3-Hydroxy-2-alkyl carboxylic acids related to mycolic acid

3-HYDROXY-2-ALKYL CARBOXYLIC ACIDS RELATED TO MYCOLIC ACID and A P. cHAuDHuRl WALTER J. GENSISR*.IpreKHNlAMI, R S. P~A~.AD, A. I. -A mat of Chemistry,BostonUniversity,Clwka RiverCampus,Bo~toa.MA02215,U.S.A. (Rakd

in USA 14 &cmber

Pm)

Abdnct-A ncriesof loog&aia ulydroxy-2dkyl acida struchrrallyandto mycolic acids arc rqmtcd. The syntbeaii appw, alakiu# use of 3.Let0extersa.9iatermadirtes, penaitathe h4!rtionof 2-alkyl gmups havill# II0

Tbemyooticacids(7,witbRIraneiaesshiebes60and R=C,, to C24) are high molecular weight microSilo&al products showing, either as such or as derivatives. diverse bkbemkl and immuilolo&al properties.’ One of the reasons for undertaking the work repoti below was to make analogues available for testing,theresultsofwhichwouldhelpinde&inga stIuctureactivity r&tionship. wrth this in mind, we have now synth&ed a group of longchain 3-hydroxy-2alkylacids7,allm&Maftermycolicacid.Sinceat least in the 6rst go arouQd,stereocbemical indivkhIality was only of seoondary importamze,we defared separations and resohltions and focussed instead on the stnlctures. Of the several approaches to /3-hydroxy acids, prob ablytbemostfamiliaristbeRefomlatskycondensatbn.z However, we did not pursue this m&X& since our target compounds would require reactant pairs-longchain akkhydes with long-chain a-brow, es-mbioations of which are known to give poor results.’ Other pote6ti~Jy useful syntheses are known,’ but hecause none offered the directness of the process formulated as l-to-7. we chose to rely on this sequu~cc.~This sequence may be regarded as a variation of the Claisen ester ~n,onethateoablestheRIand&groopsin7 tobeckseaatwill.Theresultsweobtakdservea secoodpurposeinfuturedekeatingthescopeofthis fkuile approach, with particularreference to the length of tJle substitllcat groups.

Theinitialstagemadeuseoflithiumenolatesduived eitherfromacetateorfromhigberesters(2or3).‘T&se wmacylatcdwithanacidcbbridc1’mgive,respectidy, the3-k& ester4 or the 2-alkyWketo ester 5. The lattacompoumicoulda!sobereacbedbyalkylationof4 witbalkylhalides,someofwhkhhadtobesyntksizaL Reduction with borohydride followed by saponikation of the intermed& Mydroxy-2-alkyl esters 6 fiuni&d tbe desired my&c B 7. The 6nal products 7 had R, =&Ha C,,Hn and

C17H~;R,ransedtrOm HtoCaSts. A-p&km -iIltJlisseqSenM involvedthedefxea&%Milityandlowern?a&ityof thelOllg4klcompcnmds, particularly io the acylation step. Ju&& witb the solvents and operat& at sowwhathi$uxtempenturestbanusuaihc@lheretosolIx extent. Another problem was in the saponilkation step,

R,COCi

+-&9COOR*

IL

1

2

1

R,COCI +-&HCOORa

I + R,COCHCOO~

%

3

1 HCOORP

IWO ‘i R

4

5 H 1

H R, ? HCHCOOH

-

R, E HCHCOORn k 6

7

6-ti7, where the yields could not be brought above 70%. Presumably this is the consequence of a side reaction, cleavage of hydroxi esta 6 by a base~talyxed retrograde aldol process. AnattractivealtenWemethodforsynthe&ingf hydroxy-2-alkyl acid 7 in essentially one step involves adding the dilithium derivative of a fatty acid to an aldehyde carbonyl. This was tried with the dilitbium derivative of ocWkanoic acid and octadecanal, hot was unsuccessful.9

