COMMUNICATIONS

The cyanoborate process on limonene [Eq. (e)], successful [i] A. Pelter, K. Smith, H. C. Brown, Borane Reufents. Academic Press, London,
with the~ylborane,[~] essentially failed for mesitylborane. This 1988.
might be due to the greater facility of thexylborane, as com- [2] E. Negishi, H. C. Brown, Synthesis 1980, 153.
[3] A. Pelter, K. Smith, M. G. Hutchings, K. Rowe, . I Chem. Soi,. Perkin Trunx 1
pared to mesitylborane, for the hydroboration of more hindered 1975, 129.
alkenes. It defines at least one area in which thexylborane will [4] H. C. Brown, E. Negishi, Synthesis 1972, 196.
retain an advantage over mesitylborane. [5] A. Pelter, B. Singaram, L. Warren, J. W. Wilson. Tetrahedron 1993. 49. 2965.
[6] A. Pelter. K. Smith. D. Buss, Zhao Jin, Hewwur. Chem. 1992, 3. 275.
[7] A. Pelter, S. Singaram. H . C. Brown, Tctruhedron Lett. 1983, 24. 1433.
[S] M. Srebnik, T. E. Cole, H. C. Brown, J. Org. Chem. 1990, 55. 5051.
[9] S. U. Kulkarni, H. D. Lee, H. C. Brown, J. Org. Chem. 1980, 45. 4542.
[lo] G . Zweifel. N. R. Pearson. . A m . Chem. SOC.1980, 102, 5919.
I
[ l l ] A. Pelter, M. G. Hutchings, K. Rowe, K. Smith, J. Chrm. Soi,. Perkm Truns. I
[Link]; TFAA; 1975. 138.
H20,I NaOH
A
We next examined the displacement of the mesityl group from
dialkylmesitylboranes, and found two unusual and highly selec- A Concise Total Synthesis of
tive processes that further enhance the usage of mesitylborane. ( & )-Mamanuthaquinone by Using an
The first method consisted of the addition of bromine in em-Diels- Alder Reaction**
methanol to the dialkylmesitylboranes [Eq. (f)] . This caused
cleavage of the mesityl group with production of bromome- Taeyoung Yoon, Samuel J. Danishefsky," and
sitylene and dialkylmethoxyboranes, oxidation of which gave Susan de Gala
the corresponding alcohols in overall yields of 82-87%.
An interesting class of sesquiterpenoid quinones and hydro-
quinones has been isolated from various marine sponges.['' Cer-

MesBR1R2
0r2/
- MesBr + MeOBR1R2 - PI RlOH + R20H
tain members of this family exhibit profiles of cytotoxity,['' an-
timicrobial 31 and inhibition of reverse transcriptase
action.[41Among the novel compounds in this subgroup is ma-
manuthaquinone 1 Is]isolated from Fasciospongia s ~ . ,a sponge
More importantly, the intermediate dialkylmethoxyboranes isolated from Mamanutha Island in the Fiji chain. Mamanutha-
can be reacted directly with a Grignard reagent to give fully quinone exhibits moderate cytotoxity toward human colon tumor
mixed organoboranes, in which all the alkyl groups can be pri- cell lines (HCT-116). Also, the structurally related avarol2 and its
mary. These in turn can be readily converted to the correspond- quinone analogue have been proposed as promising anti-AIDS
ing tertiary alcohols['11 [Eq. (g)]. When R' = n-octyl, RZ = n- drugs,[4a1though their clinical usefulness is far from demonstrat-
hexyl, R3 = n-butyl, the overall yield of isolated carbinol, based ed.l6]
on MesBH, is 69%, which implies a highly efficient process
given the seven steps involved.
HO

