Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Synthesis and conformational studies of 10, 10-dimethyltridecanolide Hu, Thomas Qiuxiong 1988

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1988_A6_7 H82.pdf [ 4.07MB ]
Metadata
JSON: 831-1.0060354.json
JSON-LD: 831-1.0060354-ld.json
RDF/XML (Pretty): 831-1.0060354-rdf.xml
RDF/JSON: 831-1.0060354-rdf.json
Turtle: 831-1.0060354-turtle.txt
N-Triples: 831-1.0060354-rdf-ntriples.txt
Original Record: 831-1.0060354-source.json
Full Text
831-1.0060354-fulltext.txt
Citation
831-1.0060354.ris

Full Text

SYNTHESIS AND CONFORMATIONAL STUDIES O F 1 0 , 1 0 - D I M E T H Y L T R I D E C A N O L I D E By Thomas Qiuxiong Hu B . S c , South Ch ina Institute of Technology, 1985 A T H E S I S SUBMITTED IN PARTIAL FULFILLMENT O F THE REQUIREMENTS FOR THE D E G R E E OF MASTER O F SCIENCE i n THE FACULTY OF GRADUATE STUDIES Department of Chemist ry W e accept this thesis as conforming to the required standard T H E UNIVERSITY O F BRITISH COLUMBIA August 1988 Thomas Qiuxiong Hu, 1988 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of QAB^^T^Y The University of British Columbia Vancouver, Canada Date C > c l ' f f i DE-6 (2/88) i i ABSTRACT The synthesis of 10,10-dimethyltridecanolide (42) was achieved via a fifteen-step sequence in 9% overall yield. The hydrolysis of macrolides 412., 3_5_, and ester 109 was used to probe the conformational behavior of macrolide 42- The results of this study were rationalized through molecular mechanics (MM2) calculations of conformations for macrolide 42-MM2 studies confirmed initial conformational analyses that macrolide 42 should exist predominantly in the [3434] conformation 42a. More importantly, they also revealed the existence of a [3344] conformation 421-Hydrolysis studies showed that macrolides 42 and 3J> hydrolyzed more slowly than ester 109 due to the steric effect of the intermediates. They also suggested that the minor conformation 421 very likely controlled the hydrolysis process of macrolide 42. 21 101 i i i Table of Contents Page Abstract i i List of Figures iv List of Abbreviations -v Acknowledgements v i Introduction 1 1.1 Lactonization of Long-Chain Hydroxy Acids 5 1.2 Remote Asymmetric Induction Via Conformational Control 1 0 1.3 Conformational Behavior of 14-Membered Lactones 1 3 1.4 Restricting the Low-Energy Conformations of 14-Membered Lactones by Geminal Disubstitution 1 9 Results and Discussion 2 0 2.1 Synthesis of 10,10-Dimethyltridecanolide 2 0 2.2 Hydrolysis of 10,10-Dimethyltridecanolide,Tridecanolide and n-Octyl Pentanoate...3 8 2.3 Molecular Mechanics Calculations of Conformations for 10,10-Dimethyltridecanolide 4 2 2.4 Conclusion 54 Experimental 55 3.0 General 5 5 3.1 Preparation of 10,10-Dimethyltridecanolide 5 8 3.2 Preparation of n-Octyl Pentanoate 73 3.3 Preparation of Tridecanolide 74 3.4 Hydrolysis of 10,10-Dimethyltridecanolide.Tridecanolide and n-Octyl Pentanoate...75 References 7 6 Appendix 1 (MM2 calculations), Appendix 2 (NMR and IR spectra) 8 0 i v List of Figures Figure Title Page 1 Structures of macrol ide antibiotics isolated in the 1950's 2 2 Conformation models for cyclotetradecane and erythronolide B 4 3 Magnitude of hydrogen interactions in the [3434] conformation of cyclotetradecane 1 4 4 Top and side views of the [3434] conformation of cyclotetradecane 1 5 5 Retrosynthetic analysis of macrol ide 12 21 6 Plot of the rates of the hydrolysis of 4 2 , 2 5 . and 109 at room temperature 4 0 7 Computer plots of the [3434] conformations for macrol ide 42 4 5 8 The polar map conventions 4 6 9 The [3434] conformation of cyclotetradecane and the polar map of its torsional angles 4 7 1 0 Superposi t ion of the polar maps of the ideal [3434] conformation (broken line) and the [3434] conformations of macrol ide 4J> 4 8 1 1 The [3344] and [3335] conformations of cyclotetradecane and their polar maps 5 0 1 2 Computer plots and polar maps of the [3344] and [3335] conformations of macrol ide 4J> 51 V List of Abbreviations AcO acetoxy] AC2O acet ic anhydride C D c i rcu lar d i ch ro ism D C C 1 ,3 -d i cyc l ohexy l ca rbod i im ide D M A P 4 -d ime thy lam inopy r i d i ne e ther diethyl ether g lc gas- l iqu id chromatography h h o u r ( s ) H M P A hexamethylphosphoramide IR i n f r a red m i n m i n u t e ( s ) M C P B A meta-ch loroperoxybenzoic ac id M S mass spectrometry N M R nuclear magnetic resonance P P T s pyridinium p_-toluenesulphonate TEA t r i e t hy l am ine THF te t rahydro fu ran T H P te t rahydropy ran t i c thin layer chromatography T B D M S t e r t - b u t y l d i m e t h y l s i l y l T M S t r i m e t h y l s i l y l v i ACKNOWLEDGEMENTS I w ish to exp ress my s incere grati tude to my superv isor , Dr. Larry Wei ler , for his guidance and advice during the course of this work. I am indebted to Dr. Edward Nee land for his va luab le d i scuss ions and construct ive suggest ions concerning the progress of this research and the preparation of this thesis. Thanks are also due to Karin Albert, Sigr id Albert, Pau l Cheung , R o s s Lonergan, Margot Purdon, Annie Wong , G race W u , Jackson Wu and Gera ld Yeung whose friendships have made the past two years a time to remember. In addit ion, f inancial ass is tance in the form of graduate fel lowships from the University of British Co lumbia and the efficient cooperation of the staff of the N M R and mass spectrometry service are gratefully acknowledged. Finally and most importantly, I would like to thank my parents for their encouragement, pat ience and support throughout the course of my educat ion. My gratitude to them is beyond mere words. This thesis is dedicated to my parents. 1 CHAPTER ONE I n t r o d u c t i o n Perhaps one of the most significant breakthroughs in the development of natural product chemistry was the isolation of pikromycin (1) by Brockmann and Henkel in 1950.1 It was the first macrolide antibiotic isolated from a bacterial source. Soon afterwards, several other microbially produced antibiotics were discovered which were thought to be structurally related to pikromycin. By the end of 1957, chemical degradation studies led to the revelation of the gross structures of methymycin (2J, erythromycin A (3J and B (4) and carbomycin A (5J (Figure 1 ) . 2 - 5 Each of these antibiotics shared a common feature - a lactone incorporated in a medium or large-ring system. This was the genesis of macrolide chemistry. O II 1 The term "macrolide" originally referred to the above antibiotics but, as time passed, it has gradually been used in a broader sense to define all organic compounds with a large lactone ring (12 or more atoms). 2 Figure 1. Structures of macrolide antibiotics isolated in the 1950's, R, R-j and R 2 represent different carbohydrates. Over the next 20 years, the field of macrolide chemistry was firmly established with the isolation of more than 100 large ring lactone natural products possessing diverse biological activities.^ Approximately one-half of these natural products are classified as "polyoxo" 3 macrolides, which normally contain 12, 14 or 16-membered lactone rings with numerous ring substituents including one to three glycoside units. An interesting structural feature of the macrolide antibiotics, which may be related to their biological activities, is the relative conformational rigidity of the molecules in solution. These molecules often exist in one identical conformation whether in solution or in the solid state.7 Presumably, the high degree of substitution on the macrolide plays an important role in this rigidity in that the accommodation of many steric and electronic factors leads to a single minimum energy conformation for any particular macrolide. The conformations of complex 14-membered macrolide antibiotics have aroused considerable interest among chemists and a number of models have been proposed to explain the conformations of various macrolides. One of the most studied conformations is that of erythronolide B, the aglycone of erythromycin B. In 1963, Dale introduced a minimum energy diamond-lattice conformation 6_ (bold line) for cyclotetradecane.8 A model for the preferred conformer of erythronolide B (7_) was advanced by Celmer in 1965.9 Although Celmer's model was in close agreement with the 1 H NMR data, it did not agree with X-ray crystallographic data. Celmer's conformation was also very strained because of transannular hydrogen interactions and a 1,3-diaxial methyl interaction. A second diamond-lattice conformation 8_ was proposed for erythronolide B (9_) from NMR and CD spectra data. This eventually led to the Perun model IQ..10 The Perun model was favourably compared to the solid state conformation of erythromycin A 1 0 and is now widely used as a basis for assigning stereochemistries of substituents in 14-membered macrolide antibiotics. C H , 1Q Figure 2. Conformation models for cyclotetradecane and erythronolide B. Although a considerable number of conformational studies of macrolide antibiotics have been carried out over the past 30 years, the total synthesis of macrolide antibiotics has been slow to develop. There were two major problems associated with the synthesis of macrolide antibiotics: one was the construction of a medium or large-size lactone and the other involved the introduction of numerous chiral centres into the molecule. Efficient lactone ring construction procedures are now available due to the development of new synthetic methodology and advanced experimental techniques. However, selective introduction of chiral centres on a large ring remains challenging and still demands creativity and imagination for its satisfactory completion. 5 1.1 Lactonization of Long-Chain Hydroxy Acids The first problem associated with the synthesis of a macrolide antibiotic is the construction of the lactone functionality. Although many methods to generate lactones have been developed over the past 30 y e a r s , 1 1 " 1 9 lactonization of long-chain hydroxy acids is by far the most common. This stems from the ready availability of the acyclic precursors and from the belief that the biosynthesis of a macrolide antibiotic very likely proceeds through a final lactonization stage to form the aglycone. However, the employment of the lactonization procedure, while conceptually simple, is not always an easy task to accomplish. Ring closure of a long-chain precursor is disfavoured due to an entropy loss as the two distant chain ends approach each other. As a result, intermolecular rather than intramolecular cyclization often occurs. Nevertheless, efforts to develop highly effective lactonization procedures have proven to be successful over the years. The first lactonization of long-chain hydroxy acids i i was carried out by Stoll and Rouve in 1934 . 2 0 In their studies, the use of dilute solutions (e.g. 2.0 - 8.0 x 10~ 4 M) and catalysts such as p-toluenesulfonic acid were found to be essential. The synthesis of lactones has been conducted in this manner with yields ranging from approximately 1% for nonanolide (12.) to 8 7 % for hexadecanolide (13_). 2 0 However, this method suffers from the awkward high dilution conditions needed to effect lactonizations and from the resulting poor yields of the lactones. o 12 The desire to perform cyclizations under milder conditions and with higher yields led to a strategy which activated the carboxylic acid before the lactonization. It was anticipated that the intramolecular attack of the free hydroxyl functionality on the newly activated carboxylic acid would occur under relatively mild conditions. The reaction of hydroxy acids with phosgene-triethylamine or with trifluoroacetic anhydride has successfully produced lactones through the formation of a mixed anhydride 14. For example, treatment of the hydroxy acid 15. with trifluoroacetic anhydride followed :=o 14 by cleavage of the methoxyl ethers gave zearalenone in moderate yield. 21 M e O O 1) ( C F 3 C O ) 2 0 2) BBr 3 H( O 15 l i 7 A different strategy for the cyc l izat ion of hydroxy ac ids has a lso been reported by Kur ihara et a l . , wherein the hydroxyl function was activated and the carboxylate anion acted as the n u c l e o p h i l e . 2 2 In this way, the reaction of the hydroxy ac ids with tr iphenylphosphine and d i e t h y l a z o d i c a r b o x y l a t e (17) at room tempera tu re e f fec ted r ing c l o s u r e s v i a an a lkoxyphosphonium carboxylate IS.. + Ph3P=0 18 This procedure has been employed by White and co-workers in their synthes is of vermicu l ine ( 1 9 ) . 2 3 Cycl izat ion of the hydroxy acid 2Q. with t r iphenylphosphine and diethyl azod icarboxy la te p roduced vermicul ine (19J in 1 5 % y ie ld. This process also resulted in an inversion of the stereochemistry of the original a l c o h o l . 2 3 20. 12 The advantage of act ivat ing the carboxy l ic ac id or hydroxyl funct ion prior to the esterif ication step w a s obv ious. A n even more expedient procedure involved simultaneously 8 activating both the hydroxyl and carboxylic acid groups. Corey and Nicolaou 2 4 envisioned a "double activation" through a carboxylic derivative that was able to transfer a proton from the hydroxyl to the carboxylic oxygen. One such derivative was the 2-pyridinethiol ester 21 prepared from the reaction of a hydroxy acid and di(2-pyridyl)disulfide (22) in the presence of triphenylphosphine. 2^ The proton transfer from the hydroxyl to carbonyl oxygen in 21 was greatly facilitated by the basic nitrogen of the pyridine ring. The zwitterionic intermediate 22. thus generated underwent a facile cyclization to eventually yield the desired lactone 24. G: OH Ph3P 22 21 22 r 'H 24 o N xs H Corey and co-workers have successfully applied this methodology in the synthesis of several complex macrolides. 2 4 In their synthesis of zearalenone (1_6J, the key lactonization step was accomplished by heating the hydroxy acid 23. with di(2-pyridyl)disulfide and triphenylphosphine in benzene. Subsequent hydrolysis of the ketal and tetrahydropyranyl ether protecting groups produced zearalenone (16.) in 7 5 % yield overall. 