Good timing in total synthesis: The case of phoslactomycin A

Article abstract of DOI:10.1002/chem.201000104

The importance for the right order of functional group introduction and manipulation (good timing) was demonstrated in the course of a total synthesis of phoslactomycin A. The synthetic strategy comprised a Cu^Ithiophene carboxylate (CuTC, Liebeskind's reagent)-mediated coupling to introduce the Z, Z-diene at the final stage of the synthesis in the presence of a protected phosphate. Key features for the assembly of the C1-C13 fragment were an asymmetric dihydroxylation, an Evans-aldol reaction and an advanced protective group strategy. The C14-C21 fragment was accessible via an asymmetric 1, 2-addition to cyclohexenone and a subsequent diastereoselective ketone reduction. One crucial task was the dihydroxylation of the C8-C9 alkene, the introduction of the C6-C7 double bond and the generation of the C25-nitrogen functionality. A second example consisted of the best sequence for the generation of the functional groups in the core part (first phosphorylation, second iodo-olefination, third azide/carbamate conversion). The synthetic solutions from this approach are compared with the already existing contributions in the phoslactomycin area.

Full text of DOI:10.1002/chem.201000104

DOI: 10.1002/chem.201000104
Good Timing in Total Synthesis: ACTHNUGRTNEUNGThe Case of Phoslactomycin A
Bjçrn Gebhardt, Christian M. Kçnig, Cornelia Schleth, Mario Dauber, and
Ulrich Koert*[a]
Abstract: The importance for the right
order of functional group introduction
and manipulation (good timing) was
demonstrated in the course of a total
synthesis of phoslactomycin A. The
synthetic strategy comprised a CuI–
thiophene carboxylate (CuTC, Liebe-
skindꢀs reagent)-mediated coupling to
introduce the Z,Z-diene at the final
stage of the synthesis in the presence
of a protected phosphate. Key features
for the assembly of the C1–C13 frag-
ment were an asymmetric dihydroxyla-
tion, an Evans-aldol reaction and an
advanced protective group strategy.
The C14–C21 fragment was accessible
via an asymmetric 1,2-addition to
the C6–C7 double bond and the gener-
ation of the C25-nitrogen functionality.
A second example consisted of the best
sequence for the generation of the
functional groups in the core part (first
phosphorylation, second iodo-olefina-
tion, third azide/carbamate conver-
sion). The synthetic solutions from this
approach are compared with the al-
ready existing contributions in the
phoslactomycin area.
AHCTUNGTRENGNcUN yclohexenone and a subsequent dia-
stereoselective ketone reduction. One
crucial task was the dihydroxylation of
the C8–C9 alkene, the introduction of
Keywords: cross-coupling · natural
products · phoslactomycin · total
synthesis
Introduction
ble biological activities (Figure 1). Phoslactomycins A–F
were isolated from Streptomyces nigrescens.[3] Phoslactomy-
cin B (phospholine) was also found in Streptomyces hydro-
scopicus.[4] Leustroducsins A–H, which differ from the phos-
lactomycins in the C16 side chain of the cyclohexyl fragment
were isolated from Streptomyces platensis.[5] Fostriecin was
the first compound of this class of isolated natural prod-
ucts[6] and therefore most intensively studied.[7] Cytostatin[8]
and sultriecin[9] have a similar structure and biological pro-
file.
The biological activities of the phoslactomycines and the
related compounds are of pharmaceutical interest (antitu-
mor, antibacterial and antifungal) and have been connected
with a selective human protein phosphatase 2A (PP2A) in-
hibition.[10] SAR studies showed the importance of the unsa-
turated lactone.[11]
The total synthesis of natural products forms a main chal-
lenge for synthetic chemistry. A good synthetic strategy is
one key factor for a successful synthesis.[1] Of equal impor-
tance are timing and tactics.[2] A target structure with a mul-
titude of functional groups can be assembled by different se-
quences of functional group introduction. It is important to
choose the right timing for the introduction of functional
groups. The right order of functional group manipulations
including protective operations often determines success or
failure. The phoslactomycins bear a large number of differ-
ent functional groups. Efforts towards the total synthesis of
this class of natural products are a valuable case study for
timing in total synthesis. The phoslactomycins, leustroduc-
sins form together with fostriecin, cytostatin and sultriecin a
class of structurally similar natural products with compara-
Two total syntheses of leustroducsin B by Fukuyama[12]
and by Iminashi[13] as well as two successful routes to phos-
lactomycin B by Kobayashi[14] and by Hatakeyama[15] have
been reported. Cossy has achieved formal total synthesis of
leustroducsin B[16] and phoslactomycin B.[17] Kobayashi et al.