NaCN; TFAA;
* HOCR1F?R3
H202I NaOH

1 marnanuthaquinone 2 avarol
A different method of releasing the mesityl group is through
direct reaction with an alkyl or aryllithium (e.g. Eq. (h)). Butyl-
hexyloctylborane produced from the initial reaction was con- In considering a total synthesis of mamanuthaquinone I an
verted to the corresponding carbinol by the cyanoborate pro- obvious, if speculative possibility presented itself (Scheme 1 ) . In
cess in 78% yield showing that there had been little, if any,
scrambling of the alkyl groups. The mesityl group, presumably [*I Prof. S. J. Danishefsky,'" T. Yoon
Department of Chemistry, Yale University
displaced as mesityllithium, was recovered as mesitylene in 88 YO New Haven, C T 06511 (USA)
yield. Telefax: Int. code + (212)772-8691
S. de Gala
Yale Chemical Instrumentation Center
NaCN; TFAA; New Haven, C T 06511 (USA)
MesB(Ckt)(Hex) BuLi_ B u w N H e x ) H,02 / NaOH * H°C(Bu)(Hex)(Oct) (h) [' ] New addresses: Memorial Sloan-Kettering Cancer Center
1275 York Avenue, New York, N Y 10021 (USA)
Telefax: Int. code + (212)772-8691
and
Our studies show that mesitylborane is readily available and Department of Chemistry, Havemeyer Hall
its hydroborating properties, as well as the properties of the Columbia University, New York, NY 10027 (USA)
derived organoboranes, should render it a useful addition to the Telefax: Int. code + (212)854-7142
armoury of organic reagents. [**I This research was supported by Public Health Service (PHS) Grant CA28824.
N M R spectra were obtained through the auspices of the Northeast regional
Received: November 13, 1993 [Z 6500 IE] National Science Foundation (NSF)/NMR Facility at Yale University, which
German version: Angew Chem. 1994, 106, 913 was supported by NSF Chemistry Division Grant CHE 7916210.

A n g e w Chem. l n t . Ed. Ennl. 1994, 33, No. 8 (c) VC'H Verlugs~esellschaftmbH, 0.69451 Weinheim, 1994 0570-0833194jOX08-0853$10.00+ .25,'0 853
COMMUNICATIONS
principle, the generalized system 5 can be assembled from diene We were also mindful, on the other hand, that with the parent
3 and dienophile 4. With a properly chosen aroyl function. 1 1-vinylcyclohexene, Diels-Alder reaction had been reported to
inight well be quickly derivable from the adduct 5. However, in yield cleanly the en& product.[141However, in the case at hand
such a Diels-Alder reaction an e.w transition state is required (3 + 4), we hoped to extend the "cyclopentadiene effect" by
to reach the stereochemistry required for 1. taking particular advantage of the quaternary carbon of the
diene in conjunction with a sterically demanding aroyl group on
the dienophile. In such a cycloaddition, the mdo mode might be
highly disfavored due to repulsion of the aryl group with the
dimethylcyclohexenyl array (Fig. 1). Thus. the use of a maxi-

