 was readily done under acidic conditions to give compound 22.5aThe Stereochemistryof the other major products(17c,d)was confirmed by comparison of the HPLC data withthat of the compound derived from the dehydrative condensation reaction between lb-thia-mlidine-2-thione(or 4,4dimethyl-l,3-thiezolidine2-thione)and carboxylic acid 20.(Acid 20 was obtained by acidic hydrolysis of 19.13)The absolute configuration of a minor C-4-alkylated product(Ma)was determinedby its chem-ical correlation with known compound 255aas depicted in Scheme IV.Compound 26 was prepared from 233a in the following manner.Methyl ester 24,obtained by methanolysis of 23,was treated with 2 mol equiv of LDA to form the enolate.The enolate was treated with Me1 to give methylated product 25.5a Alkaline hydrolysis of 25and subsequent dehydrative condensation of the resultantcarboxylicacid 26 with(4S)-ethyl-l,3-thiazolidine2-thione [(4S)-ETT]gave Ma.The absolute configuration of 18c was determined by its conversion to 26 under alkaline methanolysis conditions(Scheme IV).The stereochemistry of the other minor alkylation products(18b,d)was confirmed by the fact that their HPLC retention times were identical to those of the compounds prepared by dehydrative condensation of carboxylic acid 26 and the corresponding 1,3-thiazolidine-2-thiones. Because epimerization of the newly formed 0-methyl group of the major products 1 7 8 4 has never been observed under the alkylation conditions described above, the stereochemistry of major products 17a-d and minor products 18a-d can be explained as follows.Major products 17a-d could be obtained from the corresponding Z-tin(Ⅱ)enolates 13a-d18 uia six-membered,chelated transition states 16a-d.To form transition states l5a,b, the chiral acylimine derived from 3 must approach the&al Z-tin(II)enolates 13a,b from the less hinderedβ-side (α-R1=Et,i-Pr).To form transition states 15c,d,the achiral 2-tin(II)enolates 13C,dmmust approach the chiral acylimine from the less hindered a-side,opposite theβ-(sily1oxy)ethylgroup at C-3 of the&lactam.In a similar process,minor producta(18a-d)could be formed uia transition states 16a-d involving E-tin(II)enolates 14a-d19,20and the acylimine obtained in situ from 3.Further evidence for these mechanistic speculationswas obtained when the substituentsof the thiazolidine-2-thionegroup were changed.The bulkiness of the R1and/or R2group(s) of the thiazolidine-2-thionemoieties affects the product ratios of major compounds 17a-d and the minor compounds 18a-d(see Table I).Thus,kinetic enolization giving Z-enolates 13a-d and formation of rigid transition states such as 15a-d seem to be essential to obtain the desired stereochemicaloutcome for alkylation of the cyclicacylimines.In our cases,a transition state leading to kinetic 2-enohtion should be more stablethanthat leading to kinetic E-enolization because the latter bears severe steric repulsion between the methyl group and the R1and/or R2group(s)(see Figure 1).