has published an improved synthesis of the C7–C13 frag-
ment of phoslactomycin B.[18] All phoslactomycin syntheses
reported thus far gained benefit from the numerous synthe-
ses of fostriecin.[19] Here, we focus in a comparative way on
the problems of timing in the synthesis for phoslactomycins
[a] Dr. B. Gebhardt, Dr. C. M. Kçnig, Dr. C. Schleth, M. Dauber,
Prof. Dr. U. Koert
Fachbereich Chemie, Philipps-Universitꢁt Marburg
Hans-Meerwein-Strasse, 35032 Marburg (Germany)
Fax : (+49)6421-28-25677
E-mail: koert@chemie.uni-marburg.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000104.
5934
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5934 – 5941

FULL PAPER
Figure 1. Structures of phoslactomycins and related natural products.
Scheme 1. Comparison of the synthetic strategies to phoslactomycins.
and report the details of our total synthesis of phoslactomy-
cin A.[20]
for C11, a stereoselective dihydroxylation of a trisubstituted
ꢀ
alkene was selected for the C8 C9 diol function. Three dif-
ferent alkenes 12, 14, and 16 were synthesized (Scheme 2)
and evaluated (Scheme 3). S-glycidol-PMB ether 8[21] was
transformed via a CuI-mediated epoxide opening and a sub-
sequent TIPS-protection into the alkene 9. Oxidative cleav-
age of the double bond gave aldehyde 10 which led with
ylide 11[22] in an E-selective Wittig reaction to alkene 12.
Opening of the g-lactone via the hydroxy carboxylic acid 13
gave ethyl ester 14. Compound 13 could also be transformed
into hydroxyl ester 15, which delivered azide 16 under Mit-
sunobu conditions.[23]
Results and Discussion
A comparative retrosynthetic consideration of the five pub-
lished routes to phoslactomycins exhibits two conceptionally
different approaches (Scheme 1).
The first route chosen by Kobayashi and Fukuyama con-
nects both fragments with simultaneous generation of
stereo
stereoACHTUNGTRENNUNG
ACHTUNGTRENNUNG
reaction (2 + 3) while Fukuyama adds alkynyl zinc bromide
5 to aldehyde 4. The introduction of the amino group and
the phosphate is postponed in both approaches past the con-
vergent skeleton construction step.
The Sharpless dihydroxylation[24] of trisubstituted alkenes
12, 14 and 16 proceeded with remarkable different dia-
ACHUTNGRENsNUG tereoHCATUNGTRENNsUGN electivities (Scheme 3). A low diastereoselectivity of
2:1 was observed for the lactone case (12!17), while a 9:1
selectivity could be achieved with the ethyl ester (14!19).
The major stereoisomers could be obtained in pure form at
the stage of acetals 18 and 20. An excellent stereoselectivity
(17:1) was observed for the dihydroxylation of the azide-
substituted alkene 16. Diol 21 was accessible in pure form
by chromatography and could be converted into acetal 22.
Compound 22 is due to the presence of the nitrogen func-
tionality a more advanced synthetic intermediate than TBS
ether 20. However, problems during the subsequent intro-
duction of the C6/7 double bound (triazoline formation, see
below) disfavored further use of 22. In terms of timing, it
was advantageous to position the dihydroxylation step prior
The second approach relies on a cross-coupling reaction
to prepare the Z,Z-diene. An alkenyl iodide 6 and a stan-
nane 7 were used by three groups (Miyashita/Imanishi, Ha-
takeyama, Koert). Here all stereocenters are introduced sep-
arately in earlier steps. The disadvantage of these additional
steps is paid off by a higher convergency. Miyashita/Imanishi
and Hatakeyama install the amino group before the cou-
pling step. We achieved a maximum of convergency with the
amino group and the phosphate already present in the cou-
pling step.