-p~pncortedWitb saica~(0.halaad0.25mmttlidK~.Tlcspotsweremade visibkbyexposiagt#eplatetoiodiaevqora,orbyspmyiag witb~%fiSO,andtbcabeatiagat25WOtP,oroccashaUyby ~UVtigbtTbeFbhiluscdforcohnchromPtopepby WUtiO-lOOmesb.DryCOhUllUChromrtqsnpb

y”

was pufomlL?d

ICNsiticaa14(BmckmnuagradeIII;0.243 mm putkk size); PbM or stradud silka pl (0.06-0.2mm)gave pdnbly

~itb

kLSStrtktrtayl8dkRUMWiOglOW+OiliBgSdVentsmSrCfz4mphMmatiQdyby~inrrot8tingevPpMltorat

water-pumppnumaaDfJatextenlaltempstibiaedatsQor below. A4Jhu&&CoaaMchl~ylrlcobdwkentrutt?d &thlXWithSOChiabeUWCohaiagPyCidiMOrWitll~

W.J.Gmsmetd.

Table 1. Rorohydkk mdctioo of ktoatm T

I to by&oxy-estm 6

mm1 of Na8H 4

solvmlt (1111)

yield

m

of &

Of L

CllH23

C2"s

CA1

3.07

7.27

CH3C#(25)

919

oil

CllH23

cH7

C24"49

o.42

0.21

CH3OH(20)

El!+

699-70.

Cl&3

ez"s

C24"49

o-y)

2.3

C2H50H(10)

74rc

40-45.

Cl755

C%

"

3.4

0.85

CH30H(60)

89zd

56-57.

C17"35

CH3

CH3

0.85

0.41

CH3'N25)

85%'

42-45.

C17"35

u5

C24"49

0.25

3.2

CH30H(35)

90#

62&

y)bcliaedbydrY~aitiaIsl

chomrtolnpbY~~.TbePrutnct~twoovaLppissspota

onrlkpUe*ith%‘arpprox.0.23radO~6:1heuabCthylrcstrte). bplaiasd by dry c&mu cbromrtolnphy (CHCI,). A sccod pass acpmtai tbc pm&t into tbc two dimstaomm. one with 0l.p. 74-7Y (Rj 0.8 with CHCl3, and tbc second witb m.p. 67.5-m (& 0.7 with CIfCl~. ~by~tmdardchrormtolrrpby~ploriril~tberolvcatrhc(1:1xbcnzcoe,ad ~HcI,(1:1).Tbe~rboMd~oovalrppir(ltle~1pIoJlado.uwitb1:4ethyl~ hexme. ~AA~prodPd,whichmrnmovedcoavisatfy~be~Mterhl,provedtoktbe~ hydroxy-acid 7 (It, = C,,H,,; R, = H). The l&my-eater miu pdnc!, pm&d eitbe by dry cohmn chmnntogmphy (CHCl,) or by atadd chronUognph~ Unqh dim H .(CHCl&abowal& O$S witJ~CHG. “” ofke&d’k~&~;~ydn& were taken, tbc ydd dmppedto 61%;WI&exass bombydndc, .fIhromrtogrpby with Fbrisil md CHCl, kd to product &owing a sin& tk spot, 4

0.45 with 85: 1S

ChbOforabetba.

~productcouklberepurtedby~pelchro~y(CHCl~intotvoforms,tbemorepohrofwhieh bd R, 0.25 (CHCI,)ad the kaa polarof which, m.p. 75’-80’,had 4 0.3 (CHCI,).

Tabk 2. CarbooamIhydrogenaqdion

for Mydroxy-wtyl

R2

R3

%"23

C2Hc

Yll

'll"25

c2H6

w35

Fomd

Calcd.

Hydroxy-ator 2

Rl

c&m 6

Mol. Fomlr

C

H

C

H

C21H4203

73.69

12.28

78.83

12.34

C24"49

C40H8003

78.88

13.16

79.19

13.26

Ql3

"

C21H4203

73.63

12.36

73.65

12.35

ClP35

CH3

QI3

w4403

74.16

12.36

74.00

12.68

CA5

Q(3

C24"49

w9003

79.57

13.36

79.58

13.21

3-Hydroxy2-dkyl carboxyiicrddr r&al to mycdic acid

Tabk3. Hvdroxvacii17bv-dh~droxvertard ml

T

ClP23

C&l

CllHz3

C24H49

C17"35

upmfllcrtlom solutlal (rfqwox.)

ratlo

rlkalf:cster

H

Cl,%5

C4

Cl7%

C24H49

C17%5

'16%

Cl7"33

'lL"31

749

72-w’

109 llrol(1:19 H20-C2wfi,

66b

92-94

3sc

109 Lali (1:9 $o-CH30H)

60-69

al-&

rod

106 Neal (CH3W

60-70

5c56d

26e

10% tm

70

66-66e

25f

10s nm

72

79&

119

4% KOM (dloxane, cH3aH.+sane

89

llq.