外文原文-2:New Straightforward Synthesis and Characterization of a Unique1β-Methylcar-bapenem Antibiotic Biapenem Bearing a ó-Symmetric BicyclotriazoliumthioGroup as the Pendant MoietyToshio Kumagai,*,1a Satoshi Tamai,1a Takao Abe,1a Hiroshi Matsunaga,1aKazuhiko Hayashi,1a Ikuo Kishi,1a Motoo Shiro,1b and Yoshimitsu Nagao*,1c-The Chemical and Formulation Research Laboratories, Lederle (Japan), Ltd., Kashiwa-cho, Shiki,Saitama 353-8511, Japan, Rigaku Corporation, 3-9-12 Matsubara-cho, Akishima-shi,Tokyo 196-8666, Japan, and Faculty of Pharmaceutical Sciences, The University of Tokushima,Sho-machi, Tokushima 770-8505, Japan Biapenem1(1R,5S,6S,)-2-[(6,7-dihydro-5H-pyrazolo[1,2-a][1,2,4]triazolium-6-yl)thio]-6-[(R)-1-hydroxyethyl]-1-methylcarbapen-2-em-3-carboxylate, is a new non-natural 1â-methyLcarbapen-emantibiotic which exhibits a wide range of antibacterial activity, remarkable chemical stability, andextensive stability against human renal dehydropeptidase-I. Mercaptobicyclotriazolium chloride2 useful for the pendant moiety of 1 was successfully synthesized starting from hydrazine hydrate3 along an economically available synthetic route. The thiol 2 was efficiently exploited for anexpeditious synthesis of biapenem 1. Characterization (crystal structure, nonbonded S- - -Ointeraction, conformational analysis, and CH- - -O hydrogen bonds) of 1 was investigated by its X-ray crystallographic, 1H NMR, and deuteration experiment analysesIntroduction Since the development of a remarkable non-natural 1â-methylcarbapenem antibiotic by a Merck Sharp & Dohme research group,2 there have been several reports on new characteristic 1â-methyl-carbapenems, such as Meropenem,3a biapenem,3b lenapenem,3c and others.3b,d,4a,bThese 1â-methyl-carbapenems are promising as newgeneration non-natural â-lactam antibiotics because oftheir exc-ellent biological and chemical behavior.3,4 We have developed a unique biapenem 13b bearing a ó-symmetric (6,7-dihydro-5H-pyrazolo[1,2-a][1,2,4]triazolium-6-yl)thio (“bicyclotriazolium”thio) group at C2 as the pendant moiety that is strikingly different from that of other 1â-methylcarbap-enems.3a,c It is generally knownthat the C2-pendant moiety of the carbapenem antibiotics plays an important role in their antibacterial activity, chemical stability, and stability against renal dehydro-peptidase-I (DHP-I).5 Biapenem 1 exhibits a wide range of strong antibacterial activity,remarkable chemicalstability, and extensive stability against human renalDHP-I.4c,d In the earlier synthesis of 1, we adopted aroundabout synthetic route by which the desired bicyclotriazoliummoiety was con-structed after introductionof 4-mercaptopyrazolidine bis-N-p-nitrobenzyloxycarbonylderivative6 onto the 1â-methylcarbapenem skeleton14.3b We now describe an alternative straightforward synthesis of biapenem 17 and its characterization basedon X-ray crystallographic and 1H NMR analyses and a deuteration experiment.Results and DiscussionIn designing the pendant molecule, a new heterocyclic compound, mercaptobicyclotriazolium chloride 2, was chosen on the basis of the following consideration.Namely, we wished to establish a reaction cascade (vide infla) for construction of this particular heterocycle by treatment of a pyrazolidine derivative with ethyl formimidate hydrochloride. Thus, we attempted to define an economically viable synthesis of 2 starting from