In the present case the synthesis of the C7–C12 phoslacto-
mycin core structure with the three stereocenters C8, C9
and C11 was investigated first. With a chiral pool solution
ꢀ
to the introduction of the C6 C7 double bond and the nitro-
Chem. Eur. J. 2010, 16, 5934 – 5941
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5935

U. Koert et al.
Scheme 4. Miyashiata/Imanishi
dihydroxylation.
a) K2OsO2(OH)4,
(DHQD)2PHAL, tBuOH/H2O.
(Scheme 5). The conversion of the C6 ester into the corre-
sponding aldehyde via LiBH4 reduction and Dess–Martin
was applied to compounds 20 and 22. The ethyl ester 20 led
smoothly to aldehyde 25, which delivered alkene 27 in very
good yield. The same sequence applied to azido ethyl ester
22 led to a dead end. The conversion into the aldehyde 26
proceeded with a lower yield. More important, the Wittig
product 28 could not been isolated, because a subsequent in-
tramolecular 1,3-dipolar cycloaddition gave triazoline 29.
The electron-deficient double bond in 28 exhibits a high di-
polarophile reactivity. Thus, if the nitrogen functionality
shall be introduced via an azide, no electron-withdrawing
Scheme 2. Synthesis of the alkenes 12, 14,and 16. a) i) H2C=CHMgBr,
CuI, THF, ii) TIPSOTf, 2,6-lutidine, CH2Cl2; b) i) K2OsO2(OH)4, NMO,
acetone/H2O, ii) NaIO4, THF/H2O; c) THF, 508C; d) LiOH, THF/H2O;
e) i) TBSCl, imidazole, CH2Cl2, ii) EtI, K2CO3, actone/H2O; f) EtI,
K2CO3, actone/H2O; g) Zn(N3)2·2py, PPh3, DIAD, toluene.
ꢀ
group has to be present at the C6 C7 double bond. An at-
tempt to reduce the azide into an amine led to a different
problem. The Staudinger reduction of azide 22 gave com-
pound 30 after TEOC protection. The latter was reduced to
the alcohol and subsequently oxidized to the aldehyde stage.
No free aldehyde was detected, but complete hemiaminal
formation to 31 was observed. Attempts to use the hemiami-
nal for the following Wittig reaction were not successful.
With the identification of ester 27 as the most suitable in-
termediate, the completion of the C1–C6 substructure was
addressed next (Scheme 6). DIBAH reduction to the alcohol
and MnO2 oxidation gave aldehyde 32. Subsequent Evans-
aldol reaction with oxazolidinone 33 to syn-aldol 34 pro-
ceeded with 95:5 diastereoselectivity and very good yield.[25]
By this route the stereocenters C4 and C5 were introduced.
A related aldol solution for both stereocenters was used by
Kobayashi in his synthesis of phoslactomycin B.[14] The sub-
sequent protection of the secondary hydroxyl group in 34 re-
quired a silyl ether, that could be selectively deprotected in
the presence of the secondary TIPS group at HO-C11 to
allow the ring closure to the six-ring lactone. TES should
fulfill this criterion and therefore compound 34 was protect-
ed as a TES ether. The removal of the Evans auxiliary was
planned at the aldehyde oxidation state. Towards this end
the route via a thiol ester was chosen. Treatment of TES-
protected 34 with lithium ethyl thiolate gave thiol ester
35.[26] The high nucleophilicity of the sulfur allows removal
of the auxiliary at 08C which prevents epimerization. Thiol
ester 35 could be reduced to aldehyde 36 in 94% yield. The
Weinreb-amide alternative from 34 to 36 gave a considera-
ble amount of epimerization and was therefore disfa-
gen functionality. TBS ether 20 proved to be the most suita-
ble synthetic intermediate.