34'

106 ml

37b

(rthenol)~

plus roa l$o,

(1:9 H20-CH3aH) @H30H plus som

R3

l$o, H20,

Celcd.

llydmnprc~d_L

R1

Yield of 1

lbl. Fomulr

Found

C

Ii

C

H

'llH23

C&l

Cl9w3

72.61

12.11

72.63

12.26

%'23

c24n49

C3#7693

76.62

13.10

76.49

12.91

Cl935

QI3

'Zl"d3

73.69

12.26

73.66

12.42

CA

"

C2A0°3

73.11

12.26

73.33

12.43

CA5

czrH49

CuW3

70.57

13.09

76.66

13.36

2598

W.J. GENSLERet al.

chloride contained a large proportion of octadecyl alcohol, and was unsuitable for the Grignard reaction. Tetracosanyl iodide was available from tetracosanyl bromide or from tetracosanyl alcohol. The bromide was synthesized by extending the carbon chain in octadecyl chloride as follows. With dry argon or nitrogen as an inert atmosphere, octadecylmagnesium chloride was formed by refluxing and stirring a mixture of octadecyl chloride (7.1 g; 25 mmol), MeI (0.03 ml) as a starter, and Mg turnings (0.63 g; 26 mmol) in 15 ml of THF overnight. The Grignard reagent was slowly introduced (1 hr) at 0~ to a vigorously stirred soln of 1,6-dibromohexane (17.8 g; 73 mmol) in dry THF (20 ml) containing 1.5 ml of a soln of dilithium tetrachlorocuprate II previously prepared from LiCI (1.3 retool) and copper(II) chloride (0.7 mmol) in 10 ml of THF. Residual Grignard soln was washed in with 10ml of THF. After 2hr of stirring, NH4CI aq was added, and the mixture was processes as usual to obtain the product, tetracosanyl bromide, which after several crystallizations from acetone weighed 3.1 g (30%) and showed m.p. 51-52.5 ~ (lit.~2 m.p. 52-53.5~ and was homogeneous according to tic. A less soluble fraction, m.p. 71-73 o (or from other runs as high as 84~ which was isolated from the acetone recrystallizations, was taken as the C42Hs6 normal hydrocarbon (lit) 3 m.p. 83-84~ Trials using different molar ratios of reactants gave no better yields of tetracosanyl bromide. This was converted to tetracosanyl iodide by treatment with excess NaI in acetone. Silica gel chromatography (benzene solvent) afforded pure tetracosanyl iodide (90%), m.p. 52-53 ~ (lit) 4 m.p. 54.5-55.5~ An alternate sequence made use of methyl tetracosanoate and tetracosanyl alcohol as intermediates. The ester was obtained by coupling tridecanylmagnesium bromide with methyl 11 - iodoundecanoate as follows) 5 A bright yellow suspension of methylcopper(I) was formed from ethereal methyllithium (llml or 19.8 mmol) and 98% copper(I) iodide (3.90g; 20.5 mmol) in dry THF. After mixing at -65 ~ the reaction was stirred at this temp. for 1 hr; it was then slowly warmed to 0~ and then returned to -65 ~ A Grignard soln, which was formed by stirring and refluxing tridecyl bromide (5.42 g; 20.1 mmol), Mel (0.1--0.2mmol, added initially), and Mg turnings (0.482 g; 20.16 retool) in 10 ml of THF, was injected (10min) at -65 ~ with 5 ml of THF used to complete the transfer. After stirring at -65 ~ for I hr, the mixture was allowed to warm to 10~ as soon as a clear soln developed, it was cooled to -65 ~ Methyl 11-iodoundecanoate (7.0g; 21.5 mmol) in THF was injected (10 min) to this tridecanyl copper complex, and the mixture was stirred at -650 for I hr and then at room temp. for 2 hr before hydrolyzing and processing. Crystallization from acetone afforded white methyl tetracosanoate (4.3 g; 56%, and in other runs up to 75%), m.p. 57-58~ (lit) 6 58.4~ This ester was reduced with LAH in ether to tetracosanyl alcohol, obtained in 86% yield as white, well-formed, homogeneous crystals, m.p. 76-76.5 ~ (litfl 76-77~ Treatment of the alcohol at 110~ with iodine and red P (molar ratios, 1:2:0.8) led to the desired homogeneous tgtracosanyl iodide (81% after acetone crystallization), m.p. 52.5-53.5 ~ Undecyl chloride, tridecyl bromide, and pentadecyl bromide were purchased. Acyl chlorides. Octadecanoic acid with thionyl chloride or with oxalyl chloride gave octadecanoyl chloride,: which was generally distilled (b.p. 165~ at 0.I ram) before use. Dodecanoyl chloride, a commercial product, was redistilled. Ethyl 3 - ketotetradecanoate (4, Rl = CH3(CH2)1o-- and R~ = CH3CH2-). The lithio derivative of isopropyl - cyclohexylamine (distilled from calcium hydride) was prepared under N2 by slowly injecting BuLi (41 ml, 2.40 M in hexane, or 98.4 mmol) into a stirred, -20 ~ solution of freshly distilled amine (20 ml; 110 mmol, in THF (45ml). Adding pure EtOAc (4.9ml; 51.9mmol, followed by 5 ml of THF) at -60 o and then holding the mixture at -20 ~ for 30min allowed the enolate 2 to form: A soln of dodecanoyl chloride (11.6 g; 53.3 mmol) in THF (10 ml plus 5 ml) was added by syringe, and stirring was continued for another 2hr. The mixture, acidified at ice-bath temp. with HCI, was extracted with ether. The extract was shaken with portions of dil. HCI, water, NaHCO3aq, and brine before drying and removing all volatiles at room temp. Vacuum distillation of the residue gave