hydra-zine hydrate 3 and an expeditious synthesis of biapenem 1, as shown in Schemes 1 and 2. Treatment of 3 with HCO2Et and then acetone gave crystalline compound 4 in 96% yield. Allylation of 4 with allyl bromide employing K2CO3 in AcOEt at 80 °C followed by deprotection of ketimine 5 with excess HCO2H at 80 °C afforded bisformylated allylhydrazine 6 in 68% yield from 4. Bromination of 6 with Br2 in the presence of catalytic LiBr in CH2Cl2-MeOH at 0 °C followed by cyclization of the resultant dibromide 7 with K2CO3 in warm AcOEt furnished cyclic diformyl product 8 as colorless prisms in 70% yield from 6. Substitution reaction of 8 with potassium thioacetate in warm AcOEt gave crystalline acetylthiolate 9 (95% yield) which was submitted to methanolysis in MeOH containing KOH followed by acidification with a small excess of HCO2H to obtain thiol 10 in 98% yield. Air oxidation of 10 with catalytic FeCl3 in MeOH followed by deformylation of the resultant disulfide 11 with concd HCl in MeOH gave pyrazolidine disulfide dihydrochloride 12 as colorless prisms in 81% yield from 10. Then the pyrazolidine derivative 12 was subjected to the reaction cascade toward the desiredbicyclotriazolium derivative 13, which has been previously reported by our group.3b,7 Namely, the compound 12 was allowed to react with ethyl formimidate hydrochloride3b, 7 in the presence of KHCO3 in water at 0 °C, and the crude product was purified on a Dowex 50W-X4 column with 50% MeOH and 6 N HCl-MeOH (1:1) to furnish a crystalline compound bicyclotriazolium disulfide dichloride 13 in 75% yield. An aqueous solution of 13 was treated with a solution of Bu3P in THF at 0 °C, and the reaction mixture was repeatedly washed with AcOEt. After evaporation of the resultant water layerin vacuo, the crude product was crystallized in isopropyl alcohol to give mercaptobicyclotriazolium chloride 2 as the hydrate compound (C5H8N3SClâH2O) in 70% yield. Subsequently, the mercaptobicyclotriazolium moiety was introduced to the 1â-methylcar-bapenem skeleton in a Michael type reaction manner, as shown in Scheme 2. The compound 14, obtained by our earlier synthetic method,3b was treated with 2 in the presence of diisopropyle-thylamine in a solution of MeCN-acetone-DMF (1:1:0.1) at 0 °C to afford thioether 15 as color-less needlesin 91% yield. Finally, deprotection of the p-nitrobenzyl group (PNB) of 15 was effic-iently performed by our method8 employing Zn dust as follows. Treatment of 15 with excess Zn dust in 0.35 M phosphate buffer solution (pH 5.6) at room temperature followed by the usual workup3b,8 of the reaction mixture gave the desiredbiapenem3b in 80% yield. Successful recrystallization of 1 from aqueous EtOH furnished excellent crystals [colorless needles; mp 210- 218 °C (decomp); [R]20 D -33.7° (c 0.5, water)] that werethen submitted to X-ray analysis. The crystallographicstructure of 1 is depicted in Figure 1. The conformation of C6-hydroxyethyl group of 1 is different from that of imipenem9a and meropenem.9b An intramolecular nonbonded O12- - -S16 interaction (“close contact”) is recognized in the crystalline structure of 1.10 This O12- - -S16 close contact must be common even in