Miyashita and Imanishi used in their synthesis the dihy-
droxylation of alkene 23 with a different oxidation state at
C7 and obtained diol 24 as
a single stereoisomer
(Scheme 4).[13]
The next chapter of our synthesis focused on the elabora-
tion of the C1–C6 substructure. An E-selective Wittig olefi-
ꢀ
nation was selected to introduce the C6 C7 double bond
vored.[27] The following introduction of the C3 C4 double
ꢀ
bond was achieved using (PhO)2POCH2CO2Et and
NaHMDS.[28] A little excess of base to phosphonate and
HMPA as a cosolvent prevented epimerization of the alde-
Scheme 3. Synthesis of the alkenes 12, 14,and 16. a) K2OsO2(OH)4,
(DHQD)2PHAL for 12, (DHQD)2PYR for 14 and 16, tBuOH/H2O;
b) dimethoxypropane, CSA, CH3CN; c) DMPCHACHTUNGRTNEUNG(OMe)2, CSA, CH3CN.
5936
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5934 – 5941

Phoslactomycin A
FULL PAPER
ꢀ
Scheme 5. Introduction of the C6 C7 double bond. a) i) LiBH4, ii) Dess–
Martin oxidation; b) Ph3PCH=CO2Et, toluene, 708C; c) PPh3, THF/H2O,
TEOCOSu.
Scheme 6. Synthesis of the C1–C6 substructure. a) i) DIBAH, CH2Cl2,
ꢀ788C, ii) MnO2; b) nBu2BOTf, CH2Cl2, NEt3, ꢀ78 ! ꢀ208C; c) i) TE-
SOTf, 2,6-lutidine, ii) EtSH, nBuLi, THF, 08C; d) DIBAH, toluene,
ꢀ788C; e) (PhO)2P(O)CH2COOEt, NaHMDS, THF/HMPA, ꢀ788C;
f) i) (HF)3·NEt3, ii) Ti
ACHTUNGTRNE(NUNG OiPr)4, benzene, 808C; g) Zn(N3)2·2py, DIAD,
hyde and led to alkene 37 with complete Z selectivity. After
selective deprotection of the TES and the TBS ether in the
presence of the TIPS group with [HF]3·Et3N, the resulting
hydroxyl ester could be cyclized with titanium tetraisoprop-
oxide in boiling benzene[14] to lactone 38. At this stage of
the synthesis the introduction of the nitrogen functionality
via an azide-Mitsunobu reaction (38!39) could be achieved
without the undesired dipolar cycloaddition encountered in
the case 26!28!29 (Scheme 5).
PPh3, toluene, 208C.
phate introduction (44!45!43) gave a significant lower
yield and was therefore disfavored.
The problem of timing for the alkenyl iodide introduction
and azide reduction was investigated next (Scheme 8). Oxi-
dative PMB deprotection worked well in the presence of the
azide (42!46) and the corresponding carbamate (43!47).
The subsequent Dess–Martin oxidation gave aldehydes 48
and 49 for both cases in good yields. The next step, the Z
olefination[31] of the aldehyde to the alkenyl iodide gave a
completely different outcome for the azide and the carba-
mate case. Azide-containing aldehyde 48 provided the de-
sired product 50 in 76% yield, which could be converted
smoothly into carbamate 51. In contrast, the olefination of
aldehyde 49 resulted in an undesired lactamization of the
carbamate (49!52). The results from these model studies
pointed to the optimal order of first phosphorylation,
second iodo-olefination and third azide/carbamate conver-
sion.
The remaining synthetic tasks left before the final cross-
coupling were the introduction of the C9 phosphate, the
azide/amine reduction and the generation of the C12–C13
alkenyl iodide. The right timing for the phosphate introduc-
tion/azide reduction was evaluated at the test system 21
which lacks the C1–C5 lactone (Scheme 7). TMS protection
of diol 21 to bis-TMS ether 40 and subsequent selective de-
[29]
protection of the C11 silyl ether using H2SiF6 at ꢀ308C
gave alcohol 41 in excellent yield. Phosphorylation using di-
allyl phosphoroamidite followed by oxidation with tert-butyl
hydroperoxide provided the bisallyl phosphate 42.[30] After
Staudinger reduction and Alloc protection carbamate 43
was obtained. The alternative sequence of first Staudinger
reduction/Alloc protection (40!44) and subsequent phos-
With the information obtained concerning the right
timing for the introduction of the functional groups in the
Chem. Eur. J. 2010, 16, 5934 – 5941
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5937