11 g (78%) of colorless 4(Rt = C~lH23 and R2 = Et), b.p. 123-125~ (0.15mm). This product, before as well as after distillation, showed a single spot on tlc (R: 0.5 with 1:4 EtOAc-hexane) (Found: C, 71.30; H, 11.21. Calc. for C16H3oO3: C, 71.11; H, 11.11). Use of methyl instead of EtOAc afforded the corresponding ~-keto ester product in about the same yield. Ethyl 3-keto-2-tetracosanyltetradecanoate (5, Rt=C~H23, R2 = C2H~, R3 = C24H49) by alkylation. Compound 4(RI = C11H21, R2 = Et) in THF (0.290 g or 1.07 mmol in 10 ml) was injected gradually into a stirred slurry (-10 ~ of Nail (49.5mg or 1.18mmol in the form of a 57% oil dispersion) with 15ml of THF. After stirring the soln at 0~ for 30rain, tetracosanyl bromide (466.5 mg; 1.12mmol) in THF (5 ml) plus hexamethylphosphoramide (5 ml) was injected. The alkylation mixture was stirred overnight at room temp. and then refluxed 6-12 hr. After quenching with ice-cold sat NH4CIaq, and processing in the usual way, the crude product was chromatographed on a Florisil column at first with hexane as solvent and then progressing to 9:1 hexane-benzene, 1:1 hexane-benzene, benzene, 4:1 benzene-chloroform, and lastly chloroform. Unchanged tetracosanyl bromide (95 mg) emerged as an early eluate. The desired product appeared in the 1:1 hexane-benzene and the benzene fractions, followed later by impure starting ester. Crystallization of the product from hexane afforded pure 5(Rl = CllH23, R2 = Et, R3 = C24H49) as a solid (385mg; 59%), m.p. 57-58~ RI 0.61 with 1:4 EtOAc-hexane; IR (CHC13) 1735, 1710cm-l; NMR (CI)C13) 84.3 (q, J = 7.5 Hz, 2, OCH2), 3.38 (t, O O O