other â-lactam antibiotics.9a,b,11,17 To learn the conformation of C2 and C6 side chain groups of 1 in D2O, a nuclear Overhauser effect (NOE) experiment [1H NMR analysis (400 MHz)] was performed. Selected NOE enhancements are illustrated in Figure 2. Interestingly, an NOE enhancement between H1 and H8 was confirmed on the spectrum of meropenem by Sumitomo Pharm. Co. research group9c but was not recognized on that of biapenem 1. On the other hand, an NOE enhancement between H6 and H8 was observed on the spectrum of 1 but was not on that of meropenem.9cThus, the predominant conformation of C6-hydroxyethyl and C2-bicyclotriazoliumthio groups in the molecule 1 in D2O may be similar to that in the crystalline structure (Figure 1). Particularly, the conformational stability of the bicyclotriazoliumthio group at C2 may be caused by the intramolecular nonbonded S- - -O interaction due to n (O12) ó* (S16-C17) orbital overlap.10 Interestingly, four kinds of possible intermolecular hydrogen bonds (O13- - -H-O15, O12- - -H-C20, O14- - -H-C22, and O13- - -H-C22) were suggested in the crystal structure of 1. The interatomic distances (A- - -H: ca. 2.03-2.28 Å) between the specific acceptor atom (A: O12, O13, or O14) and hydrogen atom (H) of the specific donor (D: O15, C20, or C22) are unusually shorter than the sum (2.72 Å) of van der Waals radii between oxygen and hydrogen atoms, as shown in Figure3. The angles (<A- - -H-D: 123.6-149.5°) are also sufficiently large for the hydrogen bonding describedabove. In order to get an insight into the acidity of both the C20 and C22 hydrogen atoms of the bicyclotriazolium moiety, deuteration experiments (200 MHz 1H NMR analysis) on compounds 1, 13, and 15 were carried out.The results of these experiments are listed in Table 1. When these compounds were exposed to CD3OD or D2O containing 1 mol equiv or 5 mol % of Et3N at roomtemperature for 2-5 or 20 min, 100% deuteration of both C20 and C22 hydrogens was recognized. Surprisingly, the similar deuteration proceeded even without Et3N atroom temperature or 50 °C. The easy deuteration of the C20 and C22 hydrogens can be explained by the contributionof ylide 16 to the mechanism of the deuterationreaction.In conclusion, we have established useful, practical,and large-scale synthetic procedures for the preparationof mercaptobicyclotriazolium chloride 2 and biapenem 1and clarified some interesting physical and chemical characteristics of 1. (1)(a)Lederle(Japan).(b)Rigaku Corporation.(c)The University of Tokushima.(2)Shih,D.H.;Baker,F.;Cama,L.;Christensen,B.G.Heterocycles 1984,21,29.(3)(a)Sunagawa,M.;Matsumura,H.;Inoue,Y.;Fukasawa,M.;Kato,M.J.Antibiotics 1990,43,519. (b)Nagao,Y.;Nagase,Y.;Kumagai,T.;Matsunaga,H.;Abe,T.;Shimada,O.;Hayashi,T.;Inoue,Y.J.Org.Chem.1992,57,4243.(c)Ohtake,N.;Okamoto,O.;Mitomo,R.;Kato,Y.;Yamamoto,K.;HagaY.;Fukatsu,Y.;Nakagawa,S.J.Antibiot.1997,50,598.(d)Wildonger,K.J.;Ratcliffe,R.W.J.Antibiot.1993,46,1866.(4)(a)Guthikonda,R.N.;Cama,L.D.;Quesada,M.;Woods,M.F.;Salzmann,T.N.;Christensen,B.G.J.Med.Chem.1987,30,871.(b)Kim,C.U.;Luh,B.Y.;Misco,P.F.;Hichicock,M.J.J.Med.Chem.1989,32,601.(c)Ubukata,K.;Hikida,M.;Yoshida,M.;Nishiki,K.;Furukawa,Y.;Tashiro,K.;Konno,M.;Mitsuhashi,S.Antimicrob.AgentsChemother.1990,994.