3 4 5
Scheme 1 . [Link]-Diels-Alder approach for 1

The well-known preference for endo topology in the Diels-

Alder reaction['] is most pervasive with relatively unhindered H,d
cyclic dienophiles (e.g., maleic anhydride, p-benzoquinone, etc.)
endo ex0
and homoannular dienes.['] Nevertheless. endo orientation is
still preferred in most intermolecular processes and if it is not FIE.1 [Link]
addition of 3 to 4 IS favored over mdo addition.
dominant, the cycloadditions tend toward stereorandomness.
However. the attractive forces of secondary orbital interaction
to which the selectivity is generally attributedL9]are rather weak mally substituted aroyl function in the Diels-Alder reaction
and can be easily overridden by seemingly minor structural seemed to be in keeping with the stereochemical rationale, while
changes.["] Indeed, almost exclusive e m additivity has been holding promise for simplifying the late stages of reaching 1. In
exhibited by the reaction of cyclopentadiene and acyclic this communication we describe the first total synthesis of ma-
dienophiles having an alkyl substituent on the carbon bearing manuthaquinone in a remarkably concise way drawing upon
the activating group (e.g., methacrolein)." While the origin of this reasoning (Scheme 2).
this "abnormal" tendency is not clear, the anomaly had been The requisite dienophile 7 was obtained in one step (77%
mostly confined to reactions utilizing cyclopentadiene as the yield) from the reaction of the aryl lithio derivative of 1.2,4,5-te-
diene partner. Conceivably, it could be the consequence of the tramethoxybenzene[''] and 6, which was in turn prepared from
steric repulsion between the methylene group of cyclopentadi- the reaction of tigloyl chloride with N,O-dimethylhydroxyl-
ene and the r-substituent of dienophiles in the transition state amine following the protocols of Weinreb.f'61Cycloaddition of
leading to endo adduct.["] For instance, the same catalytic sys- 7 with 3" 'I under mediation by ethylaluminum dichloride/THF
tem that produces only exo adduct from the cycloaddition of in dichloromethane afforded an 85 % yield of a single cycload-
methacrolein with cyclopentadiene" l e l gave rise to nearly com- duct. The presence of T H F resulted in a doubling of the yield;
plete endo preference with acyclic dienes such as l-methoxybu- without it, decomposition of the diene competed quite signifi-
tadiene." 3] cantly. which was so extensive with other Lewis acids such as
boron triflouride etherate or titanium tetrachloride
that the desired cycloaddition did not proceed at all.
The regiochemistry of the cycloaddition was well
compliant with the precedents and no other regio- or
stereoisomers were detected. Compound 8 was char-
acterized by X-ray crystallography.[**]
With the three stereogenic centers installed in the
7
A required configuration, a straightforward end game
sequence was implemented. Treatment of 8 with lithi-
um aluminum hydride resulted, not only in reduction
of the ketone, but also in demethylation of one of the
ortho methoxy groups. Acetylation of diastereomers 9
Me?
gave 10, whose de-acetoxylation as shown afforded
11, which was then acetylated to give 12 (55% based
on 8). Oxidative demethylation of 12 with ceric am-
monium nitrate followed by de-acylation ( K 2 C 0 , in
MeOH) afforded ( )-mamanuthaquinone (85 %
based on 12; m.p. 114-116"C, the enantiomerically
pure natural product:['] m.p. 108.5-109.5 "C).[19]
(+1 11 (R = H) 9 (R = H) In summary, the total synthesis of I has been ac-
dK12 (R = Ac) "c,o (R =AC)
complished in eight steps in 32% yield from 2,2,4,5-
Schrinc 7. Synthesis of (_f)-maniaiiuthaquinone I . a ) izBuLi. LICI. T H E 0 C +room tempera- tetramethoxybenzene. We hope to sort out the factors
lure. 1 h; then (E)-CH,CH=CCH,CON(CH,)OCH, 6, -78°C -1-oom temperature. 1 2 h; b) 3 which were responsible for the apparent ex0 selectivi-
(Zequiv). EtAICI, (1.5 equiv.). THF (1 equiv), CH,CI,. room temperature. 6 h: c ) LiAIH,.
ty of this Diels-Alder reaction. For instance, assign-
T H E reflux. 2.5 d ; d) Ac,O, Et,N. cat. 4-dimethylaminopyridine. CH,CI,. room temperature.
4 h; e) Li(10 equiv). NH,. -78-C. 10 min: NaOBr; NH,CI; f)(NH,),Ce(NO,),(2.5 equiv),aq. ing the relative importance of the geminal dimethyl
MeCN; g ) K,CO,, MeOH. room temperature, 1 h . center of 3 and the two vinyl methyl groups of 7
 COMMUNICATIONS
would be helpful in planning extensions of these findings. ence map and included to set the orientation of the other two hydrogen atoms.
Further details of the crystal structure investigation are available on request