2 4 THPO a . 1) Ph 3P, ^ N ^ S ^ 2) H 30 + 21 16 9 Recent ly, a modification of Corey 's double activation method was reported by Ger lach and T h a l m a n n 2 6 who found that the presence of silver ions ( A g C I 0 4 or A g B F 4 ) complexed with 2-pyridinethiol esters as shown in 2j6_ which then were able to undergo a rapid cycl ization at room tempera ture . A problem associated with Corey 's double activation method is that the reaction products must be separa ted from thiopyr idone, t r iphenylphosphine oxide, dipyridyl sul f ide, and in the c a s e of Ag-act ivat ion, from excess silver thiolate. To circumvent this problem, another simple and highly effective cycl izat ion method has been developed by Schmidt and D i e t s c h e . 2 7 In this method, 1-phenyl-2-tetrazol ine-5-thione and tert-butyl isocyanide were used for the double activation of the hydroxy acids to produce 16, 17, 18, and 20-membered lactones in over 9 0 % yield. The purification of lactones so produced was uncomplicated. Previously, our laboratory has used a modif ied form of Corey 's double activation method for the construction of 14-membered lactones of interest. Al though the lactonization reactions proceeded smoothly with moderate to good yields, the difficulties in purifying the final products remained. In this project, Schmidt and Dietsche's method was appl ied to the cycl ization step in the syn thes i s of 10 ,10-d imethy l t r idecano l ide , a st rategical ly d imethy la ted 14 -membered lactone. 10 1.2 Remote Asymmetr ic Induction V ia Conformational Control The second problem encountered in the synthes is of macrol ide antibiot ics was the establ ishment of the relative stereochemistry of numerous asymmetr ic centres on an acyc l ic precursor. This is a problem that continues to chal lenge synthetic chemists. There are a var iety of synthet ic methods ava i lab le to es tab l i sh v ic ina l or 1,2-s te reochemica l re la t ionsh ips us ing react ions that p roceed with high internal or relat ive asymmetr ic induction, and to set up 1,3 and a few 1,4-relationships using reactions based on well understood 5 and 6-membered ring conformations. However , the es tab l ishment of correct s te reochemica l re lat ionships be tween widely separated or remote asymmetr ic centres usually requires some form of absolute stereochemical control. Each set of remote asymmetr ic centres can only be prepared from an enantiomerical ly pure starting material or by a reaction proceeding with high enantioselectivity. Tradit ional ly, the large number of asymmetr ic centres are introduced into the acyc l ic precursors of the macrol ides using enantiomerical ly pure starting mater ials, but this strategy is very difficult to accompl i sh . A new strategy for the es tab l i shment of remote s te reochemica l re la t ionsh ips in macro l ide synthes is has been deve loped by Stil l and c o - w o r k e r s 2 * * - 2 9 who exploi ted the conformations of medium or large-membered lactones as a source of stereocontrol in generating new asymmetr ic cent res . In their synthes is of 3-deoxyrosaranol ide ( 2 7 ) . 2 9 the s imple 16-membered lactone 2JL was first constructed. The other six asymmetr ic centres in the target molecule 2Z were then establ ished with excellent diastereoselectivity using the two asymmetr ic centres at C i 4 and C - j 5 , and the conformational preference of the 16-membered lactone. 2& 22 A l t h o u g h the c o n f o r m a t i o n u l t imate ly r e s p o n s i b l e for the s u c c e s s of this stereoselect ivi ty w a s not determined, this synthesis clearly i l lustrated the concept of remote asymmetr ic induction v ia conformat ional control : the pre-exist ing asymmetr ic centre(s) in a l a r g e - m e m b e r e d lac tone c a u s e d the mo lecu le to adopt a par t icu lar con format ion or conformations which then directed the stereochemical outcome of the reaction (e.g. by allowing the attack of a reagent to occur from only one face of the molecule). Interestingly, Stil l has also extended this conformat ional stereocontrol strategy to the preparation of s te reochemica l ^ complex acycl ic compounds. This strategy entails the synthesis of a medium or l a rge -membered lac tone, e laborat ion of the des i red s tereochemist ry via conformational control and c leavage of the lactone ring. For example , in the synthesis of the tr is(tetrahydrofuranoid) c o m p o u n d 2J.,30 a 16-membered lactone 20. was first constructed. Treatment of the lactone 20. with meta -ch lo roperoxybenzo ic ac id gave the tr iepoxide 21, in greater than 9 0 % d iastereoselect iv i ty . Sapon i f i ca t ion of 21 fo l lowed by ac id ca ta lyzed polycycl izat ion led directly to the desi red product £ £ . 2Q 21 21 Unfortunately, the conformation of the reacting lactone was not determined. However, it is c lear that the use of conformat ional control in the introduction of asymmetr ic centres will eventually provide easy a c c e s s to a large number of complex macrol ide antibiotics and some acyc l ic natural products which otherwise may have been difficult to synthesize. Obv ious ly , in order to accelerate and extend the appl icat ion of this new strategy, an understanding of the conformational behavior of large lactone rings and the factors controll ing their conformat ional p re ferences i s : important. Th is knowledge can a id in predict ing and controll ing the stereochemical outcome of a reaction involving in the lactone ring. A s a contr ibut ion to the deve lopment of this conformat ional control strategy, our laboratory has been involved in the studies of the conformat ional behavior of 14-membered lactones and the fundamental principles governing the stereochemistry of react ions of these compounds . The 14-membered lactones were chosen as our primary targets because many macrol ide antibiotics contain such a lactone skeleton and also 14-membered rings are the next ring s ize after cyc lohexane which can adopt a strain-free con fo rma t i on . 3 1 1.3 Conformational Behavior of 14-Membered Lactones The simplest 14-membered ring is the hydrocarbon cyclotetradecane whose conformation has been described in detail by D a l e . 8 , 3 1 From an inspection of space-filling models, Dale recognized a tendency for saturated large rings to adopt compact conformations consisting of two parallel methylene chains linked by bridges of minimum length. These rectangularly shaped conformations could be generated from the three-dimensional framework of a diamond-lattice 22 where the ideal tetrahedral carbon geometry and the favorable torsional angles of 60° or 180° were obtained. 8 Dale predicted a strain-free (i.e. free of bond angle strain, torsional angle strain and severe transannular hydrogen interactions), lowest-energy conformation 22. for cyclotetradecane and named it as the [3434] conformation. 3 2 The numbers within the square brackets represent the number of carbon-carbon bonds between atoms having two adjoining gauche angles of equal sign each followed by an anti a n g l e . 3 2 , 3 3 Dale's prediction was later confirmed by X-ray analysis of cyclotetradecane in the solid state. 3 4 At first glance, Dale's [3434] conformation of cyclotetradecane shares many similarities with the well-known chair conformation of cyclohexane. However, a close comparison of both strain-free conformations reveals an important difference between them. The 14-membered strain-free conformation is less symmetric and as such contains four diastereotopic methylene environments. The four unique methylenes are numbered in Figure 3 14 which a lso i l lustrates the relat ive magni tude of t ransannu lar in teract ions that internal hydrogen atoms exper ience in this conformation. o Least Severe Mos Severe Figure 3. Magnitude of hydrogen interactions in the [3434] conformation of cyclotetradecane. The different magni tude of the hydrogen interactions has a profound effect on the preferences for substitution of a ring carbon atom as well as substitution of the hydrogen atoms. S ince replacement of an s p 3 carbon atom in the ring with an oxygen or with an s p 2 carbon atom will reduce the transannular hydrogen repulsions, the preference for such a substitution should therefore fol low the order > C 4 > C 2 > C 3 . For hydrogen rep lacement , two different subst i tut ion patterns ex is t : a s ing le subst i tut ion or a gemina l d isubst i tu t ion. A s ing le substituent attached to any of the carbon atoms can only occupy exterior posit ions. Otherwise the transannular interactions would be prohibitively l a r g e . 3 3 , 3 4 This will lead to a mixture of four conformers which differ from one another only in the choice of the substituent posit ion. The two geminal substituents, however, will be restricted to a corner posit ion C3 s ince it is only at this posit ion that the substituents exper ience little transannular interactions. Dale defined the corner atom as having two adjoining gauche angles of equal sign each followed by an anti angle ( e .g . 180°, -60° , -60° , 1 8 0 ° ) . 3 2 A corner posit ion represents the sole posit ion 1 5 avai lable for geminal disubstitution in that substi tuents on this posit ion exper ience the least amount of steric interaction from the r i n g . 3 3 ' 3 4 A top v iew diagram of the [3434] strain-free conformation 24. immediately reveals the corner posit ions as carbon atoms 3, 6, 10 and 13. Figure 4. Top and side views of the [3434] conformation of cyclotetradecane. The ring sys tem which is of primary interest in this project, the 14-membered lactone, can now be cons idered. The introduction of an ester l inkage into the cyclotetradecane ring should not distort the [3434] conformation of the ring itself, s ince no new angular strain is introduced. Dale 's strain-free [3434] conformation for cyclotetradecane can also be reasonably expected to be the lowest-energy conformat ion for most of the 14-membered lac tones, except for highly substi tuted ones with demand ing steric and geometr ic requirements. O n e such except ion is erythronolide B ( l f i j where the Perun model must be adopted in order to explain experimental r esu l t s . For reasons of brevity, only 14-membered lac tones which we expect to adopt the [3434] conformat ion will be d i s c u s s e d in detai l and publ icat ions concern ing those with conformat ions other than [3434] are g iven for i n t e r e s t . 3 5 , 3 6 It should be emphas ized that most simple 14-membered lactones seem to adopt the [3434] conformation as demonstrated by our laboratory through deta i led conformat ional ana lyses using X- ray crysta l lography, N M R spect roscopy and molecular mechan ics (MM2) c a l c u l a t i o n s . 3 7 " 3 9 13 10 21 16 E v e n if our ana l ys i s is restr ic ted to the [3434] conformat ion , the s implest 14-m e m b e r e d lac tone t r i decano l ide (35) can still be expec ted to exist in seven different con fo rmers . However , the preference of a planar s-trans geometry for esters immediately el iminates four of the possib le conformers. It has been reported by D e s l o n g c h a m p s 4 0 that the planar s-trans geometry in ester 3_fL is approximately 3.0 kcal /mole more stable than the planar s-c is isomer 2Z. In fact, there is not a single example of an acycl ic s-cis e s t e r . 4 1 O A s - cis 21 The 14-membered ring is large enough to accommodate a p lanar s-trans l inkage as conf i rmed by dipole moment measurements 4 2 , 4 3 Tr idecanol ide (25.) has a dipole moment of 1.86 D e b y e s , 4 2 similar to those of acycl ic esters which range from 1.6 to 2.0 D e b y e s . 4 3 On the other hand, 8-valerolactone (3Ji) where the lactone group is held rigidly in the s-cis l inkage has a dipole moment of 4.22 Debyes. s - cis For tr idecanolide only three [3434] conformations 35a, 25b. and 25c can accommodate a planar s-trans lactone. The relative steric energies of these conformations have been found to be 0.2 kca l /mole , 0.0 kcal /mole and 0.1 kca l /mole, respect ively, by M M 2 c a l c u l a t i o n s . 4 4 In contrast, the relative steric energ ies of the four conformers with s-c is lactone l inkages have b e e n found to be 7.2 kca l /mo le , 7.8 kca l /mo le , 7.9 kca l /mo le and 8.2 k c a l / m o l e . 4 4 Accord ingly , conformat ions possess ing s-trans lactones should predominate and tr idecanolide (25) should exist as a mixture of conformers 3 5 a . 35b and 35c. in nearly equal amounts. 252 a£b 35£ Obviously, in order to exploit the conformations of 14-membered lactones as a source of s tereocont ro l , the number of low-energy react ing conformat ions must be reduced to a minimum. This can be achieved by introducing rigidity to the lactone rings through appropriate substi tut ions and/or funct ional izat ions. B a s e d on the simple conformational analyses of 14-membered lactones d iscussed above, our laboratory has been able to plan the synthes is of con fo rmat iona l ^ rigid 14-membered lactones and take advantage of their expected [3434] conformat ions to successfu l ly control d iastereoselect ive react ions. For examp le , en route to the syn thes is of zerano l ( 3 9 h the requi red 9 R * , 1 3 S * stereochemist ry was es tab l ished v ia a [3434] con fo rmat iona l ^ control led d iastereoselect ive reduct ion react ion. The preference of a p lanar s-trans lactone l inkage and the preferred occupat ion of a carbonyl at a non-corner posit ion resulted in a single [3434] conformation 4 0 a for macrol ide 4Q.. When this macrol ide was treated with L-Selectr ide, reduction occurred from the more o p e n e x o c y c l i c f a c e , l ead ing to the d e s i r e d product 4 J _ with exce l l en t d ias te reose lec t i v i t y . 