(d)Hikida,M.;Kawashima,K.;Nishiki,K.;Furukawa,Y.;Nishizawa,K.;Saito,I.;Kuwao,S.Antimicrob.Agents Chemother.1992,481(5)Andrus,A.;Baker,F.;Bouffard,F.A.;Cama,L.D.;Christensen,B.G.;Guthikonda,R.N.;Heck,J.V.;Joh nston,D.B.R.;Leanza,W.J.;Ratcliffe,R.W.;Salzmann,T.N.;Schmitt,S.M.;Shih,D.H.;Shah,N.V.;Wild onger,K.J.;Wilkening,R.R.In Recent Advances in theChemistry of β-Lactam Antibiotics;Roberts,S.M.,Brown, A.G.,Eds.;The Royal Society of Chemistry;London,1984;pp 86-99.(6)Kumagai,T.;Tamai,S.;Abe,T.;Nagase,Y.;Inoue,Y.;Nagao,Y.Heterocycles 1994,37,1521.(7)(a)Kumagai,T.;Matsunaga,H.;Machida,R.;Nagase,Y.;Hikida,M.;Nagao Y.Japan Kokai Tokkyo Koho(Japanese Patent),1989,JPS64-25779.(b)Abe,T.;Tamai,S.;Nagase,Y.Japan Kokai Tokkyo-Koho(Japanese Patent),1992,JP H4-230267.(c)Abe,T.;Tamai,S.;Nagase,Y.Japan Kokai Tokkyo Koho(Japanese Patent),1992,JP H4-230286.A Merck research group3d also achieved an expedit-ious synthesis of biapenem 1 utilizing our cyclization procedure3b,7 for the bicyclotriazolium derivative(8)Kumagai,T.;Abe,T.;Fujimoto,Y.;Hayashi,T.;Inoue,Y.;Nagao,Y.Heterocycles 1993,36,1729.(9)(a)Ratcliffe,R.W.;Wildonger,K.J.;Michele,L.D.;Douglas,A,W.;Hajdu,R.;Goegelman,R.T.;Springer,J.P.;Hishfield,J.J.Org.Chem.1989,54,653.(b)Yanagi,K.;Takeuchi,Y.;Sunagawa,M.ActaCrystallogr.1992,C48,1737.(c)Takeuchi,Y.;Inoue,T.;Sunagawa, M.J.Antibiot.1993,46,827.(10)The nonbonded O12---S16 atoms distance(2.976?)is significantly shorter than thesum(3.35)of the van der Waals radii of oxygen and sulfur atoms.In addition,a linear relationship between the O12 atom and the S16-C17σbond and a plane conformation of the O12-C11-C3-C2-S16 moiety can support the O12---S16 close contact.See,Kucsman,A.;Kapovits,I.Organic Sulfur Chemistry:Theoretical and Experimental Advances;Bernardi,F.,Csizmadia,I.G.,Mangini,A.,Eds.;Elsevier Scientific Publishing Co.:Amsterdam,1985;pp 191-245. Diazotization2of 22 with p-dodecylbenzeneadfonyl azidein the presence of Et2N in MeCN gave diazo compound27 in 90%yield.A solution of 27 in AcOEt was heated at 80 ℃in the presence of rhodium(I1)octanate2to give cyclization product 28.Compound 28 was treated with diphenyl chlorophosphateand diisopropylethylaminein MeCN to afford a solution of(dipheny1phosphono)-oxy derivative 29.Chromatographicpurification of the residue obtained by evaporation afforded pure 29 as colorless needles in 80%yield(Scheme V).However,the MeCN solution of 29 could be used directly for the subsequent Michael-type reaction with thiols 30 and 32.Thus,a solution of 29 was treated with 4-mercapto-N,N-bis(pnitrobenzyloxycarbony1)Pyrazolidine (30)21in the presenceof diisopropylethylaminein MeCN to give thioether 31 in 75%yield.After hydrogenolysis of the PNZ and PNB groups of 31,the resultant pyrazolidine compound was treated with ethyl formimidatehydrochloride in an alkaline solution to produce the desired triazolium carboxylate la,in 44%yield from 31,as a pale yellow,amorphous powder.N-Methy1-1,2,3-thiadiazolium-carboxylate lb was prepared from 29 according to the following procedure. The 29-containing solution prepared as described above was dissolved in THF and 0.35 M