JI

II

II

J=7.5, 1, C-CH-C), 2.50 (t, J=6.75, 2, CHzC), 1.27 and 0.88 ppm (t overlapped by s and t, 73, remaining H's). (Found: C, 79.45; H, 13.02. Calc. for C4oH7sO3: C, 79.21; H, 12.87). No alkylation occurred when THF alone was taken as solvent. The corresponding 5(Ri = CliH23, R2 = Me, R3 = C24H49), m.p. 62--64~ was prepared in essentially the same way. Trials making use of lithium isopropylcyclohexylamide in a procedure similar to that described for the synthesis of ethyl 3-keto-2-pentyltetradecanoate failed when applied to the one-step acylation of methyl hexacosanoate with dodecanoyl chloride. At least 80% of the starting ester was recovered. Solubility may have been the problem here, since large amounts of the hexacosanoate failed to dissolve at -60 ~ and as a result, enolate formation had to be carried out at -30 ~ Also, during the acylation step (-55 ~ the reaction mixture persisted as a thick slurry. Using hexamethylphosphoramide with THF did not help. Related attempts to acylate methyl docosanoate with dodecanoyl chloride gave results suggesting that the balky stage was enolate formation. Ethyl 3-keto-2-pentyltetradecanoate (5, Rj=CHH23, R2 = C2H5, R3 = C5H10. The reactants were ethyl heptanoate (0.87 g; 5.5 retool), isopropyl-cyclohexylamine (ll mmol), BuLi (11 mmol), and THF solvent (15 ml). The ester in 5 ml of solvent was added at -60 ~ followed, after I hr, by dodecanoyl chloride (1.31 g; 6.0 mmol) in 10 ml of THF. The acylation was allowed to proceed for 2 hr at -50 ~ before quenching. The product, ethyl 3keto - 2 - pentyltetradecanoate was isolated as a homogeneous material (R: 0.52 with 9:1 hexane-ethyl acetate) by dry-column chromatography I~ (silica gel with benzene) in the form of an oil (l.5g; 81% based on the ester taken); IR (neat) 1740 and 1710cm-~; NMR (CDCIa) 8 4.14 (q, J = 7.5 Hz, 2, OCH2), 3.35 (t, O O O

II

II

II

J=7.5Hz, l, C - C H - C ) , 2.5 (broad t, 2, CH2C), 1.25 and 0.88 ppm (s and t, 35, CH3's and CH2's). (Found: C, 74.24; H, 11.60. Calc. for C21H4003: C, 74.tl; H, 11.77). Methyl 3-keto-2-methyleicosanoate (5, RI = CITH3s, R2 = R3 = CH3). The direction for acylating propionate were essentially the same as those used for acylating acetate. The starting materials were methyl propionate (5.95 mmol), BuLi (equimolar amount), and octadecanoyl chloride (1.8g or 5.95mmol). Dry-column chromatography through silica gel using 2:1 chloroform-hexane served to purify the product. Zones were detected under UV light. Preparative plate chromatography (1 mm silica gel layer

34Iydroxy 2-&i

arhoxyUc acidar&tat

with 1: lO.EtGAcXHCl, dev&Q& lotvent) was ako e&!etiv~ althou& kss convenknt. Coqound S(R, = &Ha. RI = R, = Me) was u bl this way al a N solkl(1.3~ 61%). m.p. 37*; 4 0.5 with CHCI, &v&per; IR (CHCI,) 1725 and 1700cm-‘; NMR (CDCI, with CHU, reference) 8 3.57 and 0 0 3.5 (a phls q. J =45 Hz. 4, GCH3 phu t! CH- 1 ), 255 (1, J = 4 Hz, 5 CHzCO), 1.32 and 0.95ppm (m, 36, all otbw H’s). (polled: C, 74.19; H. 11.62. Cak. for C&i&: C, 74.52; 8, 11.941.

ester. lln

reactantswac hopropyl - cydobcxykrdnc (39.8mmofl. BuLi (39.6mmolL THP (1oOml) fro&Iv dktilkd from LAHMeGAc i19.8mmok and oc&e&oyl &ride (6.09 or2Ommdilll5mlofTHF).Tbeacyb&onmixtMebecamc hom~~luat-ISto(P,wbueit~bddfM#)min.Product waspwifkdbycdumnchroma@raphyon&a~.start& withsolveatbexaneandthen&lgtohexauecQntail&inueas& pqortio~ of CHCI,. Compoud 4(R, -C,,H,s, and & = Me) m.p. 29-313 was obt&d in 63% (4.4n) ykkL Later it wnsfound&atpo&ationbydry+dumac~yon silka ael with 2:l CHCLhexane was more convenient. The pmdw-i rbowed one apot UC(R, 0.57with CHCI, advent). ~~~ C, 73.75; H. 12.20. Cak. for CaH&: C, 74.06; H.