U. Koert et al.
nyl–cupration of acetylene according to the Pulido proto-
col[32] produced cuprate 53 which gave the 1,4-addition prod-
uct rac-55 in very good yields (Scheme 9). The NaBH4 re-
duction of ketone rac-55 delivered cis-cyclohexanol rac-56
with an 18:1 diastereoselectivity. Although an asymmetric
variant[33] of the cuprate addition was examined no signifi-
cant enantioselectivity was achieved. However, the asym-
metric Rh-mediated 1,4-addition of alkenyl boronic acids
developed by Hayashi[34] could be applied successfully for
the enantioselective synthesis of the cyclohexyl fragment.
The asymmetric addition of (E)-styrylboronic acid to cyclo-
hexenone 54 gave ketone 57 (ee 94%). The subsequent
NaBH4 reduction produced cyclohexanols 58 and 59 with a
7:1 diastereoselectivity in favor of the desired cis-disubstitut-
ed product 58. Noticeably is the lower stereoselectivity for
the reduction of ketone 57 bearing the E-alkene than for
the ketone 55 with the Z-alkene. The TBS protection of 58
led to 60, which upon ozonolysis of the double bond and a
subsequent Z-selective Stork–Zhao olefination gave Z-al-
kenyl iodide 61. The latter was transformed via iodine–lithi-
um exchange into the corresponding Z-alkenyl stannane
which was desilylated to produce alcohol 62. Esterfication
with isobutyric acid gave the desired cyclohexyl fragment
63.
Scheme 7. Evaluation of the timing for phosphorylation and azide reduc-
tion: a) TMSCl, imidazole; b) H2SiF6, CH3CN, ꢀ308C; c) (iPr)2NP-
ACHTUNGTRENNUNG(OAllyl)2, tetrazole, CH3CN/CH2Cl2, then tBuOOH; d) PPh3, THF/H2O,
then NaHCO3, Alloc2O; e) pTsOH, THF/H2O, 20 8C.
Scheme 8. Evaluation of the timing for olefination and azide reduction:
a) DDQ, CH2Cl2/H2O; b) Dess–Martin periodinane, CH2Cl2, pyridine;
c) ICH2PPh3I, NaHMDS, THF, ꢀ78!08C; d) PPh3, THF/H2O, then
NaHCO3, Alloc2O.
Scheme 9. Racemic und asymmetric synthesis of the cyclohexyl fragment
63: a) (Bu3Sn)2, nBuLi, CuCN, HCCH, THF, ꢀ708C; b) NaBH4, THF/
H2O, 0 8C; c) [RhACHTUNGTRNENUG(acac)AHCTUNTGREN(NGUN C2H4)2], (R)-BINAP, (E)-PhCHCHB(OH)2, di-
oxane/H2O, reflux; d) TBSOTf, 2,6-lutidine, CH2Cl2; e) i) O3, CH2Cl2,
ꢀ788C; PPh3, ii) ICH2PPh3I, NaHMDS, THF/HMPT, ꢀ788C; f) i) tBuLi,
Et2O, ꢀ788C, Bu3SnCl; ii) TBAF, THF; g) iPrCO2H, DCC, DMAP,
CH2Cl2.