1 9 1.4 Restricting the Low-Energy Conformations of 14-Membered Lactones bv Gemina l Disubst i tu t ion It is c lear that 14 -membered lac tones are potent ial ly usefu l as th ree-d imens iona l templates for the introduction of asymmetr ic centres. In our laboratory, var ious 14-membered lactones have been synthesized and investigated for conformationally controlled reactions over the past few years . During the course of these studies, one common problem a rose : often, the 14-membered lactone under investigation p o s s e s s e d more than one stable low-energy conformation which compl icated the analys is and reduced the utility of the results obtained. It was felt that a synthesis and study of a lactone with one stable conformation would provide valuable insight into the conformat ional and chemica l behavior of 14-membered lac tones. The total synthes is of 10,10-d imethy l t r idecanol ide (42) was conducted with this goal in mind. One of the many efficient ways to introduce rigidity into the lactone rings, and therefore reduce the number of conformat ions avai lab le , is by gemina l disubst i tut ion. The gemina l subst i tuents shou ld force a 14 -membered lactone to adopt a conformat ion in wh ich the quaternary carbon atom occupies a corner posit ion. 10,10-Dimethy l t r idecanol ide (4.2) could be expected to exist predominantly in a single rigid [3434] conformat ion 4 2 a in which the gemina l dimethyl group occup ies the corner posit ion and the lactone function has a planar s-trans geometry. Conformat ional and reactivity studies of this compound should lead to a better understanding of the conformat ional and chemica l behavior of 14-membered lactones. 20 CHAPTER TWO  Results and Discussion This chapter cons is ts of four sec t ions : (1) the synthesis of the target molecule 10,10-d imethy l t r idecanol ide (42); (2) hydrolysis studies of two macro l ides 42, 25., and an ester 1 0 9 : (3) MM2 ca lcu la t ions of conformat ions for macro l ide 42. and (4) conc lus ions from these investigations and future considerat ions. 2.1 Syn thes is of 10.10-Dimethy l t r idecanol ide (42^ T h e syn the t ic route to 10 ,10-d imethy l t r idecano l ide (42) was dev i sed using the fo l lowing cons idera t ions . A gemina l d imethyl group was to be in t roduced prior to the construction of the lactone ring so that it could serve to reduce the number of conformations avai lable to the remainder of the r ing. The lactone r ing, on the other hand , was to be constructed v ia the lactonization of a long chain hydroxy ac id precursor. Thus , the penultimate target for the synthesis became the co-hydroxy ac id 42 with a geminal dimethyl group 6 to the hydroxyl function (Figure 5). S ince the geminal dimethyl group was an isolated one, direct alkylation was not feasible. Neve r the less , it cou ld be genera ted from a keto functionali ty v i a a ser ies of chem ica l transformations. The carboxyl ic ac id functionality in 42 could be prepared from the oxidation of a pre-exist ing hydroxyl group. Hence , the synthetic problem was further reduced to the preparat ion of the 7-hydroxy ketone 44- Further d isconnect ion of the synthetic intermediate 44. w a s made in v iew of the fact that the most logical precursor of a -y-hydroxy ke tone derivative was y-butyrolactone (45) coupled with the corresponding Gr ignard reagent 45. The Gr ignard reagent 45 cou ld be prepared from the commercial ly avai lable 1,9-nonanediol (47). Th is comple ted our retrosynthetic ana lys is and it sugges ted an efficient approach to the synthesis of macrol ide 42. 21 Figure 5. Retrosynthetic analysis of macrolide 42.-Based on the synthetic scheme outlined above, our synthesis was initiated with the monobromination of 1,9-nonanediol (47). This reaction was achieved by continuously extracting a mixture of the diol and aqueous hydrobromic acid with heptane. The yield obtained after column chromatographic purification was 84%. The infrared (IR) spectrum of the bromo alcohol 48. showed a hydroxyl stretching absorption at 3337 cm" 1. The 1 H NMR spectrum of 42. exhibited two sets of triplets at 5 3.42 and 5 3.66 which were attributed to the a-methylene protons of the bromo and hydroxyl groups respectively. HBr 42 41 22 Protection of the hydroxyl group of the bromo alcohol 45 was carr ied out by treatment of this material with dihydropyran in the presence of pyridinium p-toluenesulphonate ( P P T s ) . The reaction proceeded smoothly to give the tetrahydropyranyl ether 42 in 9 1 % yield. Absence of a hydroxyl absorption in the IR spectrum of 42. indicated a successfu l react ion. A triplet at 8 4.60 in the 1 H N M R spectrum of 42 confirmed the presence of the acetal methine proton. OTHP The next step in the synthesis was to convert compound 42 into a nucleophi le and prepare a 7-hydroxy ketone derivative v ia a nucleophi l ic addition to 7 -butyro lactone. When the bromide 42 was treated with magnes ium and the resulting Gr ignard reagent added slowly to 7-butyrolactone, the des i red 7-hydroxy ketone could not be detected. Instead, we isolated the diol 52 which was the product of the addition of two equivalents of Gr ignard reagent to the lactone. It appeared that the second Grignard addition to the ring opened compound was faster than addition to the 7-butyrolactone. OTHP In another attempt to prepare the des i red 7-hydroxy ketone, organol i th ium reagents were invest igated. The simple alkyl bromide 51 reacted with lithium to give the organolithium reagent 52. which successfu l ly opened 7-butyrolactone to give the 7-hydroxy ketone 52 in 5 6 % 23 yield. Surprisingly, when the tetrahydropyranyloxy bromide 49_ and the benzyloxy bromide £4. were subjected to this procedure, no reaction was observed and the starting bromides were recovered. *> u y , 2 ) Q At this point, we were forced to search for alternate conditions for this reaction. Our attention turned to the use of sulfone chemistry. It had been reported by Covicchioli et a l . 4 5 that a.oc-dilithio derivatives of alkyl phenyl sulfones react with small and medium-size lactones to give the corresponding hydroxy ketones. The a.rx-dilithio derivatives of alkyl phenyl sulfones were prepared by treatment of the sulfones 5JL with two equivalents of a-butyllithium. The attack of lactones by these gem-dimetallic compounds afforded intermediate enolates 5£. which upon workup gave the single addition products 5_Z. The formation of diols by a second organometallic addition was therefore avoided. OLi 2n-BuLi p ^ y ^ o i i ] aqueousNH<CI. ^ Y ^ S O H R R 55 0 56 52 24 To apply Covicchio l i 's method in our synthesis, the bromide 42 must first be converted into its corresponding phenyl sulfone. Severa l methods are available for such a convers ion. Direct alkylat ion of alkal i metal sal ts of b e n z e n e su lph in ic ac ids is a widely used p rocedu re . 4 * * The reaction is usually performed in refluxing alcohol or in dimethylformamide at room temperature, but it proceeds slowly and often gives only moderate yields of sulfones. An improved synthesis of sulfones has been reported by Veenstra and Z w a n e n b u r g 4 7 who u s e d tetrabuty lammonium p-to luenesulphonate (55) as the nucleophile in reactions with alkyl hal ides to give the corresponding sul fones 52 in good yields. Compound 55 was prepared by extraction of a concentrated aqueous solution of tetrabutylammonium bromide (52) and sodium p- to luenesulphinate (51) with methylene chlor ide. ( n - C ^ J ^ B r + N a S O z - Q - CH 3 ( ^ N + SC* CH 3 R ~ ~ X R — SO2—^-CH 3 + ( n - C ^ ^ t f X 52 Recent ly , a versat i le procedure for the preparat ion of phenyl su l fones has a lso been publ ished by Manesca l ch i et a l . 4 8 Th is method involved the alkylation of benzenesu lph inate an ion suppor ted on Amber lys t A - 2 6 , a macroret icu lar anion exchange resin conta in ing a quaternary ammonium group. The Amber lyst A -26 supported benzenesulphinate anion 52 was prepared by the exchange reaction of sodium benzenesulphinate (63) with the resin in chloride form £4.. The alkylat ion react ion was ach ieved by stirring 52. with alkyl hal ides in refluxing benzene . The phenyl sul fones 55 thus produced were isolated simply by filtering the resin and removing the solvent under reduced p ressure . The y ie lds with primary alkyl ha l ides were reproducibly above 9 0 % and the unpurified products showed very low levels of by-products by spectra l a n a l y s i s . 4 8 25 64 62 62 S02—R + CH2N+(CH3)3X" 65 = polymer Although each of these three methods d iscussed above could be applied to our synthesis, we favou red M a n e s c a l c h i ' s p rocedure in v iew of its super io r y ie lds and relat ively straightforward workup. Us ing this procedure, the phenyl sulfone ££. was prepared from the bromide 12. in 9 4 % yield. The 1 H N M R spectrum of 6J1 exhibited a multiplet at 8 7.52-7.92 for f ive aromat ic protons and a triplet at 8 3.08 for the a - m e t h y l e n e pro tons of the benzenesulfonyl group. In addition, the mass spectrum of £ £ showed the expected parent peak at m/e 368 . •OTHP + \" CH2N+(CH3>3 SO* ~0 " 12 62 66 Having obta ined the phenyl sul fone in high y ie ld , we set out to perform the alkylation reaction aga in . The sulfone ££. was first treated with two equivalents of n-butyllithium to give the a.a-dil i thio spec ies which was then stabl ized with a smal l amount of H M P A and al lowed to react with Y-butyrolactone. The react ion p roceeded smoothly and rapidly to afford the y-hydroxy ketone £ Z in 81% y ield. 26 ^ ^ ^ ^ ^ 2 n-BuLi, HMPA ft OTHP The IR spectrum of £7. showed absorptions at 3440 cm"1 and at 1720 cm"1, indicating the presence of a hydroxyl and a carbonyl group respectively. In addition, the diastereotopic methylene protons, a to the carbonyl group, gave rise to two sets of triplets at 8 2.64-2.78 and 8 2.98-3.10 in the 1 H NMR spectrum of 6_Z, Reductive desulfonation of £7 to the hydroxy ketone was smoothly accomplished in 7 9 % yield by means of aluminum amalgam in refluxing aqueous tetrahydrofuran. 4 9 The 1 H NMR spectrum of £2. supported the desulfonated structure of this product. OTHP Al-Hg THF-H2O OTHP The desulfonated product 6JL exhibited two spots by tic after purification. The less polar fraction was identified as the hemiketal 6JL which was in equilibrium with the y-hydroxy ketone £8_. This equilibrium was not observed in compound £Z, possibly due to the steric effect of the benzenesulfonyl group. •OTHP O 68 OTHP The presence of £9_ was not expected to interfere with our synthetic scheme since differences in reactivity between the primary and tertiary alcohols or the carbonyl functionality could be used to shift the equilibrium to the desired acyclic form 6_8_- However, reaction of ££, might proceed slowly due to the existence of the less reactive hemiketal ££. This was indeed found to be the case in subsequent steps. 27 With the carbon framework of the 14-membered lactone constructed, continuation of the synthesis next required the introduction of the geminal dimethyl group. The replacement of a ketone by two methyl groups is an attractive strategy and a number of methods for such a transformation have been reported. One of the most popular methods uses a three-step s e q u e n c e . 5 0 , 5 1 A Wittig olefination of a ketone produces an alkene which is then transformed into a cyclopropane v ia a S immons-Smith react ion. The cyc lopropane ring thus obtained can be hydrogenated to give a geminal d imethyl group. Money and c o - w o r k e r s 5 0 have reported this procedure in the synthes is of (+)-longiborneol (ZOJ. The ketone Z I was first treated with methy lenetr iphenylphosphorane to give the alkene 7-2. which was then subjected to the cyclopropanat ion react ion. Subsequent hydrogenat ion of the cyc lopropane ring in Z3. and reduction of the acetoxyl group furnished (+) - long iborneo l (70). 20. However , certain problems accompany the use of this procedure: (1) Wittig react ions are very sensit ive to the steric environment around the carbonyl group undergoing r e a c t i o n . 5 2 (2) T h e bas ic character of the yl ide reagent is often incompat ib le with easi ly eno l izab le k e t o n e s . 5 3 (3) Cyclopropanat ion reactions frequently give low yields due to the interference of other functional groups with the cyclopropanat ing r e a g e n t s . 5 4 28 In 1980 , Reetz et al. 5 ^ reported a different approach to the preparation of a geminal dimethyl group from a carbonyl function. This method involved the use of methyltitanium trichloride ( 7 4 ) which was prepared quantitatively by treatment of methyllithium or methylmagnesium chloride with titanium tetrachloride. MeLi (MeMgCl) + Ti04 — CH3TiCl3 24 Methyltitanium trichloride ( 7 4 ) is a non-basic reagent and reacts chemo- and stereoselectively with carbonyl compounds. Thus, the reaction of the titanium reagent with a ketone produces a tertiary alcohol 25. which can be readily converted into the tertiary chloride 26- The addition of another equivalent of 24 produces the geminal dimethyl compound 22-9 CH3TiCl3 f V -P H HC^ \~P CH3TiCl3 w R ^ ^ R ' R ^ ^ R ' R ^ ^ R 1 R ^ ^ R ' 25 26 22 One year later, Reetz et a l . 5 6 reported that dimethyltitaniumdichloride (22), from the reaction of titanium tetrachloride with two equivalents of methyllithium, exhaustively methylated ketones to form geminal dimethyl compounds 22. TiCL, + 2 MeLi M^TiCfe 71 o \y R A> R . + Me2Tia2 - R ^ R 1 21 22 Interestingly, another titanium reagent 29_ has also been reported capable of directly dimethylating a ketone. This reagent was first synthesized by Tebbe et al. in 1 9 7 8 5 7 and has become known as Tebbe's reagent. It was prepared by treatment of titanocene dichloride (8J2) with two equivalents of tr imethylaluminum (81). The reagent was soon made commercia l ly avai lable due to its versati le applications in organic synthesis. a a + 2 A1(CH3)3 A1(CH3)2 £ 1 22 One of the most useful propert ies of Tebbe 's reagent is its superiority in methylenating carbony l compounds . It c a n methylenate a ldehydes , ke tones, and the carbony l g roups of carboxyl ic acid d e r i v a t i v e s . 