-on

h&l 3-kuo-2.tumw;wyfd~cosanoate (S, R, = C”H,.q, R, CH,.R,=CbJIe)by~Amixtono10.315g(l.~mmd) of 1(R, - C,,Ha,. Rz - Me) in THF (27 IIIII distiikd over tAH (0.1191; 1.05mmol). A sohl of t&awaanyl iodide (0.518; Llmmon ill llml of THP that had been dihltuJ with lmam&ylphosp&runi& was illjcctal Ontin), aftcrwhich stirringwracontinusdfalominatoo.ptIwmtemp.forl.7~hr. andatrdluxfM6hr. Themixture,concentr8tedtoabonthatfitsvoiume,was ~~tratedwitblSOmlof25%HCI,PIld~sedf~ to isolate the pale ycnow crude alkyktioo pro&t. chromato-by through silica gel Q:l hexaoo-CHCU ppve on&rued t&cosanyl i&de (O.i2g; 24%). m.p. SO-SF, &w out t&t, followed by the slower mo\ina 5Gt, - C,,L. R,= Me. RI= CA). %a pmduct (O&g; 1196),- which Ld m.p. ti%%, showed a single qot oa a tic plate (R, 0.5 with CHCl,); IR (CHCI,) 1735, 1705cm-‘; NMR (CDCI, with CHCI, reference)8 3.74 (a, 3, GCH,), 3.4 (1, J = Hz, 1,

1, 2.5 (1, J -7,

CH$), 1.4 and O.%ppm (s and 1. 82, remain& H’s). An additiomll4Ologof~m.P.69-~,~obbined~a mixtore by proc&og various chmmatognphy fractions. (Fowu& C. 80.13; H, 13.24. Cak. for t&H&: C, 79.81; H, 13.10). Isolation of some dbdkylation ptodact. which showed oo NMR 0 0 2:CH- r? sigodandahighCH,coatentstressedtbeimportance of avoid& more thaa eqoimokr amoonts of BuOK. Usiq~ tetracosanyl bromide in plaa of the iodide awl t-BoGH btud ofTHPgaverisetotbede!+dakyktionprodu&althoughio lower yield. Neither NaH oar LiH in THP cootaioiog bexamethylpbosphora&k wan e&tive. 3-Hya’toxy utas 6 by botvhydhk n&don of the Meto Wt~S.ThC3-htoCStCfSSW~tfWYdWithN8B)4iDXhXillg

MeGHorEtGHfM24hr.Tbemokrrat&ofborohvdr&tokotonc varied widely-from 0.2s to n-germ& without atTectiogtheyklddra&aBy.Afterthe~tionp&od,soivent wasremovedatroomtemp.anderreducedprwumua.Tbsr& duewaSeertedwitbwrta,NH&kqL?rHCbt&~with ether. and processed tbmrtta io the used w8Y. cohmm