phosphate buffer solution(pH 7.0).The mixture was treated with 4-(mercaptomethyl)-N-methyl-1,2,3-thiadiazolium trifluoromethanesulfonate(32)22in alkaline,aqueous methanol to give the thiol adduct.The thiol ahydrogenolysiswith Zn powder to remove the PNB group to afford the desired product lb,in 54%yield from 27,as a pale yellow,amorphous powder.Both(-)-lβ-methylcarbapenems la and lb exhibited potent and broadspectrum antimicrobialactivities,.as do thienamycin and imipenem.23 Excellent stabilityof la against human renal dehydropeptidase-I was observed.23(1)(a)The University of Tokushima.(b)Lederle(Japan),Ltd.(2)Salzmann,T.N.;Ratcliffe.R.W.;Christensen,B.G.:Bouffard.F.A.J.Am.Chem.SOC~.1980,102~exit:6161.(3)(a)Nagao,Y.;Kumagai,T.;Nagase,Y.Tamai,S.;Inoue,Y.;Shuo,M.J.Og.Chem.,first of three papers in this issue.(b)Nagao,Y.;Nagase,Y.;Kumagai,T.;Kuramoto,Y.;Kobayaehi,S.;Inoue,Y.;Taga, T.; Ikeda,H.J.Org.Chem.,second of three papers in this issue.(4)Reference 3 and references cited therein.(5)(a)Shih,D.H.;Baker,F.;Cama,L.;Christensen,B.G.Heterocycles1984,21,29.(b)Shih,D.H.;Cama,L. Christensen,B.G.Tet-rahedron Lett.1985,26,587.(6)Nagao,Y.;Kumagai,T.;Tamai,S.;Abe,T.;Kuramoto,Y.;Taga,T.;Aoyagi,S.;Nagaae,Y.;Ochiai,M.;In oue,Y.;Fujita,L.M.;Shinkai,I.;Salzmann,T.N.J.Am.Chem.SOC.108,4673.(7)Fuentes,L.M.;Shinkai,I.;Salzmann,T.N.J.Am.Chem.SOC.1986,108,4675.(8)(a)Iimori,T.;Shibasaki,M.TetrahedronLett.1986,27,2149.(b)DCziel,R.;Favreau,D.Zbid.1986,27, 5687.(c)Kawabata,T.;Kimura,Y.;Ito,Y.;Terashima,S.Ibid.1986,27,62419)Nagao,Y.;Abe,T.;Shimizu,H.;Kumagai,T.;Inoue,Y.J.Chem.SOC.C,heh.Commun.1989,821.(10)(a)Nagao,Y.;Hagiwara,Y.;Kumagai,T.;Ochiai,M.;Inoue,T.;Hashimob,K.;Fujita,E.J.Org.Chem. 1986,51,2391.(b)Nagao,Y.;Dei,W.-M.;Ochiai,M.;Shiro,M.J.Org.Chem.1989,54,5211.(11)Nagao,Y.;Dai,W.-M.;Ochiai,M.;Tsukagoshi,S.;Fujita,E.J.Am.Chem.SOC.1988,110,289.(12)Iwasawa,N.;Mukaiyama,T.Chem.Lett.1983,297.(13)Nagao,Y.;Seno,K.;Kawabata,K.;Miyasaka,T.;Takao,S.;(14)Kokaltokkyokoho(Japanese Patent)1990-108664.(15)Kronenthanl,D.R.;Han,C.Y.;Taylor,M.K.J.Org.Chem.1982,(16)Evans,D.A.;Sjogren,E.B.Tetrahedron Lett.1986,31,4961.(17)Murahashi,S.;Naota,T.;Kuwabara,T.;Saito,T.;Kumobayashi,H.;Akutagawa,S.J.Am.Chem.SOC 1990,112,7820(18)After publicationof our previouspaper(seeref 2a),similar results were published by another group,see ref 8b.(19)lH NMR(400-MHZ)analysis of the tin(Ⅱ)enolates of 12b inTHF-d,at 0℃:d 1.76(d,J=6.8 Hz,allylic Me of the 2-enolate),1.62(d,J=6.8 Hz,allylic Me of the E-enolate),4.49(q,J=6.8 Hz,olefinic H of the 2-enolate),and 5.05(br q,J=6.8 Hz,olefinic H of the E-enolate).Assignment of the signale was confirmed by decoupling experiments.(20)’H NMR(400-MHz)analysis of the tin(I1)enolates of 12c inTHF-d,at-40℃:δ 1.76(d,J=6.3 Hz,allylic Me of the 2-enolate),1.68(d,J-6.3 He,allylicMe of the E-enolate),4.88(br q,olefinicH of the2-enolate,signalsoverlapped thwe of the thiazolidine SCH,-),and 5.07(brq,J=6.3 Hz,olefinicH of the E-enolate).Aeaignment of the signalsWBB confiied by the decoupl-ingexperiments.Cf.Heathcock,C.J.Org.Chem.1980,45,106(21)Kokaitokkyokoho(JapanesePatent)1989-25779(22)Kokaitokkyokoho(JapanesePatent)1989-25778.(23)(a)Reported by:Hikida,M.;Yoshida,Y.;Nishiki,K.;Furukawa.Y.and Mitsuhashi,S.28th Inters-cience Conference on Antimicrobia.Agents and Chemotherapy,Loa Angeles,CA,Oct (24)1988.(b)Hikida M.;Nishiki,K.;Furukawa,Y.;Nishizawa,K.;Saito,I.;Kuwao,S.submitted for publication in Antimicrob.Agents Chemother