5 8 It does not appear to enol ize ketones as Wittig reagents somet imes do. Thus, the optically active ketone £2 can be converted to the corresponding methylene product 8 3 in 9 3 % yield without racemizat ion 5 8 In 1983, Grubbs et a\?° reported that the conversion of a ketone to a geminal dimethyl group cou ld , in some c a s e s , be achieved by using two equivalents of Tebbe 's reagent. These authors cited a single example - this being the dimethylat ion of cyc lohexanone 8JL- In this reaction, a smal l amount of methylenated product &5_ was also isolated 5 8 O S2 S2 O (95%) + (5%) In light of these results, it s e e m e d worthwhile to investigate the use of the titanium reagents in our synthesis . W e first chose to explore the possibil i ty of direct dimethylation of 30 our molecule using Tebbe 's reagent. Thus , the hydroxyl group in compound £ £ was first benzylated to give the keto compound £ £ in 7 5 % yield. •OTHP NaH PhCH2Br OTHP £6 When the keto compound £ £ was treated with 2.4 equivalents of Tebbe 's reagent using Grubbs ' p r o c e d u r e , 5 8 only the methylenated product £ 7 was isolated. Attempts to effect the dimethylation reaction were then carr ied out with varying equivalents of Tebbe 's reagents using different reaction condit ions. Unfortunately, in no case was any dimethylated product detected. However, the methylenation reaction did proceed rapidly in yields of greater than 80%. OTHP 22 PhCH20' OTHP £6 22 Although Tebbe 's reagent proved to be a powerful methylenating reagent in our studies, its high cost detracted from its use in a large sca le synthesis . Our attention next turned to the reagent descr ibed by Reetz and c o - w o r k e r s . 5 6 Titanium tetrachloride was treated with two equivalents of methyll i thium according to R e e t z ' s p r o c e d u r e . 5 6 The resulting titanium reagent was then al lowed to react with one-half equivalent of cyc lododecanone (££) which was used as a model. The reaction proceeded smoothly to give 1 -methy lcyc lododecanol (££) instead of the des i red 1,1-dimethylcyclododecane (90). Efforts to effect the direct dimethylation reaction under a variety of condit ions again proved to be fruit less. However , the dimethylated product 9J1 was obtained by convers ion of the tertiary alcohol £9. into the chloride 9J_ and subsequent treatment of £1 with the titanium reagent again. 31 £2 21 2Q The strong acid ic condit ions required for the convers ion of 8j9_ to 2 1 and its low yield presumably due to the compet ing dehydration reaction also made this route impractical in our synthesis. This method was, therefore, abandoned. The failure of direct dimethylation with titanium reagents led us to return to the more c lass ica l procedures of converting a ketone group into a geminal dimethyl group. The first step was a Wittig olefination react ion. Thus , the y-hydroxy ketone £ £ and its cycl ic form 6JL were a l lowed to react with three equivalents of methylenetr iphenylphosphorane. A s expected, the reaction p roceeded slowly due to the presence of the unreactive hemiketal 612.. However, a prolonged reaction time of 80 h gave the alkene 2 2 in 8 0 % yield. The absence of the carbonyl absorpt ion and appearance of an olef inic absorpt ion at 1644 c m " 1 in the IR spectrum of compound 22. indicated a successfu l Wittig reaction. The absorption at 8 4.75 in its 1 H N M R spectrum was ascr ibed to the terminal vinyl protons. Ph3P=CH2 •OTHP 32 At this stage, it was necessary to protect the hydroxyl group in compound 2£. An acetate protecting group was chosen , s ince it would survive the subsequent oxidation reaction. Treatment of the alcohol 22. with acet ic anhydride and pyridine in the presence of a catalytic amount of 4-dimethylaminopyridine (DMAP) gave the desi red ester 22 in 9 4 % yield. AC2O, C5H5N _ • O T H P » A c O ^ ^ ^ ^ ^ ^ ^ ^ D M A P II 22 22 The IR spectrum of 22 showed a carbonyl stretching band at 1743 c m " 1 . The 1 H N M R spect rum of 22. exhibited a sharp singlet at 8 2.08 attributed to the methyl protons of the acetate and a triplet at 8 4.08 ascr ibable to the methylene protons a to the acetoxyl group. In addit ion, the mass spectrum of 22 showed a molecular ion peak at m/e 354. The next step in the synthesis was to convert the alkene into a cyclopropyl ring. One of the first effective syntheses of a cyc lopropane unit from an alkene was performed in 1958 by S i m m o n s and S m i t h 5 9 who treated an olefin with methylene iodide in the presence of z inc-copper couple. These authors proposed that methylene iodide reacted with z inc-copper to form an intermediate iodomethylz inc iodide (24), the carbon atom of which was electrophil ic and thus attacked the double bond to give cyclopropane 26. through the transition state 9 5 . 5 9 c ' r Y« - 2 M 1 c v 7 x 1 f + C H . [ t ] . , „ C H , + j / \ \ i A / \ 1 24 25 26 In 1 9 6 8 , F u r u k a w a a n d c o - w o r k e r s 6 0 p u b l i s h e d a p a p e r c o n c e r n i n g the cyclopropanation of a lkenes using diethylzinc and methylene iodide. These authors showed that cyc lopropanes could be obtained more easi ly with this new cyclopropanat ing reagent. S ince 33 then, the Furukawa modif ication of the S immons-Smi th cyclopropanat ion procedure has been used in organic synthesis with considerable s u c c e s s . 6 1 When the olefin 23. was heated with diethylzinc-methylene iodide in toluene at 60 °C , the corresponding cyc lopropane compound 2 Z was produced, but in low yield (about 25-30%). Capi l lary glc analys is of the crude reaction mixture indicated the presence of more polar by-products. Al though these by-products were not isolated and character ized, it seemed likely that they were formed by react ion of the tet rahydropyranyl ether in c o m p o u n d 2 3 . with the cyclopropanat ing reagents. OTHP Clear ly , the low yield of this reaction warranted alternative plans for the preparation of c o m p o u n d 2 Z - Speci f ical ly , it was felt that the alcohol 23. obtained by the c leavage of the tetrahydropyranyl ether in 2 2 might be a better substrate for the cyclopropanat ion react ion. Treatment of 23. with pyridium p-toluenesulfonate afforded the hydroxy a lkene 23. in 9 0 % y ie ld . When the hydroxy olefin 23. was subjected to the modif ied S immons-Smi th react ion, the corresponding cyclopropyl compound 22. was produced in 4 1 % yield. This yield was higher than the one obtained earlier, but still left much to be des i red. Et2Zn, CH2I2 ^ ^ / V V V ^ ^ S A O H A c O - ^ ^ V ^ ^ ^ ^ ^ O H 34 At this point, we were encouraged by the improved yield and dec ided to investigate the reac t ion by further modi fy ing the subst ra te by protect ing the a l coho l 2JL as a ter t -buty ld imethyls i ly l (TBDMS) ether. Si ly l ether derivatives of a lcohols have a number of useful propert ies. The volatility of tr imethylsi ly l ethers m a k e s them sui table for separa t ion and structure e luc idat ion by a combinat ion of gas chromatography and mass spec t r ome t r y . 6 2 On the other hand, their lability to mild condit ions limits their use as protecting g r o u p s . 6 3 However , the stability of T B D M S ethers to a wide range of react ion condi t ions makes them particularly effective protecting groups for the hydroxyl f u n c t i o n . 6 4 Th is stability results from the fact that most of their react ions p roceed by nucleophi l ic attack at s i l i con, which is , therefore, sensi t ive to steric h indrance. React ion of the a lcohol 9JL with tr iethylamine (TEA) , i e i l -bu ty ld imethy ls i l y l ch lo r ide and a catalytic amount of 4-dimethylaminopyridine gave the silyl ether 100 in 9 5 % yield. AcO' TBDMS-Cl t TEA. DMAP AcO" •OTBDMS 2S lffi The 1 H N M R spectrum of compound 100 exhibited a six-proton singlet at 8 0.06 due to the si lyl methyl protons, whi le a second singlet ascr ibable to the tert-butyl methyl protons appeared at S 0.90. To our satisfaction, the cyclopropanation reaction of the silyl ether a lkene 100 proceeded c leanly, under the condit ions used earl ier, to give the cyclopropyl compound 101 in 8 8 % yie ld. The IR spectrum of 101 showed a cyclopropyl-hydrogen stretching absorption at 3065 c m " 1 . In addit ion, the cyclopropyl protons of 1Q1 gave rise to a doublet at 8 0.22 in the 1 H N M R spectrum. OTBDMS Et2Zn •OTBDMS C H 2 l 2 1QQ 101 35 Hydrogenolysis of the least hindered cyclopropyl bond b J ) of compound 101 was accomplished by treatment of a solution of 101 in glacial acetic acid with a catalytic amount of platinum oxide under hydrogen. The geminal dimethyl silyl ether 102 thus produced (in 7 9 % yield) was hydrolyzed to the corresponding alcohol 1Q2, in 8 7 % yield by treatment of 1Q2. with pyridinium p-toluenesulfonate. OTBDMS 103. The IR spectrum of 103 showed a hydroxyl absorption at 3560-3204 cm" 1. The 1 H NMR spectrum exhibited a sharp singlet at 5 0.84 which was attributed to the six protons of the geminal dimethyl group in the molecule. The next two steps in the synthesis were rather straightforward. Oxidation 6 6 of alcohol 103 gave the carboxylic acid 104 in 7 7 % yield. The IR spectrum of 104 showed the acid absorptions at 3480-3014 cm* 1 and 1712 cm"1. In addition, the 1 H NMR spectrum of 104 exhibited a triplet at 8 2.36 which was ascribable to the a-methylene protons of the carboxylic acid function. The acetate protecting group in compound 104 was hydrolyzed with potassium carbonate in methanol. The ©-hydroxy acid 105 thus produced (in 9 8 % yield) was highly pure and was carried directly to the lactonization reaction. KM 105 With ample quantit ies of the hydroxy acid 105 in hand, we were ready to perform the final cycl izat ion step in our synthesis. The literature records an extensive list of methods for effecting the lactonizat ions of hydroxy ac ids , and many of these have been d i scussed in the Introduction. Among the several existing methods, Schmidt and Dietsche's procedure has proven r e l i a b l e . 2 7 Our lactonization reaction was therefore carr ied out using this procedure. T rea tmen t of 1 -pheny l -2 - te t razo l ine -5 - th ione (106 ) with one equivalent of tert-buty l isocyanide (107) in toluene at room temperature gave compound 1 0 8 a and 108b which upon addition of 0.8 equivalent of the o-hydroxy acid 105 afforded the 14-membered lactone 42. in 6 6 % yie ld. Ph N — N P h — N — N P h — N — N + + \ N H H — O N 106 101 108a 108b 42 The IR spectrum of macrolide 42 showed a lactone carbonyl absorption at 1718 cm" 1. The 1 H NMR spectrum of 42 was readily assignable. The six methyl protons gave rise to a sharp 37 singlet at 8 0.84. The a-methylene protons of the lactone oxygen and the a-methy lene protons of the carbonyl group resulted in two sets of multiplets at 8 4.18 and 8 2.40 respect ively. In addit ion, the mass spectrum exhibited a molecular ion peak at m/e 240. Thus , the synthesis of 10,10-d imethy l t r idecanol ide (42) was comple ted . 38 2.2 Hydro lys is of 10.10-Dimethyl t r idecanol ide (42). Tr idecano l ide (35) and n-Octvl Pen tanoate M091 To investigate the conformational behavior of the macrol ide synthesized in this project, the h y d r o l y s i s of 1 0 , 1 0 - d i m e t h y l t r i d e c a n o l i d e ( 4 2 ) . t r i decano l ide (35 ) and n-octyl pentanoate (109) was per formed. From simple conformational ana lyses , macrol ide 42 was expected to exist predominantly in conformation 4 2 a . O n the other hand, tr idecanolide (35) was known to exist as a mixture of conformers 3 5 a . 35b and 35£ in nearly equal a m o u n t s . 4 4 35s 25J2 25£ Examinat ion of the environments surrounding the lactone functionality in each of the above conformat ions revea led that 4 2 a and 3 5 a directed the lactone carbony l towards the sterical ly more hindered interior of the ring. Cons idera t ion of the steric interact ions of the tetrahedral intermediates in the base ca ta lyzed hydrolysis reaction led to an initial prediction that hydrolysis of conformers 4 2 a and 3 5 a would be s lower than that of conformer 3 5 b or 3 5 c . S ince macrol ide 35. exists as a mixture of 3 5 a . 3 5 b and 3 5 c . it shou ld hydro lyze faster than macro l ide 42. which ex is ts predominantly in conformation 4 2 a . The intermediates involved in the hydrolysis of 422. and 35a are 110 and H I where the oxyanion is forced into a sterically crowded environment. The intermediates encountered in the 3 9 hydrolysis of 3 5 b and 3 5 c (112 and 113 respectively) are less steric h indered. Al l of these intermediates are a s s u m e d to maintain the [3434] ring conformations of the starting materials. HO HO 112 In addition to studying the hydrolysis of macrol ides 42. and 35_, the hydrolysis of a 13-carbon ester was also conducted. It was expected that the ester would hydrolyze faster than either of the macrol ides. The synthesis of macrolide 42. has been detailed in Sect ion 2.1. The other two compounds used for hydrolysis studies were prepared as fol lows: Ester 109 w a s prepared by treatment of 1-octanol (114) and valer ic ac id (115) with d icyc lohexy lcarbod i imide ( D C C ) in the p resence of 4-dimethylaminopyr id ine ( D M A P ) . The esterification reaction proceeded smoothly to give the desired product 10j9_ in 9 4 % yield. O DMAP O 115 114 102 Macrol ide 25. was synthesized using a modified Baeyer Vill iger r e a c t i o n . 6 7 Treatment of cyc lo t r i decanone (116) with peroxytr i f luoroacetic ac id gave the des i red product 2 5 in 6 7 % y i e l d . 40 116 25 Hydrolysis of macrolides 12. and 25., and ester 109 with potassium carbonate in methanol was first carried out at room temperature. The relative rates of hydrolysis were determined by monitoring a reaction mixture which contained the hydrocarbon dodecane as an internal standard and following the disappearance of the ester and the macrolides by gas-liquid chromatography (glc). The data obtained from this reaction are plotted in Figure 6 as relative intensities of the hydrolysis compounds to the internal standard versus reaction time. 0 29 50 73 Time (min) Figure 6. Plot of the rates of the hydrolysis of 12,25. and 109 at room temperature. 