Studies addressed to such issues as well as to proposals From The Director of the Cambridge Crystallographic Centre. 12 Union Roiid.
relating to the mode of action of mamanuthaquinone are in Cambridge CB2 1EZ (UK) on quoting the full jounal citation
progress. 1191 While we were unable to obtain a sample of the natural product. our assign-
ment is secure given the crystallographically verified structure of 8 and the
complete correspondence of the high-field ' H and " C NMR spectra ofsyn-
thetic 1 with those reported.
[Link] Procedure
8: To :in ice-cold solution containing 7 (208 mg. 0.74 mmol) and T H F (0.06 mL.
1 equiv) iiiCH,CI, (5 mL) was added 1.8 M EtAICI, in toluene (0.60 mL. 1.5 equivj.
followed hy clow dropwise addition of 3 (0.20 g. 2 equiv) in CH,CI, (2 m L ) . After
the addition was complete. the mixture was allowed to warm to room temperature
and stirred for 6 h The reaction was quenched by cautious addition of saturated
NH,CI iit 0 C. l'heaqueous layerthat separated wasextracted with CH,CI, and the
combined organic phases were dried (Na,SO,), concentrated in w c u o , and chro- On the DNA Recognition Role of the
inatograpiied on silica (l0:l to 4:l 1ienane:'EtOAc) to give 263 mg of 8 as a white
solid (X4"b). n i p . 122-124'C (CH2CIZ,'hexane);IR (CDCI,): Fmds= 2939. 1680,
Carbohydrate Sector in Calicheamicin:
1484.1465.1424.1257.1225.11Y7cm~i:'HNMR([Link],):b=6.55(s, A Comparison of DNA Cleaving Capacity of
1 H. phenyl). 5.43 (d, .I = 4.3 Hz. 1 H. vinyl). 3.87 (s. 6 H). 3.73 (s. 3 H).3.70 (s.
3H).~.~O~d..l=l2..[Link]).2.25([Link]).2.0~1.7(m,3H).1.2-1.0(m.3Hj,
Enantiomeric Calicheamicinones**
1 . 0 7 ( s . 3 H ) .1.~12(~.3H).0.92(d,J=[Link](s.3H):'~CN~4R(75MHz. Jayshree Aiyar, Stephen A. Hitchcock, Derek Denhart,
3.2. 149.2. 148.X. 145.7. 138.9, 138.5. 132.8, 115.6. 99.6. 62.0. 61.5,
41 2. 36.5. 33 9. 31.2. 30.3. 29.8. 28.0. 32.7. 17.1. 12.0: MS (20eV Kevin K. C. Liu, Samuel J. Danishefsky,* and
E I ) . i i i I (I-el.i n t . ) : 416 ( 1 ) [M']. 280 (1). 225 (100) [aroyl+], 210 (6). 191 (4); Donald M. Crothers
high-re\oliitioii FAB MS for C,,H,,O,. calcd 416.2564, found 416.2540: correct
elemental aiialysis. The antitumor activity of the naturally occurring drug
calicheamicin 7: (1) is believed to operate through its efficient
Received: November 26. 1993 [Z6516IE] cleavage of duplex DNA in a double-stranded, highly sequence-
German version: A i g e w . Chciii. 1994, 106. 923 selective fashion.['%*] The drug i s comprised of an aryl tetrasac-
charide carbohydrate region tethered to an enediyne-containing
[ l ] For a series of reviews on marine natural products including those from
aglycone moiety. The latter, which is termed calicheamicinone.
sponges. \ee: D. J. Faulkner. Nu/. Prod. R~[Link]/.s1992, Y. 323 and i t s forerun- was synthesized first in our laboratory as its racernater3' (2/3)
ners. and subsequently in the enantiomerically pure form correspond-
[2] 'I) M Kondracki, M. Guyot, Tetrriherlruri 1989,45. 1995: b) M. Kondracki. M. ing to the natural product, by Nicolaou et al.[41
C;u>ot. 7?iruhedroii Lrti. 1987. 5815: c) J. Rodriguez. E. Quinoa. R. Riguera,
B M Petei-s. L. M . Abrell. P. Crews. Z,rriihrriron 1992, 48. 6667.
131 B. W. Sullibaii. D. J. Faulkner. G . K. Matsumoto. H. Cun-heng, J. Clardy. J
SSSMc
Oix. C'ii<wi. 1986. 51. 4568.
[4] n) P. S. Sai-in. D. Sun. A. Thornton, W. E. G. Miiller. JNCI. .l Nu//. Cuncer
/ m / . 1987. 78. 633: h) S . Loya, A. Hiri. FEES Let,. 1990. 269, 131.
[ 5 ] J. C. Swerse), L. R. Bgirrows. C. M. Ireland, Tetruhrdron Let/. 1991. 6687.
[6]Sec tbotnote 3 of ref. [Zc].
[7] K Alder. G . Stein. A i i j i r i i . . Chem. 1937, 50. 510.
[XI " I i i i ~ ~ r i i i o l ~ ~ ~Diu1.s-
i r l t , , . A k l w Rcuciion~":W. Oppolzer in in Ciiiiipre/i~,[Link]~,~,
Or-
, q [ i i i i < ' .Siw/h.\i\. b?~/. 5 (Ed.: B. M. Trost): Pergamon. London. 1991. Chapt.
4.1
[9] R B Woodwiird. T. J Katz. rcrruhrt/ron 1959. 5. 70.
[lo] .I (i. Martin. R. K. Hill. Chriii. Rc>r. 1961, 61. 537.
[l I ] :ii L: Kobuke. T [Link]. J. Fiirukawa. J A m . Chtwi. So1 1970, 97. 6548: for
more recent examples employing Lewis acid catalysts. see b j F. Ribiere, 0.
Ri:int. H. B. K a p n . f i , m i / i ~ v / r mA . ~ ~ m ? i i w t i1990,
:i~ 1, 199, c j K. Furuta, S. 1
Shiinizu. Y. Miwa. H. Ydmamolo. J Org. C/iei?i.1989.54. 1483; d j H. Takemu-
rii. N. Komeshinxi. I Takahashi, S. Hashimoto. N. Ikota. K. Tomioka. K. '+Mon OH
Kogi. E3frulirdronLer/. 1987, 5687; ej M. Reetz, S. Kyung. C. Bolm, T. Zierke,
Ciiciii Itid. Loridoii 1986. 834.
[ I ? ] The concept of a specific oid(i-orienting preference of methyl groups by a
(5 interaction lids been suggested [l la]. However. the explanation given in the