外文翻译-1
β-内酰胺.3. 不对称全合成新的非自然1β-甲基碳青霉烯类对人类的肾脏脱氢肽酶-I展出的强大的抗菌活性和稳定性
Yoshimitsu Nagao,JaYunosuke Nagase,Toshio Kumagai,Hiroshi Matsunaga,Takao Abe,Osamu Shimada,Takaaki Hayashi,and Yoshinori Inoue
在1980年,默克公司的一个研究小组建立起了一个关于硫霉素的具有选择性实验,筛选出一个惊人的自然物质——β-内酰胺抗生素。还建立了一个极好的合成方法来合成碳青霉烯类。自那时以来,有许多研究报告了涉及合成硫酶素类和类似硫酶素类,发展出了特别引起人们的注意新的1β-甲基碳青霉烯类——非天然甲基碳青霉烯类。非天然甲基碳青霉烯类拥有强大的稳定对肾脏脱氢肽酶-I保持卓越的抗菌活性。在长尾和勒德尔(日本)集团,在富恩特斯,和其他团体,都报告了高度选择性烷基化的方法合成1β-甲基碳青霉烯类。
在前面的文献中,我们报告了高度选择性的4-乙酰氧基氮杂环丁酮和消旋的(3R*,m*)-3-[(1R*)-l-[(tert-butyldimethylsily1)-oxy]ethyl]- 4-乙酰氧基氮杂环丁酮和利用该反应产物的碳合成手性关键中间体的报告。我们继续研究一系列β-内酰胺的合成,我们现在详细介绍一种切实可行的方法使有用的碳能合成不对称的全新的非天然1β-甲基碳青霉烯类的1a和1b。这些特殊1β-甲基碳青霉烯类,附有铵作为杂环的硝酸铵作为感应在C-2位上,我们预计其将展出优秀的抗菌活性。特别是双环成分可视为手性σ -对称杂环的离域的π电子。合成1a和1b是用C-4的手性碳原子,在合成1β-甲基碳青霉烯类时建立连续四个不对称碳原子(C1,C5,C6,和C8)。我们采用了一种非对称,醛型乙醛反应与手性锡(Ⅱ)为烯醇的C6 形成C8的构成键,从而导致(3R,4R)-3-[(1R)-1-[(tert-butyldimethylsilyl)oxy]ethyl] 4-乙酰氧基氮杂环丁酮(3)具有光学活性. 一个高效的,选择性亚胺烷基的另一手性锡(Ⅱ)烯醇和手性碳原子从3原位获得通过C1形成碳的构成键。我们还利用的C - 4 -噻唑烷手性试剂的作用建立所有四个非对称中心在1a ,b,在1β-甲基碳
青霉烯类合成具有显着的特点。
3-[3-(Benzyloxy)propionyl]-(4R)-isopropyl-l,3-thiazolidine-2-thione(5),这步反应得到 3 - (苄氧基)丙酸( 4 )和(受体) -异丙基- 1 ,3 -噻唑- 2 -硫,是trifluoromethanesulfonate和N-乙基哌啶溶于CH2C12中在-78℃下反应2小时得到。过量添加乙醛并混合搅拌,通过HPLC分析在- 78℃下7a收益率达到 84 %。7a可通过合理化的方法形成过渡态6 ,其中用乙酰醛的办法是手性锡(Ⅱ)烯醇从较少阻碍β 的一边,及对面的α-异丙基噻唑基结合,形成六元环。在六元环中甲基乙醛是中间轨道。保护羟基的是7a与TBDMS,通过7a提交氨解与P -氨基苯甲醚反应,使酰胺8从7a中获得的产量为91 %。8a的氢的苄基是使化合物9产量高的主要原因。环酰胺9碳酸氢钠在4 : 1(CH2C12 :二甲基甲酰胺)的展开剂中顺利进行,使β -内酰胺10定量。脱氧化的产物10与硝酸铵反应提供了已知的化合物11,收益率可达到67 %(SchemeⅠ)。(3R,U)-3-[(1R)-l-[(tert-butyldimethylsilyl)oxy]ethyl]-4acetoxyazetidin-2-one(3) 对于转换效率11的测定是通过过氧乙酸与RuC13.nH20和已经取得的成就的Murahashi来完成的。
我们已经证明,3-acetyl-(4S)-ethyl-1,3-thiezolidine2-thione是匹配手性(3R,4R)-3-[(1R)-1-([tert-butyldi-methylsilyl)oxy]ethyl]-4-acetoxyazetidin-2-one(3) 的烷基化反应。因而,手性锡(Ⅱ) 产生原位电子化 3 -丙( 4 ) -乙基(和异丙基) -1,3 -噻唑- 2 -硫酮( 12a, b )锡(Ⅱ),在0℃下 三氟甲磺酸和 正丁基麟溶于THF中, 12a和3反应式遵循的是平常workup提供的,C -4 烷基化所带的17a和18a是通过高效液相色谱分析在90:10比率中得到的( SchemeⅡ和表一) 。 烷基化反应的3通过 12b作为类似的混合物17b和18b( 91:9 ) 。烷基化反应的手性C3 通过3 -丙衍生物还产生了17c,d的非手性的 1 ,3 -噻唑- 2 -硫酮。烷基化进展顺利,该二苯甲酰甲烷以及烷基化与C – 4手性噻唑烷的关系(见表一)。
主要产物( 17a, b )能确定它们是能够化学转换为已知化合物22,是合成1β-甲基碳青霉烯类关键中间体(见计划11)。化合物17a和17b,都是酰胺结构,分别用咪唑在MeCN在室温下形成鎓类咪唑19。化合物19发生的克莱森型原位脱羧得到β-酮酯21,由17a得到的产量为80 %和17b得到产量的86 %。在酸性约束条件下使化合物22消除TBDMS 得到21是很容易完成的。其他的主要产品(17c,d)证实了高效液相色谱法研究比较的数据得出该化合物来自脱水的l,3-thiazolidine-2-thione(or 4,4dimethyl-1,3-thiezolidine-2-thion)和羧酸20之间的缩合反应。(酸20 是19酸性水解得到的。)构成了未完成的C - 4烷基化产物( 18a) ,在方案四中描述了其化学性质与已知化合物25。化合物25准备是从23的以下列方式达到。甲酯24 ,是从23获得,用2摩尔的LDA的形成烯醇。该烯醇能形成甲基化的产品25。18a在碱性条件下水解成26及以后的dehydrative,由此得到羧酸26。18C在碱条件测定了其转换至26的过程(方案四) 。其他的微烷基化产品( 18b,d)也证实了这一事实,即他们的高效液相色谱保留时间相同得到的相同化合物,这些都是准备由脱水缩合羧酸26和相应的1,3 –噻唑- 2 -硫酮得到的。
因为差向异构化而新成立的β-甲基的主要产品由1 7a-b在从来没有转换的条件下得到,烷基化如上所述,主要产品17a-b和18a是未完成的产品,其解释如下。主要产品17a-b可从相应的烯醇化物Z -锡(Ⅱ)13a-b. 15c,d的手性的Z -锡(Ⅱ)形成过渡态,研发工作方法是从手性的13c,d侧对面的乙基的C - 3内酰胺开始。在一个类似的过程中,未完成的产物( 18a-b)可形成转型的16a,通过三维涉及电子锡(Ⅱ)烯醇化物14a-d, D和取得的不对称的原位3 。进一步的证据,这些猜测时,得到的取代的噻唑- 2 -硫基团发生了变化。该庞大的R1和/或R2的组的四氢噻唑- 2 -硫酮基影响了产品的主要成分的比例17α-D和未完成产物的18a(见表一) 。因此,烯醇的动能给予的Z –型和13c-d形成的基团,以取得理想的立体化学的烷基化的循环结果,如15a-d看来是必不可少的,。在我们研究的情况下,过渡态是导致动力学的Z -烯醇应该更加稳定,而导致动力学电子-烯醇因为后者负有严重的空间位阻斥力之间的甲基R1和/或R2的组(见表一)。