41 Figure 6 shows that ester 1 0 9 hydro lyzed faster than the macro l ides under these condit ions. This result could be explained in terms of the steric hindrance of the intermediates involved in the react ion. At tack on the macro l ides by the hydroxyl anion would force the oxyanion into a sterically crowded environment regardless of their conformations. O n the other hand, attack on a long-chain ester by the hydroxyl anion would be free of any steric interaction. One would therefore anticipate a much more rapid hydrolysis of the ester than the macrol ides. However , the observat ion that both macro l ides hydro lyzed at the s a m e rate w a s surpr is ing. If macrol ide 42 existed in a single [3434] conformation 4 2 a and the hydrolysis of these macro l ides p r o c e e d e d through intermediates maintaining the [3434] conformat ions, macrol ide 42 should hydrolyze more slowly than macrol ide 25- Hydrolysis of macrol ides 41 and 25., and ester 109 at low temperatures (0 °C and -20 °C) also gave similar results to those shown in Figure 6. To interpret the hydrolysis results and understand the conformat ional preferences of macro l ide 42. from a theoretical point of v iew, we next undertook the molecular mechan ics (MM2) c a l c u l a t i o n s 3 8 ' 3 9 of conformations for macrol ide 42.-4 2 2.3 Molecu lar Mechan ics Calcu la t ions of Conformat ions for 10.10-Dimethvrtr idecanol ide im One of the goals of this project was to construct and investigate a 14-membered macrolide which should exist predominantly in one conformation. The structural requirements we used to predict a single conformation for macrol ide 4£ were the preference for a planar s-trans lactone l inkage and the occupat ion of a geminal dimethyl group at the corner posit ion in the [3434] conformation of the 14-membered lactone. Our theoretical approach to studying the conformational preferences of this system was to ca lcu la te the ster ic energ ies of all likely conformat ions of macrol ide 4 2 . and from these energ ies to determine their approximate Bo l tzmann distr ibut ions. The calculat ion of these conformational energies was performed using the molecular mechan ics (MM2) p r o g r a m . 3 8 , 3 9 Molecular mechanics calculat ions have been gaining popularity in the past few years as r e s e a r c h e r s try to interpret their resu l ts a n d formula te syn the t i c p l ans b a s e d on conformational ana lyses . The M M 2 program was introduced by Al l inger and c o - w o r k e r s 3 8 , 3 9 as an alternative to the compl icated, sh. initio molecular orbital (MO) methods of calculat ing molecular energ ies . The energ ies ca lcu la ted by the M M 2 program are based on c lass ica l mechanics in which the equations used to calculate the energies are parameterized to best fit the experimental data . Mo lecu les are represented as though constructed from balls and spr ings with a ser ies of potential energy functions to express the "steric" energy of a molecule. The "steric" energy of a molecule is the sum of five different energies. The first energy term is assoc ia ted with bond stretching which the M M 2 program treats as a modif ied Hooke 's law equat ion. T h e remaining four terms include angle bending, torsional strain, dipole and Van der Waa l s interactions. To calculate the minimum energy conformation of a molecule, the M M 2 program employs the s teepest -descent method. A likely conformation of a molecule is first constructed using molecular models and its approximate co-ordinates are determined. These are provided as an 43 input file to the M M 2 computer program and the steric energy is calculated from this given set of co-ordinates. The computer then moves one atom to a new set of co-ordinates and recalculates the energy. If the atom movement results in a lower energy, then the atom is further moved in the same direction until the energy difference is less than or equal to a preset value. Although this process has been descr ibed using a single atom, in fact, every atom within the molecule is simultaneously subjected to this movement-calculat ion sequence until the energy of the system is min imized. T h e result ing ster ic energy of a certain conformat ion is an energy relat ive to a hypothetical, strain-free, reference system. The difference in energy between conformations of a molecule is g iven by direct compar ison of the ca lcu la ted steric energ ies . However , if a compar ison between different molecules is required, steric energies should not be used ; an alternative for this type of compar ison is to use the heats of formation, which can a lso be calculated by the M M 2 program. The conformat ions of macrol ide 4£ to be invest igated arose from the conformat ional a n a l y s e s of s imp le 14 -membered lac tones . A s desc r i bed p rev ious ly , the low-energy conformat ions for t r idecanol ide (25J are 3 5 a . 35b and 3 5 c . 4 4 These three conformers were taken as the bas ic f ramework to generate conformers 4 2 a . 4 2 b and 4 2 c for macrol ide 42. Their relative steric energ ies were determined by M M 2 calculat ions. In addit ion, the relative steric energy of conformer 42d which had a geminal dimethyl group at the corner position but a lactone linkage at a position different from that of conformer 4 2 a was also calculated. 44 Figure 7 shows the computer plots of the side and top views of conformations 4 2 a . 42b . 4 2 c and 4 2 d . and their relative steric energies. side view 4 2 a (0.0 kcal/mole) top view side view 42Ja (5.8 kcal/mole) top view side view 4£c_ (4.2 kcal/mole) side view 424 (7.8 kcal/mole) top view top view Figure 7. Computer plots of the [3434] conformations for macrol ide 42. 46 M M 2 calculations showed that the energy of conformer 42a was the lowest. Accordingly, the steric interactions in this conformation should be the least. This was in agreement with our s imple conformat ional ana lys is . The non-corner gemina l dimethyl groups in 4 2 b and 4 2 c introduced steric interaction in the rings and therefore ra ised their steric energ ies. Conformer 42b had a higher energy than conformer 42c.. This was due to the fact that the dimethyl group in 42b was at a sterically more hindered posit ion. A s expected, conformer 4 2 d . containing the s-cis lactone l inkage, was destabi l ized and resulted in a relatively high energy state. Al though the d iagrams shown in Figure 7 demonst ra ted the [3434] character of the con fo rmat ions a n d their s t ruc tures , it w a s diff icult to recogn ize the symmet ry of the conformat ions or to easi ly compare one conformat ion with another. It was a lso difficult to determine if any deviation from the ideal [3434] conformation had occurred. Fortunately, these difficulties could be solved through the use of polar m a p s . 6 8 A polar map is a circular graph which plotting the sign and magnitude of the internal torsional angles of a ring vs . the bonds along which they are found. The concentr ic circles of a polar map represent va lues of the torsional angle in ± 60° increments and the straight l ines which intersect the circ les are the bonds (numbers 1-14) where the torsional angle is formed (Figure 8). Accordingly, a 14-membered ring produces 14 data points on a polar map. These points, when connected, generate a "star pattern" representing the conformation. 14 1 12 10 8 3 5 4 8 7 Figure 8. The polar map conventions. 4 7 The value of polar maps is that the complete set of torsional angles formed by the ring atoms will a lways uniquely define the conformation. The torsional angles may be obtained from M M 2 ca lcu la t ions , from X- ray data or from an inspect ion of a molecu lar mode l . Recent developments in the determination of the signs of torsional angles have also enhanced the general use of polar m a p s . 3 5 , 6 9 The polar map of the [3434] conformation of cyclotetradecane is shown in Figure 9. The polar map clearly i l lustrates the C 2 rotation axis through bonds 4 and 11 . In addit ion, the patterns of the polar map can be used to quickly indicate the presence and location of corner a toms within a conformat ion by recogniz ing the corner 's character ist ic ant i -gauche-gauche-anti sequence of torsional a n g l e s . 3 5 The position of the corner atoms on the conformations and polar maps are marked with an asterisk. C2 axis of macrolide Figure 9. The [3434] conformation of cyclotetradecane and the polar map of its torsional angles. With the aid of polar maps, conformations can be unambiguously identified and examined. Th is is faci l i tated if the ideal conformat ion, for examp le , the ideal [3434] conformat ion, is simultaneously plotted on the same map with the one under examinat ion. The polar maps of the four conformat ions determined by M M 2 calculat ions (Figure 7) are g iven in Figure 10 which also shows the ideal [3434] conformation in broken l ines. 48 14 1 H Figure 10. Superposi t ion of the polar maps of the ideal [3434] conformation (broken lines) and the [3434] conformations of macrolide 4£. The polar maps in Figure 10 show that conformer 4 2 a c lose ly approximated the ideal [3434] conformat ion. This was in agreement with our prediction that a geminal dimethyl group at the corner posit ion would not introduce severe t ransannular interact ions into the ring and would therefore not signif icantly distort the [3434] ring framework. 49 O n the other hand , the polar maps of conformers 4 2 b and 4 2 c showed a smal l but signif icant deviat ion from the ideal [3434] conformat ion at bonds 10 and 9, respect ively. T h e s e are the bonds that contain the non-corner dimethyl subst i tuted ca rbon . The steric hindrance introduced by the dimethyl group has forced the ring to change its conformation. It is interesting to note that the strain appears to be concentrated at a few bonds rather than distr ibutes over many bonds and the main portion of the ring still adheres to the [3434] f ramework. Simi lar ly , the main portion of the ring is [3434] like for conformer 4 2 d with deviation occurring at bond 3 which contains the s-cis lactone group. Ear ly in our project, we were only c o n c e r n e d with the [3434] conformat ions for macro l ide 42- However , recent work in our laboratory revealed severa l other low-energy con format ions for cer ta in m a c r o l i d e s . 3 5 , 3 6 Among them were two non-d iamond lattice conformations descr ibed by D a l e . 3 2 In his pioneering work on conformations of cycl ic a lkanes, Dale considered only those ring conformat ions which were super imposab le on a d iamond latt ice. However , ca lcu la t ions on cyc lotet radecane later revealed two low-energy conformat ions which were not d iamond lattice super imposab le . T h e s e were des ignated the [3344] (1.1 kcal /mole higher than the [3434] base) and [3335] (2.2 kcal /mole) conformat ions (Figure 1 1 ) . 3 2 T h e s e conformat ions were found to be lower in energy than every d iamond lattice conformation with the except ion of the [3434] arrangement (base va lue of 0.0 kcal /mole) . The posit ion of the corner atoms of the conformations in Figure 11 is marked with an asterisk. 50 Figure 11. The [3344] and [3335] conformations of cyclotetradecane and their polar maps. W e were prompted to investigate these two conformat ions for macrol ide 42 by M M 2 calculat ions and we restricted our calculat ions to those conformat ions containing an s-trans lactone l inkage and a geminal dimethyl group at the corner posi t ion. S ince the [3344] and [3335] ar rangements were less symmetr ic than the [3434], two [3344] conformat ions 4 2 e and 421, and three [3335] conformat ions 42a. 4 2 h and 42i were poss ib le for macrol ide 42. Their calculated steric energies relative to 4 2 a . computer plots and polar maps are given in F igure 12. top view polar map 4 2 e (1.2 kcal /mole) top view polar map 42f (1.3 kca l /mole) top view polar map 42 i (3.5 kcal /mole) Figure 12. Computer plots and polar maps of the [3344] and [3335] conformations for macrolide 42 (relative steric energy in kcal /mole) . 52 With the ster ic energ ies of all l ikely conformat ions ca lcu la ted , the proport ions of dif ferent con format ions for macro l ide 42. cou ld now be es t imated using the Bo l tzmann distribution equat ion. For two isomers A and B in equil ibrium, the equil ibrium constant K (the ratio of the number of A molecules, N ^ , to the number of B molecules, Ng ) is given by equation 1. and E g are the energies of two isomers, R is the gas constant and T is the absolute tempera tu re . K = N A N B exp ( E A • EB) RT (D Using the calculated relative steric energies of the above conformat ions, an estimate of their distribution can be obta ined. For example , cons ider conformers 4 2 a and 4 2 c at 25 ° C , equation 1 becomes N42a N42C e x p I - ( 0.0 - 4.2 ) x 10 ca l /mo le 1.986 cal /K mole x 298 K = 1.2 x 10 Conformat ion 42c is not significantly populated at room temperature and for pract ical purposes it can be ignored. Similarly, conformations 4 2 b . 4 2 d . 4 2 g . 42h and 42i need not be cons ide red as poss ib le con fo rmers for macro l ide 42- Our attention cou ld therefore be concentrated on one [3434] conformation 4 2 a and two [3344] conformations 42£ and 42f- The Bo l tzmann distribution of these conformat ions is ; 4 2 a : 4 2 e : 42f = 80 : 11 : 9. Accordingly, macrol ide 42a should exist as a mixture of these three conformations. 