to my&c add

2999

onccawcouldbe~hltothetwodirrtenohowric forms.IbtbbirIRrbaorption~pnerPnytakeoinCHCl, &tio&tbehydroxyeatcvprodt&36abowaJabsaptkMat 360&3400 aad 172s-172Ocm-‘. PMR spectra @!ncrany io cDclJ.fort&2-~8bowai~intbe regkns a 3.P3.4 &road m, c@H) aDd 2.5-2.1 ppmm. CHCO). Comwand~,=C,,H~R-Me.R,=H)rovePMR~~1 .. -. _ 84.G3.8 (l&a& CWOH) and Z&k &orted d, J;3Hx. CHFO). Tabks1aml2gkedetaikfortheindividualrnns. Attatptswereauktofosm6direcUybyaddiogthetithium emhteof methylto the aIdehyde carbonyl Lronp of ~oroctdarnrl.Rlmsweremadewitblithinm kopropykyclohexykudde io td&ydrohnraaexamethylphospbolu&katteolparangingfrom-6Oto1ooin1~intervala andforvtiuamactionpuiods.fooocruenastbaeevidcnce ofcoadensation.Tbeestc?conldberecoveredingoodykld,aod tksbowedoolytbetwoapotswrrespolKliogtothestarting mate&k. H~ocidc7f~atarLThemabylorethylester6 wasrenuxedwithdkatiinaqneolMmetha&lM&anolfor 2-3.5 hr. tbea wokd and acid&d with s-IO% hydroddoric a&i. lIleproductwaacxtractalwithetherorchIoroforl&andthe extractwaswaabedsevarltimeswulwaterandaaltsolotio& cfriedandstriopedofonsolveat.Themidnewasouri6cdbvdrv or wet column wy. Tabk 2 gives ihe detail; foi ill-m htbehydroxyacidproduct7~r2-alkylgroup.thio kyer chromatography developed two pa&By ove&pping spots. llic tic solvents found w&d iucluded mixtures of chlorofofm pad mctlmd, 20:3 bewpbmetbanoL and I:1 chloroformether. IlIe hydroxy rcids either neat or ill chloroform solution showed IR absorption maximaiotberegiona36aL2sOoaod 171S-1695cm-‘. NMR spectra wae taken with the compouds in CDCI, or ccl& with ordiuy chloroform a the &rnal rrfannce.Inallcasesthehydroxyproto~d7p~abroed sipnrlrt117.2-S,tbepmboa(positioa-3eametsamultipletrt 4.1-3.4, the proton at position-2 M a mdtipkt at 2.7-2.45,sod the rema&g methyl ami methykne protons at 1.5-1.3 and LOSO.Wppm.Intbe~exPmpkofthehydroxy~withmethykaeat~ition-2,~methykneprotonscameoutpradis. torteddoubkt.InaUcases,tbesigoalasso&edwithtbe hydmxyl oroups was lost when Dro was added. Experimental [email protected] informatkmabouttbea&lr7. UnsnccessN attempts were made to obtain hydroxy achl7 (R, =C,,Hu; Rz= C,JIn) directly by lddipp the dilithium derivative of octadecanok acid (formedwithlitbilmldi-impropylrmide in tebnhydrofunm-bexyl phoaphoramS9 to octadararl.TkoftbecrudeprodnctmixtMeshowedtbattbe onlymatuiakpreaeotwerethelm&aqtastartingacidaIld akkhyde;furtbu,tbeacidcwldbesqamtedendcharacterix& llIepossibStythatthedSthimnderivativefaikdtoformwas excluded by isow krge amoonta of hexadecylmaknk ecid afterhubbl&carboadiox&iatotheenokteprq!4uatjoo. Atiwkdgan&a--S work was supported by the Natiooal Cauce~ Institute (N~LCP+iS783) of the National fn&utes of Health, U.S. Public Health !3avke.

‘J. Aaadiueau, lRe Bactetfal L&fds. Holden-my, Ban Fmn&co, Calif. (1966); N. Polgar, In Topics&ILipidMm&try, (Editedby P. D. Guaatooe). Vd. 2, p. 234. Wiky. New York (1971). E. Lcderer. Cti. Pky~. LQti 16, 91 (1976); E. Lukrer, A. Adam, R (Jiohm, J.-F. Petit and J. Wietmbia. &kc.Cdf.Bkd~tm.7.87(197~J.Asa&wao.Fn~.Chem. Ride, TO@CJ a &nmt Chati&

59, i (le$.

kOtGtbU88tiSf~NXUltSkWCbCCll0bthCdWithrbort dmaboaatfrs(acatatcorpropionats)iocombinotionwitb lq ddehydes. Cf. E. I’. Jenoy and C. A. Grab, ifdv.