text. M hich is based on steric interactions. seems more likely in the light of the
f i i c t that ii similar E.w preference is exhibited by z-haloacrylic acids (see [lo]). Calicheamicin yi reacts with supercoiled phage D N A to pro-
Alw. quite recently. Lewis acid controlled variation ofent/o:c.w selectivity in duce a ratio of double- to single-stranded cuts of 1:2.''' The
ii irtero-Diels-Alder reaction has been reported: L. F. Tietze. C. Schneider.
S i i i l i ~ i i1992. 755.
1131 K Mikaini, M. Tei-ada. Y Motoyama, T. Nakai. 7 i ~ i r d i ~ d r oAws y m i w i r v 1991.
[*I Prof. S.J. Danishefsky!" Dr. J. Aiyar. Dr. S . A. Hitchcock. Dr. D Denhart,
Dr. K. K. C . Liu. Prof. D. M. Crothers
647.
Department of Chemistry, Yde University
1141 E. J. Core). M . C . Desai. Tefruheiiron LPII.1985, 5747.
New Haven. CT 0651 1 (USA)
[15] F. Beiiingtoii. R. D . Morin, L. C Clark, Jr., J. Org. C k m . 1955. XJ- 102.
Telefax: Int. code + (203)432-5098
1161 S. N d i m . S. M. Weinreb, Tetrul~erlronL e i / 1981. 3815.
1171 S. P. Tani\. Y. M. Abdallah, Si.n/h. Coininun. 1986, 251. [ '1 New addresses: Memorial Sloan-Kettering Cancer Center
[lX] X-ray crystal structure analysis of compound 8: C,,H,,O,, triclinlc, space 1275 York Avenue, New York. NY 10021 (USA)
g r o u p d ( n o . ? ) . o =9.9014(6).h =10.322(1),c =12.188(1)A,a = 85.161(7). Telefax: [Link] + (212)772-8691
/{ = 69.095i6). 7 = 87.714(7) . I; =1159.4(4) A'. Z = 2. = 1.193 gem-'. and
/!(MokJ = 0.8 cm-! Of the 4548 reflections collected, 4294 were unique and Department of Chemistry. Havemeyer Hall
2526 reflections uith I > 3 n ( I ) were used in the refinement of the structure; Columbia University. New York, N Y 10027 (USA)
R = 0 045. R \ 4 = 0.055 The residual electron density was less than 0.15 e k J . Telefax: Int. code + (212)854-7142
Eiiral'-Nonius CAD4 diffractoineter. Mo,, irradiation, 2H,,,, = 50 , sturctiire [**I This rescarch was supported by the National Iiistitutes of Health (Grants No.
deterinination by direct methods (MITHRIL). all hydrogen atoms were in- CA 28824 and GM-21966) A NATOjSERC (Science and Engineering Re-
cluded in calctdnted positions. cxcept those belonging to methyl groups in search Council (UK)) Fellowship to S A. H.. a National Research Council
which case one hydrogcn atom of the methyl group w a s located in the differ- (Canada) Predoctoral Fellowship to D D.