42£ 421 53 Among these three conformations 42fi. is identical with 42a in the region of the molecule containing the lactone functionality. The chemistry of 42e should be similar to that of 4 2 a and it would be sufficient to descr ibe macrol ide 42 in terms of conformations 4 2 a and 421-The lactone groups of 4 2 a and 421 are in different environments. Conformer 42a has the lactone carbonyl group directed toward a sterically more hindered ring interior than conformer 421. The tetrahedra! intermediate from hydrolysis of 421 should be sterical ly less hindered than that of 42a and the hydrolysis of 421 should therefore be faster than 4 2 a . The results of the hydrolysis studies could now be interpreted. A possib le explanation for our observat ion of the similar rates of the hydrolysis of the two macro l ides is that the [3434] ring conformat ions were not maintained in the hydrolysis intermediates. The steric hindrance introduced by the hydrolysis intermediates may have led to a change of the [3434] conformation such that the difference of the lactone carbonyl environment in the ground state conformations was not reflected in the reaction. Another exp lanat ion for the hydro lys is s tudies is that macro l ide 42. exists as an equi l ibr ium mixture of conformat ions 4 2 a / e and 42f . The rates of conformat ional change between these conformers are much faster than that of the hydrolysis reaction and conformer 421 is expected to hydrolyze more easily than conformer 4 2 a . According to the Curt in-Hammett p r i n c i p l e , 7 0 the ratio of the conformat ions of macrol ide 42 would not be ref lected in the hydrolysis p rocess . From Le Chatel ier 's pr inciple, the select ive hydrolysis of conformer 421 would shift the equi l ibr ium from conformer 4 2 a to 42f . Therefore the minor conformer 421 was the one that control led the hydrolysis p rocess and not the predominant conformer 4 2 a . Consequen t l y , the hydrolys is of macro l ide 42 was observed to be faster than expected in compar ison to macrol ide 25, 54 2.4 Conc lus ion The synthesis of 10,10-dimethyltr idecanol ide (42) was accompl ished v ia a fifteen-step sequence in 9% overal l y ie ld. During the synthesis, it was found that direct Gr ignard attack of Y-butyrolactone fai led to produce the cor responding y-hydroxy ketone. However , this was so lved by the use of sul fone chemistry. Convers ion of the keto functionality into a geminal dimethyl group in our molecule was ach ieved by a three-step sequence (Wittig olef inat ion, S immons-Smi th reaction and hydrogenation). M M 2 studies conf i rmed the conformat ional ana lys is that macro l ide 1 2 shou ld exist predominant ly in the [3434] conformat ion 4 2 a which p o s s e s s e d a p lanar s-trans lactone l inkage and the geminal dimethyl group at a corner posit ion. However, more importantly, M M 2 stud ies a lso revea led the ex is tence of a [3344] conformat ion 1 2 1 which evident ly w a s controll ing the rate of hydrolysis of macrol ide 1 2 . In the beginning, we restricted the conformational analys is to the [3434] conformation. But during this project, it was found that the [3434] conformation was not sufficient to explain the hydrolysis results. B a s e cata lysed hydrolysis of macrol ides 1 2 and 25. appears complex for detai led conformational analys is. It can be expected that further elaboration of the lactone ring, for example, introduction of addit ional geminal dimethyl groups, should favour the existence of only one conformation and therefore simplify conformational analys is . Obv ious ly , there is much work that can and should be done on construct ions of 14-membered lactones with def ined conformation(s) and on chemical reactions of such macrol ides. Ult imately, the conformational behavior of 14-membered lactones could culminate in the total synthesis of the macrol ide antibiotics and the application of conformational control in synthetic chemistry as a whole. 55 CHAPTER THREE Experimental General Solvents, reagents and equipment setup. Solvents were dried as follows: diethyl ether (ether), benzene, toluene and tetrahydrofuran (THF) were distilled into a collecting reservoir by heating at reflux over sodium benzophenone ketyl radical under a dry nitrogen (N2) atmosphere. Methylene chloride (CH 2CI 2) and triethylamine (TEA) were distilled from calcium hydride and methanol (MeOH) from magnesium methoxide. Using an oven dried syringe, an anhydrous solvent was removed through a stopcock fitted on the reservoir. Unless otherwise specified, all reagents were supplied by the Aldrich Chemical Company and used without further purification. Bottles of n-butyllithium (in hexanes) and diethylzinc (in toluene) were equipped with used glc gas port septa wedged between the Sure/Seal crown and the twist cap. In this way, they were stored with no significant changes in molarity for up to nine months. a-Butyllithium was standardized by titration against 2,2-diphenylacetic acid in THF at room temperature to the faintest appearance of a yellow color. Acetic anhydride and 3,4-dihydro-2H-pyran were distilled over calcium hydride. 7-Butyrolactone, pyridine, and hexamethylphosphoramide (HMPA) were distilled under reduced pressure from calcium hydride and stored over molecular sieves (3 °A). Methylene iodide was distilled under reduced pressure and stored over tin metal. Nitrogen was supplied by Union Carbide and prior to use was passed through two columns of indicating Drierite (CaS0 4 impregnated with CoCI2). Syringes and needles were oven-dried at 120 °C for a minimum of 3-4 hours and stored in a desiccator. Unless stated otherwise, all reactions were carried out under an atmosphere of dry nitrogen. The glassware (including the Teflon coated magnetic stirring bar) was assembled and connected to the vacuum pump and flame-dried. After the glassware had cooled, dry nitrogen was introduced to the system. 56 C o l d temperatures were mainta ined us ing either an ice/water bath (0 °C) or an acetone/dry ice bath (-78 °C) . The concentrat ion or evaporation of solvents under vacuum refers to the use of a Buchi rotary evaporator. Petro leum ether refers to the fraction boil ing between 30-60 °C . Reaction monitoring. A l l react ions were monitored by thin layer chromatography (tic) and/or gas- l iqu id chromatography (glc). Ana ly t ica l tic w a s per formed on a luminum backed , precoated s i l ica ( S i 0 2 ) gel plates (E . Merck, type 5554). The plates were v isual ized by ultraviolet f luorescence or by heating the plates after spraying them with 3 M sulfuric ac id . Analyt ical glc was performed on a Hewlett Packard model 5880A gas chromatography using a 12 m x 0.2 mm capi l lary Carbowax column or a 15 m x 0.2 mm capi l lary DB-210 co lumn. In both c a s e s , f lame ionization detect ion was used with a helium carrier g a s . Al l samples were made up in ether and injection volumes were 2 u l . Product purification. Un less otherwise stated, all react ion products were purif ied by f lash chromatography using 230-400 mesh A S T M si l ica gel suppl ied by E. Merck C o . In most c a s e s , the si l ica gel was reclaimed after column chromatography. This involved discarding the upper 2-4 cm of si l ica gel in the column and flushing the remaining si l ica gel with methanol until c lean. A hose connected to a water aspirator was attached to the column spigot and the si l ica gel sucked to dryness (powder dry). The si l ica gel was subsequently regenerated by oven heating for 6-8 hours at 120 °C. This recycl ing procedure could be repeated 3-4 t imes before the si l ica gel turned a yellow color whereupon it was d iscarded. This procedure greatly extended the general usage of si l ica gel column chromatography. Product characterization. Infrared (IR) spec t ra were recorded on a B o m e m Miche lson 100 FT spectrophotometer. Samp les were d issolved in chloroform and the spectrum was taken and subsequent ly subtracted from a spectrum of pure chloroform. In some c a s e s , a neat sample was directly employed. Absorpt ion posit ions are given in c m " 1 and abbreviat ions used in quoting the IR bands are: st=strong, m=medium, w=weak and br=broad. 57 Low resolution mass spectra were determined on a Var ian M A T model C H 4 B or a Kratos-AEI model M S 50 spectrometer. The parent peak as well as major ion fragmentat ions are reported as percentages of the base peak. Exact masses were obtained by high resolution mass spectroscopy using a Kratos-AEI model M S 50 spectrometer. Al l instruments were operated at 70 ev. Nuc lear magnet ic resonance (NMR) spect ra were taken in deuterochloroform ( C D C I 3 ) solut ion with s ignal posi t ions given in parts per mill ion (ppm) from the internal s tandard of tetramethylsi lane (0.00 ppm) on the 8 sca le . Proton nuclear magnet ic resonance ( 1 H N M R ) spect ra were recorded at 300 M H z on a Var ian X L - 3 0 0 or at 400 M H z on a Bruker WH-400 spec t rometer and are reported in the form: chem ica l shift (number of protons, s igna l multiplicities). The abbreviat ions used in quoting the data are: s=singlet, d=doublet, t=triplet, q=quartet and m=multiplet. 58 3.1 Prepara t ion of 10.10-Dimethv l t r idecano l ide (42) 9 - B r o m o - 1 - n o n a n o l (48) A suspens ion of 1,9-nonanedio l (35.1 g , 219 mmol) in 50 mL of 4 8 % HBr w a s prepared in an 1 L liquid-liquid continuous extractor. The suspension was heated to 90 °C and was extracted with 400 mL of heptane at this temperature for 72 h. The extract was coo led , washed twice with saturated aqueous sodium bicarbonate and once with brine. The organic layer was dried over M g S 0 4 and concentrated under vacuum. The crude oil was purified by column chromatography using a mixture of petroleum ether : ethyl acetate (3 : 1) as eluent to give the monobrominated product 48_ (40.8 g , 84%) as color less crystals. 1 H NMR (300 M H z , C D C I g ) 8: 3.66 (2H, t), 3.42 (2H, t), 1.92-1.28 (15H, m); IR (CHCI3, c m ' 1 ) : 3337 (free O H , st), 1260 (C-Br , m); MS (m/e, relative intensity): 206 ( 8 1 B r : M + - H 2 0 , 9), 204 ( 7 9 B r : M + - H 2 0 , 9), 178 (9), 176 (9), 164 (13), 162 (13), 150 (19), 148 (19), 137 (32), 135 (34), 97 (58), 83 (44), 82 (21), 81 (14), 70 (13), 69 (85), 68 (20), 67 (17), 5 7 (14), 56 (17), 55 (100), 54 (12) . 59 1-Bromo-9 - rne t rahvdro -2H-py ran -2 -ynoxy1 - nonane (49) 4 2 Pyridinium p-toluenesulfonate (PPTs) (1.50 g , 5.97 mmol) was added into 120 mL of dry C H 2 C I 2 at room temperature under N 2 . 9-Bromo-1 -nonanol (13.8 g , 61.9 mmol) was d issolved in 20 mL of dry C H 2 C I 2 and injected. The mixture was cooled to 0 °C with an ice bath and freshly dist i l led d ihydropyran (8.17 mL , 89.6 mmol) w a s added d ropwise . After the addi t ion, the coo l ing bath was removed and the react ion mixture w a s st i r red at room temperature for 4 h. The mixture was diluted with ether, washed twice with co ld 1N HCI and once with brine. The organic phase was dried over M g S 0 4 and concentrated under vacuum. The crude yellow oil was chromatographed on a si l ica gel column using a mixture of petroleum ether : ethyl acetate (3 : 1) to give the desired product 4 J . (17.3 g , 91%) as a color less oi l . 1H NMR (300 M H z , C D C I g ) 8: 4.60 (1H, t), 3 .94-3.36 (6H, m), 1.92-1.26 (20H, m ) ; IR ( c m " 1 ) : 1260 (C-Br , m), 1128 ( C - O , m); MS (m/e, relat ive intensity): 307 ( 8 1 B r : M + - 1, 7), 305 ( 7 9 B r : M + - 1, 9), 85 (100), 84 (10), 83 (13), 69 (22), 67 (12), 57 (21), 56 (28), 55 (34), 4 3 (18), 41 (38), 40 (13), 32 (27), 29 (20), 28 (27); E x a c t m a s s ca lc . for C 1 4 H 2 6 8 1 B r 0 2 : 307 .1097 ; F o u n d : 307 .1107 ; ca l c . for C 1 4 H 2 6 7 9 B r 0 2 : 305.1117; Found : 305.1115. 60 1 - P h e n v l s u l f o n v l - 9 - f ( t e t r a h v d r o - 2 H - p v r a n - 2 - y h o x y 1 - n o n a n e (661 OTHP A 0.1 M aqueous solution of sodium benzenesul f inate was slowly percolated through a co lumn filled with Amberlyst A -26 (Rohm and Haas) in the chloride form until a negative test for chlor ide ion in the eluate was obta ined. T h e resin was success ive ly washed with water, a c e t o n e , ether and dr ied under v a c u u m at 50 °C for 5 h to g ive Amber lys t A -26 in benzenesul f inate form. This resin (47.5 g , 158 meq.) was added into a solution of bromide 4 3 (40.5 g , 132 mmol) in 300 mL of dry benzene . The mixture was vigorously stirred at reflux for 48 h. The resin was filtered and washed with C H 2 C I 2 . The filtrate was concentrated under vacuum and the crude oil w a s chromatographed on a s i l ica gel co lumn using a mixture of petroleum ether : ethyl acetate (4 : i ) as eluent to recover bromide 4 £ (3.80 g , 8.4%) and to give sulfone 6JL (41.3 g, 94%) as a light yellow oi l . 1 H N M R (300 M H z , CDCI3) 8 : 7.92 (2H, d), 7.70-7.52 (3H, m), 4 .58 (1H, t), 3 .92-3 .04 (6H , m), 1.90-1.20 (20H , m) ; IR ( c m - 1 ) : 3062 (phenyl C - H , w), 1310 (S=0 , st); M S (m/e, re lat ive in tensi ty) : 368 ( M + , 2) , 367 ( M + - 1, 7), 339 (13), 285 (40), 284 (13), 2 8 3 (43), 268 (12), 2 6 7 (51), 255 (23), 254(11) , 251 (20), 239 (12), 143 (56) , 125 (37), 124 (14), 101 (38), 100 (10), 85 (100) , 84 (19), 83 (24), 69 (47), 67 (13), 57 (17), 56 (10), 55 (36). Exact mass ca lc . for C 2 0 H 3 2 S O 4 : 368 .2023; Found : 368.2017. 61 1 - H v d r o x v - 5 - p h e n v l s u l f o n v l - 1 3 - r r t e t r a h y d r o - 2 H - p v r a n - 2 - v l \ o x v 1 - t r i d e c a n - 4 - o n e um P h S 0 2 0 62 a-Butyl l i thium (6.8 mL, 11 mmol) was injected at 0 °C into a well-st irred solution of sul fone 6j6_ (2.0 g , 5.4 mmol) in 60 mL of dry T H F under N 2 . After stirring for 30 minutes the react ion was c o o l e d to -78 °C and freshly dist i l led H M P A (0.6 m L , 3 mmol) and y-butyrolactone (0.44 mL, 5.4 mmol) were added. The mixture was stirred for 3 h and al lowed to warm to room temperature. It was then quenched with aqueous N H 4 C I and extracted with ethyl acetate (3 x 30 mL). The organic layers were combined, washed twice with saturated aqueous cupric sulfate and once with brine, dried over M g S 0 4 and evaporated under vacuum. The crude product was chromatographed on a si l ica gel column using a mixture of petroleum ether : ethyl acetate (1 : 1) as eluent to give the y-hydroxy keto compound £ Z (1.64 g , 81%) as a yel low o i l . 