Chim.

a500

W.J.GENNJLludd

Acta SL. 1936(1953);J. W. Fmhnfeld and J. J. Weamr, I. &g.Ckm.34,3689(1969);D.A.Comfortb,A.B.Opvrmd G. Bead, 1. Ck.m. Sac. (C), 2759 (1969). C. MaImi, D. R Mimikia and N. Polgar, /. C&cm.sot. 5562 (l!X$& lmw p&dOd~h#keumpkOfCo&O8&~IdW&ly&4d akkhyde(C-7)withacDodadyriredu~cster(C_8)inr yiddbcttertlm6O%.M.Behmad,RC~mdM. GaLldemlr,I. s mm. 36, c33 (1972), have used a-hmo SaftJ blated of &romo esters with promhg

idtr;;m3(l~;;c.T.IkwmdC.H.HatbcoclI.Am. ~Soc.n,8109(l~;G.~radT.T~lbld)), 1273(1977). u I mightfor%eatMdudc!hisencoodeMation~ wa&attr&c&elwtbodwhentwoidendal~8re~ billed. However, the sdective fomatkm af 3-kato entaI 6(B,#R,CH~byc’mdm&didgenterta,oft&~type in sot s&f-. (Cf. Il. 0. Hou6c, dludm syncwe Nuhud~, 2Edn, p. 144. W. A. Bmjmin, New York (1912). @Cf.M. w. Bathke in Ref. 2; M. w. Batlh ad J. Deitcb, Ttimk&o~ Lcnar 2953(1971);hf. W. RhthkeaaJ A. Lidat, lbik 39!95IlpIl): M. W. Ratbke and D. !hllim. Ibid m (1972). . .. ‘WebawalsowaiacylimidmkawiUtgoodmadta.Cf.H.A. !kabbdW.Bohr.Nrmhkth&ofPmat&ehadc w, (Edited 6y W. Foerat). Voi. 5, ;. 61. A&mic Pma, New York (KU). I, Cam& I? Pdlirmai ad E Fe&?&Gazz CUR. Id 65, l#,(la~;w.EBrbrmna,W.CdeladA.L~.I.Am. ~Soc.(L8U(1940);F.~mdG.~,luhu

RM-~BrUSoc.~~U1(1~;E.EvmT~ T. A. Bpmcer, D. S. Alkn, Jr. ad R L &via, Tdmkdmn 14,

a (1961).

‘P.LGn~er,~Rrp.M~~U,rm(l977);W.Adrm. J. Baem ad J.C. Liu. 1. Am. Ch. Sot. 94.2ooo (1972):P. E. Pfar d a/.#I. *. &ml. 37.451,1256 (lti); Js; 262(1!no); Tamhedmn Z&ten 699 (1970); G. W. Momcb aad A. R Bwkett, L m. C’kem.J(, 1149(1971):B. An&o. Brlt. Six. aim. fi 1848(1910). 3. Loev and K. M. Smdcr, Ghan III&R&(Lad) 15 (1%5);B. LoevradM.M.Goodrma.IbidZOZd(1969);H.E.Addmd E. thpi, 3. Chawogz ‘II.353 (lmz). “L Ffiedmn ad A. sld, 1. Am. Chl. sot. % 7lOl(l!m). Ahocf.P.Bahh.M.LF.Bahzia,C.C.FortuaodR htoa, I. %. Ckm. 41,2769 (1976). ‘RLDLsiradS.Ddclrl.~~Ulb51,~(19ST)[~ Ahh. 62.6171 (1956)J. ‘9LLukdmds.Lbbial,couc2&ch.GmwlR23,1100 (Pm ‘% Gum a.m. Phm. Brll. (hpm) 6.456.462 11958). “D.EBa&eituawJG.M.Whiksk,3.~.~4D,719 (l!#R D. E. Bergbdx ad J. M. Killwgh. JbU 41, 27SO (191a). “H. Gum and T. Makiw, Yu&ugschiZ&hi 78. 141 (1958) [C%em. Abm. S2, lOE6S(1958)]. “See H. Bicbat, Buff. SIX. CUR. 13: 52 (1946). ‘8. L&m, P. Pmtdaacc and K. Sack-Hamen, Bull Sot. C%fmFr. 413 (1952);I. Polonsky ad R L&m. Jbfd 504 (15’S);cf. V. L Hmky ad P. J. Cdiak, C&a &ng News 23,1332 (EMS).