重氮化的22在与p-dodecylbenzeneadfonyl叠氮反应,是以Et3N与MeCN能形成重氮化合物27, 27 加入AcOEt并加热到80 ℃ 能得到收率为90 %,在铑(Ⅱ)的作用下得到28。28是二苯基磷酰氯和diisopropylethylaminein 的MeCN反应的结果,( dipheny1phosphono )含氧衍生物29。蒸发残渣后纯化获得29是无色针状物收益率为80 %(计划五) 。MeCN能使29直接用于随后迈克尔型反应的硫醇30和32 。因此,分解29得到4 -巯基- N , N -二( pnitrobenzyloxycarbonyl ) pyrazolidine ( 30 ) ,二异丙基乙胺硼烷络合物在MeCN反应后得到硫醚31的产量为75 %。在31连接的PNZ和PNB中,由此产生的吡唑烷在碱性溶液反应得到盐酸乙基甲亚胺,以产生预期的三唑翁硫酸甲酯盐羧酸镧,由31计收益率为44 %,是一种淡黄色,无定形粉末。N -甲基-1,2,3 - 三唑翁硫酸甲酯羧酸的合成是从29按照下列程序开始的。29在四氢呋喃和0.35 M磷酸盐缓冲液( pH值为7.0 )反应。混合物 4 - (巯基甲基) - N -甲基-1,2,3 - 三唑翁硫酸甲酯三氟甲( 32 )在碱性条件下,甲醇水溶液给硫醇加合物。加合物的巯基在氢锌粉末与消除的PNB提供理想的产物1b,由27计收益率为54 %,是淡黄色,无定形粉末, 我对卓越的稳定性的1a对人肾脏脱氢肽酶-I进行了观察得出,这两种1 β甲基碳青霉烯类1a和1b表现出强大和广谱抗菌活性。
外文翻译-2:
新型1β-甲基碳青霉烯类抗生素比阿培南
Toshio Kumagai,Satoshi Tamai, Takao Abe, Hiroshi Matsunaga,Kazuhiko Hayashi,Ikuo Kishi,Motoo Shiro, and Yoshimitsu Nagao,
介绍
(4R,5S,6S)-3-[(6,7-二氢-5H-吡唑 [1,2-a〕[l,2,4」三唑内鎓-6-基)〕-硫-6-[(R)-1-羟乙基〕-4-甲基碳青霉-2-烯-2-羧酸是一种新型1β-甲基碳青霉烯类抗生素, 具有范围很宽的抗菌活性, 卓越的化学稳定性,是一种对人体脱氢肽酶-I相对稳定的抗生素。甲基碳青霉烯类抗生素成功的合成开始于水合肼,是一种经济的可用的合成路线。能有效的利用硫醇迅速的合成比阿培南1。比阿培南1(晶体结构,由 S- - -O构象相互作用,以及CH- - -O氢键构成),是通过X射线调查晶体结构,氢核质谱,以及重氢试验分析所得。
结果与讨论
由于非自然1β-甲基碳青霉烯类抗生素发展显著,由默沙东研究小组发表的一些报告中,新特点的1β -甲基碳青霉烯类, 如梅罗-青霉烯, 3a比阿培南,3b 来那培南, 3c和.3b ,d, 4a,b.这些1β -甲基碳青霉烯类是有希望成为新一代非自然-内酰胺类抗生素,由于其出色的生物化学特性,我们已经开发了一套独特比阿培南13b及ασ-对称( 6,7 -二氢- 5H -吡唑并[ 1,2 - a ] [ 1,2,4 ] 三唑内鎓-6-基-)硫的合成。C2作为吊坠成分是明显不同与其他1β -甲基碳青霉烯类的3a,c .一般人都知道的C2型在碳青霉烯类抗生素中发挥着重要作用,其对肾功能和稳定的脱氢肽酶I型抗菌活性强,化学稳定性好。比阿培南是默克研究小组开发的一种具有显著作用的非天然1β-甲基碳青霉烯类抗生素,有几个报告显示1β-甲基碳青霉烯类的新特征,比如美洛培南, 比阿培南,来那培南以及其他。1β-甲基碳青霉烯类由于他们极好的生物和化学特性有希望成为新一代非天然抗生素。我们已经发展出一种独特的注射剂(6,7-二氢-5H-吡啉-[1,2-α][1,2,4] triazolium-6-酮)硫醇,其构造就象垂饰的一部分,不同于其他1β-甲基碳青霉烯类。具一般所知碳青霉烯抗生素的C2-垂饰一部分起着重要的作用在它的抗菌活性,化学稳定性以及对脱氢肽酶1的稳定性。我们通过一个回旋的合成路线来合成比阿培南,其中理想是其引入 1β-甲基碳青霉烯类双(4一毗哇烷)二硫化物衍生到1β-甲基碳青霉烯类骨架的14,3b。我们现在介绍另一种简单的合成比阿培南17,其表征基于X射线晶体学和核磁共振分析和氘的实验。比阿培南1展示广泛的强大的抗菌活性,卓越的化学稳定性、对人类的肾脏DHP-1广泛的稳定性。在1β-甲基碳青霉烯类双(4一毗哇烷)二硫化物的衍生物形成1β-甲基碳青霉烯类的骨架,这在早些时候,我们采用合成了合成路线的一部分就能期望的构造出来。另一种是直接描述我们现在的合成及其表征比阿培南根据x射线结晶和核磁共振氢谱分析和一个氘的实验。
提供良好的乙醇水溶液能成功的得到结晶1,当时通过给透视分析得到的晶体是无色针;熔点210 - 218 ℃ (分解) ; [α] 20 D -33.7 ° (c0.5 ,水)。该晶体结构描述出1结构,是C6 -羟乙基,1是不同于亚胺培南及美罗培南的。一个分子内非共价键O12 --- S16相互作用 (“密切接触” )是今天识别晶体结构1 的依据。这O12 --- S16密切联系,是其他内酰胺类抗生素共同特征。在D2O中要了解构C2和胶质瘤侧链1,必须进行核超过豪瑟效应(观察的)实验[核磁共振分析( 400兆赫) ]。改进选定的观察是图2所示。有趣的是,一个观察的提高,是H1和H8能确认频谱美罗培南。Co.research研究小组,没有承认是比阿培南 1。另一方面,一个观察中显示,增强观察频谱H8和H6 之间 ,没有发现时美罗培南。因此,主要构的C6 -羟乙基和C2连接的基团在D2O分析中可能是类似的晶体结构(图1 )。尤其是形成稳定的C2连接的集团可能是由分子内非共价键--- ö相互作用,并且由于到 ( O12 ) σ * ( S16- C17 )轨道重叠而不能发现。
有趣的是,在晶体结构可能又四种分子间氢键( O13 ---氢O15 , O12 ---氢C20 , O14 ---氢C22和O13 ---氢C22 ) 。该原子之间的距离(A---H: ca.2.03 - 2.28 A ),包括具体原子(A: O12 , O13 ,或O14 )与氢原子( h )的( D: O15 , C20或C22 )是非常短的,范德华半径之间的氧和氢原子总和为 2.72 A ,如图3所示。上文所述角度 (<A---H-D:123.6-149.5°) 氢键 也是足够大。为了得到一个洞察C20和C22的氢原子的具体成分,并进行了15个二元化合物的氘实验( 200兆赫的1H NMR分析)。这些实验结果列于表1。当这些化合物受到1摩尔当量CD3OD或D2O或5 %的Et3N分子在室温下20分钟,100 %氘的C20和C22连接两个氢原子被确认。令人惊讶的是,类似的氘实验即使在没有Et3N条件下,在室温或50 ℃也能形成。最简单的氘的C20和C22氢的发现,是叶立德16的机制氘反应可以解释的。
最后,我们已经建立了有益的,真实的,和大规模合成含硫侧链酰氯2和比阿培南1的方法,并澄清了一些有趣的物理和化学特性。
|
|