1 H NMR (300 M H z , C D C I g ) 8 : 7.80 (2H, d), 7.74-7.54 (3H, m), 4 .58 (1H, t), 4 .18-4 .08 ( 1 H , m), 3 .92-3 .32 ( 6 H , m), 3 .10-2 .66 (2H , m), 1.92-1.14 ( 2 1 H , m); IR ( c m ' 1 ) : 3556 -3328 (H-bonded O H , br), 3062 (phenyl C - H , w), 1718 ( C = 0 , st), 1310 ( S = 0 st); MS (m/e, relat ive intensity): 367 ( M + - C 4 H 7 0 2 , 23) , 353 (16), 352 (17), 351 (18), 350 (17), 3 4 0 (11), 339 (45), 3 1 3 (33), 309 (12), 285 (14), 283 (17), 267 (21) , 211 (11), 144 (14), 143 (51), 142 (20), 126 (21), 211 (11), 144 (14), 143 (51) , 142 (20) , 126 (21), 125 (100) , 124 (45), 101 (58), 100 (30), 9 7 (21), 96 (15), 95 (12), 86(50) , 85 (49), 84 (73), 83 (85), 82 (18), 81 (22), 79 (12), 78 (22) , 77 (37), 71 (14), 70 (13), 69 (53), 68 (18), 67 (47), 57 (63), 56 (53), 55 (86), 54 (14), 4 3 (56), 42 (57), 41 (83), 39 (17). 62 1 - H v d r o x v - 1 3 - m e t r a h v d r o - 2 H - p y r a n - 2 - y n o x y ] - t r i d e c a n - 4 - o n e ( M ) PhS02 O OTHP O sa OTHP Aluminum foil (813 mg, 0.15 mmol) was cut into strips approximately 5 cm x 0.5 cm and immersed into a 2 % aqueous mercuric chloride solution for 15-25 seconds . The aluminum strips were r insed with methanol and ether, cut into p ieces approximately 0.5 c m 2 and added immediately to a solution of sulfone QI (909 mg, 2.01 mmol) in 60 mL of 10% aqueous T H F . The mixture was stirred at reflux for 8 h, cooled and filtered. The sol id phase was washed with T H F and the filtrate evaporated under vacuum to remove most of the solvent. The residue was extracted with ether (3 x 20 mL). The combined organic layers were dried over M g S 0 4 and concentrated under vacuum. Purification of the crude product by column chromatography using a mixture of petroleum ether : ethyl acetate (1 : 1) as eluent gave compound 6JL (495 mg, 79%) as a light yel low oi l . 1 H NMR (300 M H z , CDCIg ) 8: 4 .58 (1H, t), 3.92-3.32 (6H, m), 2.58 (2H, t), 2.44 (2H , t), 1.90-1.22 (21H, m) ; IR ( c m - 1 ) : 3441 (free O H , st), 1710 ( C = 0 , st); MS (m/e, relat ive intensi ty) : 296 ( M + - H 2 0 , 18), 211 (21), 101 (16), 97 (88), 95 (13), 85 (100), 84 (49), 81 (14), 71 (13), 69 (18), 67 (22), 57 (13), 56 (15), 55 (60) , 4 3 (19), 41 (25). 63 4 - f 9 - ( T e t r a h v d r o - 2 H - p v r a n - 2 - v n o x y n o n y l l - p e n t - 4 - e n e - 1 - o l (92) liO^^^^^^^^^^^^^^ OTHP H O ^ * V - ^ S H X ' , N / ^ ^ O 1 22 a-Butyl l i th ium (3.13 mL , 5.0 mmol) was injected into 50 mL of dry ether at room temperature under N 2 . T r ipheny lmethy lphosphon ium bromide (1.79 g , 5.0 mmol) w a s caut iously added in portions and the resulting orange solution was stirred vigorously at room temperature for 4 h. The keto compound £ £ (628 mg, 2.0 mmol) was d isso lved in 15 mL of dry ether and added dropwise to the reaction v ia an addition funnel, upon which the orange color d ischarged and a white precipitate formed. The mixture was stirred at reflux for 80 h, coo led and fi l tered. The ether filtrate was washed with 1N HCI and brine, dr ied over M g S 0 4 and concentrated under vacuum. The crude oil was chromatographed on a si l ica gel column using a mixture of petroleum ether : ethyl acetate (6 : 1) as eluent to give the alkene compound £ 2 (499 mg, 80%) as a light yel low oi l . 1H NMR (300 M H z , CDCI3) 8: 4.74 (2H , s) , 4 .58 (1H, t), 3 .92-3.32 (6H, m), 2.12 (2H, t), 2.04 (2H, t), 1.90-1.22 (21H, m); IR ( c m ' 1 ) : 3536-3252 (H-bonded O H , br), 3076 (=C-H, w), 1644 (C=C, m); MS (m/e, relat ive intensi ty): 311 ( M + - 1, 3), 210 ( M + - T H P O H , 7) , 109 (15), 101 (48), 97 (20), 96 (10), 95 (38), 86 (22), 85 (70), 84 (53), 83 (37), 82 (27), 81 (40), 69 (68), 68 (24), 67 (81), 57 (50), 56 (51), 55 (100), 4 3 (55), 42 (13), 41 (87), 39 (23) ; 64 2 - [ ( 3 - A c e t o x v ) - p r o p v n - 1 1 - r n e t r a h v d r o - 2 H - p y r a n - 2 - y n o x v 1 - u n d e c - 1 - e n e ££3_) £2 Fresh ly dist i l led acet ic anhydr ide (0.99 mL , 10 mmol) and pyridine (0.85 m L , 10 mmol) were injected into 125 mL pi anhydrous ether at room temperature under N 2 . A catalytic amount of 4-dimethylaminopyridine (DMAP) was added . The alcohol SZ (2.18 g , 7.0 mmol) was d isso lved in 20 mL of dry ether and added dropwise to the mixture v ia an addition funnel. The reaction was stirred at room temperature for 4 h. The mixture was diluted with ether and w a s h e d three t imes with br ine. The organic phase was dr ied over M g S 0 4 and concentrated under vacuum. Purification of the crude product by column chromatography using a mixture of petroleum ether : ethyl acetate (3 : 1) as eluent gave the acylated compound SZ (2.41 g , 94%) as a yel low oi l . 1H NMR (300 M H z , CDCI3) 8: 4.74 (2H, d), 4.58 (1H, t), 4.08 (2H, t), 3.92-3.34 (4H, m), 2 .12-1 .98 (7H , m), 1.90-1.22 (20H, m); IR ( c m - 1 ) : 3076 (=C-H, w), 1743 (C=0 , st), 1646 (C=C, m); MS (m/e, re lat ive in tensi ty) : 354 ( M + , 4) , 353 ( M + - H, 3) , 210 (18), 123 (15) , 121 (13), 111 (10), 110 (15), 109 (21), 108 (16), 107 (10), 101 (17), 97 (18), 96 (22), 95 (62), 94 (13), 93 (21), 85 (79), 84 (39), 83 (39), 82 (88), 81 (54), 80 (13), 79 (27), 69 (42), 68 (30), 67 (100), 57 (24), 56 (37), 55 (90), 54 (26), 53 (21), 42 (88), 41 (67); E x a c t m a s s ca lc . for C 2 i H 3 8 0 4 : 354 .2771 ; Found : 354.2769. 65 1 0 - f ( 3 - A c e t o x y l - p r o p v n - u n d e c - 1 0 - e n e - 1 - o l (98) The T H P ether 23. (4.17, 11.8 mmol) was d isso lved in 30 mL of dry M e O H and injected into 120 mL of dry M e O H at room temperature under N 2 . Pyr id in ium p-toluenesulfonate (306 mg, 1.18 mmol) was added and the mixture was stirred at room temperature for 48 h. The solvent was evaporated under vacuum and the residue was taken up in ether, washed with saturated aqueous sod ium bicarbonate, dr ied over M g S 0 4 and concentrated under v a c u u m . Purif ication of the crude product by column chromatography using a mixture of petroleum ether : ethyl acetate (3 : 1) as eluent gave the alcohol 2fi (2.85 g, 90%) as a color less oi l . 1 H NMR (300 M H z , C D C I g ) 8: 4.74 (2H, d), 4.08 (2H, t), 3.64 (2H, t), 2.12-1.98 (7H , m) , 1.86-1.22 ( 1 7 H , m); IR ( c m " 1 ) : 3 5 3 4 - 3 2 2 0 (H -bonded , O H , br), 3076 (=C-H, w), 1743 ( C = 0 , st), 1646 (C=C, m); MS (m/e, relative intensity): 252 ( M + - H 2 0 , 7), 210 ( M + - A c O H , 4), 110 (10), 109 (12), 108 (11), 97 (10), 96 (13), 95 (51), 83 (18), 82 (100), 81 (18), 79 (12), 69 (23), 68 (19), 67 (92), 57 (10), 56 (14), 55 (32), 43 (42), 41 (22). 66 2-[(3-Acetoxy^propyl1-11-tert-butyldimethvlsiloxvundec-1-ene (100) To 40 mL of dry CH 2 CI 2 at room temperature was added successively the alcohol 9_Q. (540 mg, 2.0 mmol) in 5 mL of dry CH 2 CI 2 , triethylamine (0.56 mL, 4.0 mmol), DMAP (49 mg, 0.4 mmol) and tert-butyldimethylsilyl chloride (452 mg, 3.0 mmol) under N 2 . The mixture was vigorously stirred at room temperature for 4 h, quenched with 1N HCI and extracted with ether. The organic layer was washed with saturated aqueous bicarbonate, brine, dried over MgS0 4 and concentrated under vacuum. The crude product was purified by column chromatography using a mixture of petroleum ether : ethyl acetate (6 : 1) as eluent to give the protected TBDMS ether 1Q0_ (730 mg, 95%) as a yellow oil. 1 H NMR (300 MHz, CDCI3) 8: 4.74 (2H, d), 4.08 (2H, t), 3.60 (2H, t), 2.12-1.98 (7H, m), 1.84-1.22 (16H, m), 0.90 (9H, s), 0.06 (6H, s); IR (cm"1): 3076 (=C-H, w), 1743 (C=0, st), 1646 (C=C, m), 1100 (Si-O, st); MS (m/e, relative intensity): 327 (M + - C 4 H 9 , 23), 118 (10), 117 (100), 109 (13), 95 (20), 83 (12), 81 (22), 75 (37), 73 (11), 69 (20), 67 (21), 55 (22), 43 (18), 41 (14). 67 1-ff3-Acetoxy)-propvl|-1-f(9-tert-butyldimethvlsMoxyl-nonyn-cyclopropane (101) A c O ^ — ~ Y ^ ^— ^ ^ O T B D M S A c O ^ ^ ^ r ^ ^ ^ ^ 101 To a well-stirred, heated (55 °C) solution of the alkene 1Q0_ (450 mg, 0.96 mmol) in 20 mL of dry toluene was injected a solution of diethyl zinc (Et2Zn) in toluene (1.75 mL, 1.92 mmol) and freshly distilled methylene iodide (CH2I2) (0.30 mL, 3.8 mmol) under N 2 . The mixture was stirred at 55 °C for 22 h. A further injection of CH 2 I 2 (0.15 mL, 1.9 mmol) and Et 2 Zn (0.88 mL in toluene, 0.96 mmol) followed by 11 hours of stirring was still not sufficient to complete the reaction. Two further additions of CH 2 I 2 (0.15 mL, 1.9 mmol) and Et 2Zn (0.88 mL, 0.96 mmol) and a further stirring for 20 h completed the reaction. The mixture was cooled to room temperature and quenched with 10 mL of 1N HCI. The layers were separated and the aqueous layer extracted with ether. The combined organic phases were washed with 1N HCI, brine, dried over MgS0 4 and concentrated under vacuum. Purification of the crude product by column chromatography using a mixture of petroleum ether : ethyl acetate (9 : 1) gave the cyclopropyl compound 1QJ_ (410 mg, 88%) as a yellow oil. 1 H NMR (300 MHz, CDCIg) 8: 4.08 (2H, t), 3.60 (2H, t), 2.06 (3H, s), 1.74-1.22 (20H, m), 0.90 (9H, s), 0.22 (4H, d), 0.06 (6H, s); IR (cm - 1): 3065 (cyclopropyl C-H, w), 1743 (C=0, st), 1100 (Si-O, st); MS (m/e, relative intensity): 341 (M + - C 4 H 9 , 23), 118 (10), 117 (100), 109 (13), 95 (20), 83 (12), 81 (22), 75 (37), 73 (11), 69 (19), 67 (19), 55 (20), 43 (13), 41 (10). 68 1 - A c e t o x y - 4 . 4 - d i m e t h y l - 1 3 - t e r t - b u t y l d i m e t h y l s i l o x y t r i d e c a n e (1 02 ) OTBDMS To a solution of the cyclopropyl compound 101 (580 mg, 1.46 mmol) in 5 mL of glacial acet ic ac id was added P t 0 2 (99 mg). The resultant suspens ion was stirred under an H 2 atmosphere (3.5 atm.) for 24 h. Saturated aqueous sod ium bicarbonate w a s added to the mixture with stirring until it b e c a m e bas ic . The resultant aqueous slurry was extracted with ether (4 x 20 mL). The combined ether extracts were washed with brine, dried over M g S 0 4 and concentrated under vacuum. The crude product was chromatographed on a si l ica gel column using a mixture of petroleum ether : ethyl acetate (9 : 1) as eluent to give the hydrogenation product 1Q2. (110 mg) as a yel low oil and the dimethyl a lcohol 103 (218 mg) as a color less oi l . The hydrogenation reaction yield was 7 9 % based on the yield of the next reaction. 1 H NMR (300 M H z , CDCI3) 8 : 4.04 (2H, t), 3.60 (2H, t), 2.06 (3H, s) , 1.60-1.16 (20H, m), 0.90 (9H, s), 0.84 (6H, s), 0.06 (6H, s ) ; IR ( c m ' 1 ) : 1743 (C=0 , st), 1100 (S i -O , st); MS (m/e, re lat ive in tensi ty) : 343 ( M + - C 4 H 9 , 48) , 300 (10), 299 (35), 283 (10), 255 (11), 135 (18), 118 (10), 117 (100), 111 (16), 97 (25), 83 (43), 75 (27), 73 (11), 71 (11), 69 (31), 57 (13), 55 (23), 43 (13). 69 13-Acetoxy-10.10-dimethvltridecan-1-ol M03) The TBDMS ether 1Q£ (220 mg, 0.55 mmol) was dissolved in 5 mL of dry MeOH and injected into 20 mL of MeOH at room temperature under N 2 . PPTs (15.6 mg, 0.06 mmol) was added and the mixture was stirred at room temperature for 24 h. The solvent was evaporated under vacuum and the residue was taken up in ether, washed twice with saturated aqueous sodium bicarbonate and once with brine. The organic phase was dried over M g S 0 4 and concentrated under vacuum. Purification of the crude product by column chromatography using a mixture of petroleum ether : ethyl acetate (3 : 1) as eluent gave alcohol 103 (136 mg, 87%) as a colorless oil. 1 H NMR (300 MHz, CDCIg) 8: 4.04 (2H, t), 3.64 (2H, t), 2.06 (3H, s), 1.62-1.14 (21H, m), 0.84 (6H, s); IR ( cm - 1 ) : 3560-3204 (H-bonded, OH, br), 1743 (C=0, st); M S (m/e, relative intensity): 268 ( M + - H 2 0 , 0.2), 226 ( M + - AcOH, 5), 211 (18), 111 (28), 109 (16), 101 (14), 97 (43), 96 (11), 95 (22), 85 (16), 84 (20), 83 (71), 82 (28), 81 (23), 71 (18), 70 (13), 69 (70), 67 (21), 61 (29), 57 (29), 56 (18), t5 (100), 43 (56), 42 (11), 41 (32). 70 13 -Ace toxy -10 .10 -d ime thy l t r i decano i c ac id M041 O 104 Alcohol 103 (270 mg, 0.94 mmol) was d issolved in 5 mL of acetone and cooled to 0 °C with an ice bath. Jones reagent was added dropwise v ia a smal l syr inge under N 2 until the solution stayed dark-brown (like the Jones reagent itself). The mixture was stirred at 0 °C for 20 minutes. 2 -P ropano l was added slowly until the solut ion b e c a m e c lear with a blue precipitate being formed. The sol id was filtered and washed with acetone. The filtrate was concentrated under vacuum and the residue taken up in ether. Aqueous 15% N a O H was added and the layers separated. The aqueous phase was acidif ied with 1N HCI and extracted with E tOAc . The comb ined organic p h a s e s were dr ied over M g S 0 4 and concent ra ted under v a c u u m . Puri f icat ion of the crude product by co lumn chromatography using a mixture of petro leum ether : ethyl acetate ( 3 : 1 ) and approximately 1% acetic ac id as eluent gave acid 104 (105 mg, 77%) as a color less oi l . 1H NMR (300 M H z , C D C I 3 ) S : 4.04 (2H, t), 2.36 (2H, t), 2.06 (3H, s) . 1.68-1.14 (19H, m), 0.84 (6H, s ) ; IR ( c m " 1 ) : 3480-3014 (acid O H , br), 1743 (ester C = 0 , st), 1712 (acid C = 0 , st); MS (m/e, relat ive intensi ty) : 240 ( M + - A c O H , 3), 225 (13), 212 (34), 199 (12), 181 (10), 97 (12), 84 (12), 83 (100), 82 (11), 69 (18), 61 (13), 57 (12), 56 (10), 55 (37), 43 (28), 41 (16). 71 1 0 . 1 0 - D i m e t h v l - 1 3 - h y d r o x v t h d e c a n o i c a c i d M 0 5 ) Aceta te 104 (100 mg, 0.33 mmol) was d isso lved in 5 mL of dry M e O H . Pulver ized potassium carbonate (101 mg, 0.73 mmol) was added under N 2 and the mixture was vigorously stirred at room temperature for 5 h. The solvent was evaporated under vacuum and the residue was taken up in ether, washed twice with 1N HCI and once with brine. The organic phase was dr ied over M g S 0 4 and concentrated under vacuum to give the c lean co-hydroxy ac id 105 (84 mg, 98%) which was carr ied directly to the next react ion. 1H NMR (300 M H z , C D C I 4 ) 8: 3.62 (2H, t), 2.36 (2H, t), 1.68-1.14 (20H, m), 0.84 ( 6 H , s ) ; IR ( c n r 1 ) : 3564-3014 (alcohol O H , acid O H , br), 1712 (acid C = 0 , st); MS (m/e, relat ive intensity): 240 ( M + - H 2 0 , 1), 199 (12), 181 (17), 163 (13), 125 (10), 111 (12), 101 (30), 99 (10), 97 (22), 95 (10), 85 (10), 84 (12), 83 (100), 82 (10), 81 (28), 71 (17), 70 (10), 69 (37), 67 (11), 57 (26), 56 (18), 55 (71), 4 3 (33), 41 (31). 72 1 0 . 1 0 - D i m e t h y l t r i d e c a n o l i d e (42) 1 -Pheny l - 2 - t e t r azo l i ne -5 - t h i one (71 m g , 0 .40 mmol) and te r t -bu ty l i socyan ide (0.05 mL, 0.4 mmol) were added to 3 mL of dry toluene at room temperature under N 2 . After 10 minutes of stirring the homogeneous solution was added to a solution of the hydroxy acid 10J*. (80 mg, 0.31 mmol) in 6 mL of dry toluene under N 2 . The mixture was diluted with 60 mL of dry toluene and stirred at reflux for 4 h. The mixture was cooled, evaporated under vacuum to a vo lume of approximately 4 mL. This concentrated solution was filtered through a short s i l ica gel column using benzene as eluent to give macrol ide 42 (49 mg, 66%) as a light yellow sol id. The dimer by-product (12 mg, 20%) was also isolated, which could be hydrolyzed back to acid 1 0 5 . 1 H N M R (400 M H z , CDCI3.) 5: 4 .22-4.14 (2H, m), 2 .44-2.36 (2H, m), 1.74-1.12 (18H, m), 0.84 (6H, s ) ; IR ( c m - 1 ) : 1718 ( C = 0 , st); M S (m/e, relat ive intensi ty): 240 ( M + , 11), 225 ( M + - C H 3 , 14), 212 (47), 84 (13), 83 (100), 82 (22), 69 (22), 67 (10), 56 (13), 55 (43), 43 (10), 41 (20); E x a c t m a s s ca lc . for C 1 5 H 2 8 0 2 : 240.2090; Found : 240.2085. 73 3.2 Preparat ion of n-Octyl Pentanoate M09) 109 To a well-st irred, ice coo led solution of 1-octanol (390 mg, 3.0 mmol) and valer ic acid (367 m g , 3.6 mmol) in 30 mL of dry C H 2 C I 2 w a s a d d e d s low ly a so lu t ion of dicyclohexylcarbodi imide (DCC) (655 mg, 3.2 mmol) in 5 mL of C H 2 C I 2 and a catalytic amount of D M A P . The mixture was stirred at room temperature overnight. The urea was filtered off by suct ion fi l tration. The filtrate was di luted with C H 2 C I 2 and w a s h e d twice with 1N HCI, saturated sod ium bicarbonate and brine. The organic phase w a s dr ied over M g S 0 4 and concentrated under vacuum. The crude oil was purified by co lumn chromatography using a mixture of petroleum ether : ethyl acetate (6 : 1) as eluent to give ester 109 (604 mg, 94%) as a color less oi l . 1H NMR (400 M H z , C D C I 3 ) 8: 4 .25 (2H, t), 2.50 (2H, t), 1.85-1.05 (22H, m); IR ( c m " 1 ) : 1722 ( C = 0 , st), 1173 ( C - O , m); MS (m/e, relat ive intensi ty) : 214 ( M + , 4) , 172 (7), 158 (12), 157 (22), 112 (40), 103 (100), 85 (87), 84 (41), 83 (38), 71 (23), 70 (56), 69 (30), 61 (13), 60 (10), 57 (97), 56 (51), 55 (43), 44 (13), 43 (53), 42 (27), 41 (62), 39 (13); E x a c t m a s s ca lc . for C - ) 3 H 2 g 0 2 : 214.1934; Found : 214.1934. 74 3.3 Preparat ion of t r idecanol ide (35) 15 Trif luoroacetic anhydr ide (0.68 mL, 4.8 mmol) was added slowly to a solution of 9 0 % hydrogen peroxide (0.14 mL, 4.0 mmol) in 4 mL of C H 2 C I 2 at 0 °C under N 2 . The solution was st i r red at 0 °C for 25 min and at room tempera tu re for 10 m in . T h e resul t ing peroxytri f luoroacetic ac id was added slowly to a well-stirred mixture of cyclotr idecanone (393 mg , 2.0 mmol) and d isod ium hydrogen phosphate in 30 mL of C H 2 C I 2 at 0 °C. After the addi t ion, the mixture was st irred at reflux for 2 h. The mixture w a s coo led to room temperature and poured into water. The organic layer was washed with saturated aqueous sodium bicarbonate, brine, dried over M g S 0 4 and concentrated under vacuum. The crude oil was purif ied by co lumn chromatography using toluene as eluent to g ive macrol ide 3JL (246 mg, 63%) as a color less oi l . 1H NMR (400 M H z , C D C I 3 ) 5: 4.35 (2H, m), 2.58 (2H, m), 1.90-1.40 (20H, m); IR ( c m - 1 ) : 1720 ( C = 0 , st), 1144 ( C - O , m); MS (m/e, relat ive intensi ty) : 212 ( M + , 12), 194 (14), 176 (11), 169 (8), 111 (15), 110 (21), 98 (33), 97 (30), 96 (34), 95 (17), 84 (31), 83 (42), 82 (41), 81 (22), 73 (16), 71 (13), 70 (22), 69 (58), 68 (32), 67 (27), 60 (10), 5 7 (18), 56 (28), 55 (100), 54 (13), 43 (32), 42 (21), 41 (59); E x a c t m a s s ca lc . for C 1 3 H 2 4 0 2 : 212.1777; Found : 212.1775. 75 3.4 Hydro lys is of 10.10-Dimethvrtr idecanol ide (42). Tr idecanol ide (35) and n-Octvl Pentanoate hOQ) Macro l ide 42. (5 mg, 0.02 mmol), macrol ide 2 5 (4 mg, 0.02 mmol), ester 109 (7 mg, 0.03 mmol) and dodecane (3 mg, 0.02 mmol) were d isso lved in 1 mL of M e O H . Pulver ized po tass ium carbonate (20 mg, 0.15 mmol) was added and the mixture was stirred at room temperature. The react ion was monitored by gas- l iquid chromatography (glc). The relative intensities of the hydrolysis compounds to the dodecane internal standard were recorded. The hydrolysis react ion was also carr ied out at 0 °C and -20 °C under the s a m e procedure as descr ibed above. R E F E R E N C E S (1) Brockmann, H. and Henke l , W. , Naturwissenschaften, 1950 ,2 Z . 138. (2) Djerassi , C . and Zder ic , J .A . , J. Am. Chem. Soc, 1 9 5 6 , ZS., 6390. (3) Wi ley , P . F . , G e r z o n , K., S i g a l , M.V. , Weaver , 0., Quarck , U .C . , Chauvet te , R .R . and Monahan , R., J. Am. Chem. Soc, 1957 , ZS, 6062. (4) Wi ley, P . F . , S iga l , M.V. , Weaver , O . , Monahan , R. and G e r z o n , K., J. Am. Chem. Soc, 1 9 5 7 , 7_9_, 6070. (5) Woodward , R.B. , Angew. Chem., 1 9 5 7 , 55, 50. (6) (a) Ke l le r -Sch ie r le in , W . , Fortschr . Chem. Org. Naturstoffe, 1 9 7 3 , 3JL 313; (b) Back , T . G . , Tetrahedron, 1 9 7 7 , 3J1, 3041; (c) N ico laou, K . C . , Tetrahedron, 1 9 7 7 , 3JL 683. (7) M a s a m u n e , S . , Ba tes , G . S . and Corcoran , J . W . , Angew. Chem. Int. Ed. Engl., 1 9 7 7 . 16. 5 8 5 . (8) Dale, J . , J. Chem. Soc, 1 9 6 3 , 93. (9) Ce lmer , W .D . , Antimicrob. Agents Chemother., 1 9 6 5 , 144. (10) E g a n , R . S . , Perun , T . J . , Mart in, J . R . and Mitscher, L A . , Tetrahedron, 1 9 7 3 , £9., 2525 . (11) Harr is, D.R., M c G e a c h i n , S . G . and Mi l ls , H.H. , Tetrahedron Lett., 1 9 6 5 , 679. (12) (a) Marsha l l , J . A . , Synthesis, 1 9 7 1 , 229; (b) Grob , C .A . , Angew. Chem. Int. Ed. Engl., 1 9 6 7 , 5 , 1; (c) Fel ix , D., Schre iber , J . , Ohloff, G . and Eschehmose r , A . , Helv. Chim. Acta, 1971 ,54. , 2896. (13) Berge lson , L.D., Molotkovsky, J . G . and Shemyak in , M .M . , Chem. and Ind., 1960 , 558. (14) Carnduff, J . , Egl inton, G . , M c C r a e , W . and Raphae l , R.A. , Chem. and Ind., 1960 , 559. (15) (a) Corey , E . J . and H a m a n a k a , E. , J. Am. Chem. Soc, 1 9 6 4 , £5, 1641; (b) Corey , E . J . and S e m m e l h a c k , M .F . , Tetrahedron Lett., 1 9 6 6 , 6237; (c) Co rey , E . J . and Helquist , P. , Tetrahedron Lett, 1 9 7 5 , 4091. (16) Dauben, W . G . , Beas ley , G . H . , Broadhurst, M.D. , Muller, B., Peppard , D .J . , Pesne l le , S . and Suter , C , J. Am. Chem. Soc, 1 9 7 5 , 31, 4973. 77 (17) Crombie, L , Kneen , G . and Pattenden, G . , J. Chem. Soc Chem. Commun., 1976, 66. (18) Corey , E . J . and Petrz i lka, M . , Tetrahedron Lett., 1975, 2537. (19) (a) Hurd, B . N . and S h a h , D.H. , J. Org. Chem., 1973 ,28 . , 390 ; (b) Hurd, B . N . and S h a h , D.H. , J. Med. Chem., 1973 , 1 1 , 543 . (20) (a) Stol l , M . and Rouve , A . , Helv. Chim. Acta, 1934,12, 1283; (b) Stol l , M . and Rouve , A . , Helv. Chim. Acta, 1935,11, 1087. (21) Peters , C A . and Hurd, R . N . , J. Med. Chem., 1975,11, 215. (22) Kur ihara, T., Nakaj ima, Y . and Mi tsunobu, 0., Tetrahedron Lett., 1976, 2455 . (23) Fukuyama, Y . , K i rkemo, C L and White, J . D . , J. Am. Chem. Soc, 1977,21, 646. (24) (a) Corey , E . J . and Nico laou, K . C , J. Am. Chem. Soc, 1974,21, 5614; (b) Corey , E . J . , Trybulsk i , E . J . , Me lv in , L .S . , N ico laou , K . C , Secr is t , J . A . , Lett, R., She ld rake , P .W. , Fa lck , J . R . , Brunel le , D .J . , Has lahger , M.F . , K i m , S . and Y o o , S . , J. Am. Chem. Soc, 1978. 1 0 0 . 4 6 1 8 ; (c) Co rey , E . J . , K i m , S . , Y o o , S . , N ico laou , K . C , Me lv in , L .S . , Brunel le , D . J . , Falck, J . R . , Trybulsk i , E . J . , Lett, R. and Sheldrake, P.W. , J. Am. Chem. Soc, 1978,111, 4620 . (25) (a) Muka iyama, T., Synthetic Comm., 1972 ,2, 243 ; (b) Muka iyama, T., Angew. Chem. Int. Ed. Engl., 1976, H , 94. (26) (a) Ger lach , H. and Tha lmann , A . , Helv. Chim. Acta, 1974,51, 2 9 3 ; (b) Ger lach , H. and Kunz ler , P . , Helv. Chim. Acta, 1978, H , 2503 . (27) Schmidt , U. and Dietsche, M. , Angew. Chem. Int. Ed. Engl., 1981,21, 771 . (28) St i l l , W . C , J. Am. Chem. Soc, 1979,111, 2493 . (29) Sti l l , W . C . and Novack, V . J . , J. Am. Chem. Soc, 1984,111, 1148. (30) St i l l , W . C . and Romero , A . G . , J. Am. Chem. Soc, 1986,101, 2105 . (31) Da le , J . , Angew. Chem. Int. Ed. Engl., 1966 ,1, 1000. (32) Da le , J . , Acta Chemica. Scand., 1973, 27 , 1115; 1973, 2Z, 1131 ; 1973, 2Z, 1149. (33) Duni tz, J . D . , X - ray Ana l ys i s and the Structure of Organ ic Mo lecu les . 1979, Corne l l Universi ty P r e s s , I thaca. 78 (34) (a) Bjornstad, S.L., Borgen, G., Dale, J. and Gaupset, G., Acta Chemica. Scand., 1975, B29. 320; (b) Groth, P., Acta Chemica. Scand. Ser. A., 1976,3J1, 155. (35) Neeland, E., Ph.D. Thesis, Univ. of British Columbia, 1988. (36) Keller, T.H., Ph.D. Thesis, Univ. of British Columbia, 1988. (37) Neeland, E., Ounsworth, J.P., Sims, R.J. and Weiler, L, Tetrahedron Lett., 1987, 2JL 35. (38) (a) Allinger, N.L., Gorden, B. and Profeta, S.Jr., Tetrahedron, 1980, 3JL, 859; (b) Burkert, U and Allinger, N L, Molecular Mechanics. ASC Monograph 177, Washington, D.C, 1982. (39) Allinger, N.L, Gorden, B.J., Newton, M.G., Norskov-Lauristsen, L. and Profeta, S. Jr., Tetrahedron. 1982. 38. 2905. (40) Deslongchamps, P., Stereoelectronic Effects in Organic Chemistry. 1983, Permagon Press, Willowdale. (41) Schweizer, W.B. and Dunitz, J.D., Helv. Chim. Acta, 1982, £5_, 1547. (42) Huisgen, R. and Ott., H., Tetrahedron, 1959, £, 253. (43) Jones, G.I.L. and Owen, N.L, J. Mol. Struct, 1973, lfi, 1. (44) Ounsworth, J., Ph.D.- Thesis, Univ. of British Columbia, 1985. (45) Cavicchioli, S., Savoia, D., Trombini, C. and Umani-Ronchi, A., J. Org. Chem., 1984, 4JL 1246. (46) Suter, C.H., The Organic Chemistry of Sulfur. John Wiley & Sons, New York, 1948, 667. (47) Veenstra, G.E. and Zwanenburg, B., Synthesis, 1975, 519. (48) Manescalchi, F., Orena, M. and Savoia, D., Synthesis, 1979, 445. (49) Corey, E.J. and Chaykowsky, M., J. Am. Chem. Soc, 1965, fiL 1345. (50) Money, T. and Kuo, D.L, J. Chem. Soc, Chem. Commun., 1986, 69. (51) Oppolzer, W. and Godel, T., J. Am. Chem. Soc, 1978,10JL 2583. (52) (a) McMurry, J.E. and Choy, W., Tetrahedron Lett, 1980,21, 2477; (b) Boeckman, R.K. and Silver, S.M., Tetrahedron Lett., 1973, 3497. 79 (53) (a) Maercker, A . , Org. React, (N.Y.) 1965,14, 270. (b) Lombardo, L, Tetrahedron Lett., 1982, 23. 4293. (54) Karunaratne, V., Ph.D. Thesis, The Univ. of British Columbia, 1985. (55) Reetz, M.T., Westermann, J. and Steinbach, R., Angew. Chem. Int. Ed. Engl., 1980.1JL 900. (56) Reetz, M.T., Westermann, J. and Steinbach, R . , J. Chem.Soc, Chem. Commun., 1981, 237. (57) Tebbe, F.N., Parshall, G.W. and Reddy, G.S., J. Am. Chem. Soc, 1978,1M, 3611. (58) Brown-Wensley, K.A., Buchwald, S.L, Cannizzo, L, Clawson, L, Ho, S., Meinhardt, D., Stille, J.R., Strauss, D. and Grubbs, R.H. et al., Pure & Appl. Chem., 1983, 5JL 1733. (59) (a) Simmons, H.E. and Smith, R.D., J. Am. Chem. Soc, 1958, fifi.. 5323; (b) Simmons, H.E. and Smith, R.D., J. Am. Chem. Soc, 1959,8_L 4256. (60) Furukawa, J., Kawabata, H. and Nishimura, J., Tetrahedron, 1968, 24, 53. (61) Trost, B.M. and Hiemstra, H., J. Am. Chem. Soc, 1982. 104. 886. (62) (a) Baker, K.M., Shaw, M.A. and Williams, D.H., J. Chem. Soc, Chem. Commun., 1969, 1108; (b) White, E. and McCloskey, J .A., J. Org. Chem., 1970,31, 4241. (63) Klebe, J.F., "Silylation in Organic Synthesis" in "Advances in Organic Chemistry. Methods and Results. Vol. 8". Wiley Intersciehce, New York, N.Y., 1972, 97. (64) (a) Lalonde, M. and Chan, T.H., Synthesis, 1985, 817; (b) Colvin, E., "Silicon in Organic Synthesis". Butterworth, London, 1980, 184. (65) Newham, J., J. Chem. Rev., 1963, 6JL 123. (66) Bowden, K., Heilbon, I.M., Jones, E.R.H. and Weedon, B.C.L. J. Chem. Soc, 1946, 39. (67) Emmons, W.D. and Lucas, G.B., J. Am. Chem. Soc, 1955, TL 2287. (68) (a) Ogura, H., Furuhata, K., litaka, Y., J. Am. Chem. Soc, 1978,1QJ2, 6733; (b) Ogura, H., Furuhata, K., Kuwano, H. and Suzuki, M., Tetrahedron, 1981, 3Z (suppl. 1), 165. (69) Weiler, L. and Ounsworth, J.P., J. Chem. Ed. 1987, £4, 568. (70) Seeman, J.I., Chem. Rev., 1983,33, 83. 80 Appendix 1 MM2 calculations The strain energies of different conformations for macrolide 42, were calculated using Allinger's MM2 computer program. 3 8 b The force field used in molecular mechanics calculations consists of a set of equations derived from classical mechanics, which contain adjustable parameters that are optimized to obtain the best fit of the calculated and experimental properties of the molecules. The MM2 program is much faster than quantum mechanical calculations and produces very reliable values. In addition, the MM2 program allows rapid calculations of considerably large molecules, since the equations used are sufficiently simple to be rapidly solved by modern computers. The quality of a molecular mechanics force field, and hence the reliability of its predictions, is critically dependant on the parameters used. For the calculations in this thesis, the parameters of Allinger's 1982 MM2 force field were employed. All the calculations were performed on the Amdohl computer in the Computing Centre of the University of British Columbia. The time required to generate the conformations, screen out unfavorable selections and calculate the energies were usually 5 to 10 min. Appendix 2 NMR and IR spectra 41 82 66 85 89 90 92 95 96 4000. 3400. 2B00. 2200. 1600. 1000. CM-i 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0060354/manifest

Comment

Related Items