C1–C13 fragment we turned our attention to the C14–C21
fragment. Cyclohexenone 54 was chosen as starting material
for the synthesis of the cyclohexyl substructure. The stan-
5938
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5934 – 5941

Phoslactomycin A
FULL PAPER
In order to find suitable conditions for the later final
cross-coupling, the reaction of alkenyl iodide 51 with the al-
kenyl stannane 63 was investigated at this stage of the proj-
ect. Miyashita/Imanishi applied in their leustroducsin B syn-
bis-TMS ether 67. Selective deprotection of the C9-TMS
group and subsequent phosphorylation led to protected
phosphate 68. Next, the olefination sequence was opened
with an oxidative PMB-ether cleavage and a subsequent
Dess–Martin oxidation to obtain the aldehyde 69. A Stork–
Zhao olefination gave (Z)-alkenyl iodide 70. Finally, azide
70 was converted into Alloc-protected amine 71. In order to
remove the bulky C11-OTIPS group for the subsequent
cross-coupling compound 71 was treated with H2SiF6 to
obtain diol 72. The CuTC-mediated cross-coupling of alken-
yl iodide 72 and alkenyl stannane 63 proceeded successfully
thesis[13] a Stille reaction[35] ([PdCl
CHTUNGTERN(NUGN MeCN)2], RT, 1 h, 61%)
2A
in a related situation. Hatakeyama in his phoslactomycin B
synthesis also chose Stille conditions ([PdCl2A(MeCN)2], RT,
CTHUNGTRENNUNG
1 h, 46%).[15] In both cases no protected phosphate was
present at C9-OH. The easy cleavage of allyl phosphates
with a Pd0 catalyst[30] led us focus on an alternative to the
Pd0-mediatated cross-coupling. Although used not substoi-
chiometrically such as a Pd catalyst, but in stoichiometric
amounts, CuI–thiophene carboxylate (CuTC, Liebeskindꢀs
reagent)[36] is a valuable cross-coupling reagent which has
passed the late-stage test in several total syntheses.[37] How-
ever, the CuTC-mediated cross-coupling of 51 with 63 was
unsuccessful (Scheme 10). A possible explanation could be
the steric shielding of the alkenyl iodide by the bulky TIPS
group. Miyashita/Imanishi and Hatakeyama accomplished
their Pd-mediated cross-coupling without a protective group
at C11-OH.[13,15] Therefore in our case, the TIPS group in
compound 61 was removed using 25% aqueous H2SiF6 in
CH3CN at 58C, which led to simultaneous removal of the
TMS ether at C8 and the formation of compound 64. Now
the CuTC-mediated cross-coupling with alkenyl stannane 63
gave the desired Z,Z-diene 65 in 61% yield.
The synthesis of the complete C1–C13 fragment applied
the optimal order of first, phosphorylation, second, iodo-ole-
fination and third, azide/carbamate conversion from the pre-
vious studies summarized in Schemes 7 and 8. Starting point
for the introduction of the phosphate was diol 66 which was
accessible from acetal 39 or in better overall yield from pre-
cursor 38 (Scheme 11). Double TMS protection delivered
Scheme 11. Synthesis of the complete C1–C13 fragment: a) CSA, HCl,
CH2Cl2/MeOH; b) i) Zn(N3)2·2py, DIAD, PPh3, toluene, ii) TFA, MeOH;
c) TMSCL, imidazole, CH2Cl2; d) i) CSA, THF/H2O, 20 8C, ii) (iPr)2NP-
AHCNUTRTGEG(NNNU OAllyl)2, tetrazole, CH3CN/CH2Cl2, then tBuOOH; e) i) DDQ, CH2Cl2/
H2O, ii) Dess–Martin periodinane, CH2Cl2, pyridine; f) ICH2PPh3I,
NaHMDS, THF, ꢀ78!08C; g) PPh3, THF/H2O, then NaHCO3, Alloc2O;
Scheme 10. Cross-coupling studies: a) H2SiF6, CH3CN/H2O, 5 8C;
b) CuTC, NMP, 08C, 1 h.
h) H2SiF6, CH3CN/H2O; i) 63, CuTC, NMP, 08C, 1 h; j) [Pd
HCOOH, Et3N, THF, 508C.
ACHTUNGTRENNUNG(PPh3)4],
Chem. Eur. J. 2010, 16, 5934 – 5941
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5939

U. Koert et al.
[2] a) J. W. Cornforth, Aust. J. Chem. 1993, 46, 157–170; b) B. M. Trost,
Science 1983, 219, 245–250; c) R. A. Shenvi, D. P. Omalley, P. S.
Baran, Acc. Chem. Res. 2009, 42, 530–541.
[3] a) S. Fushimi, S. Nishikawa, A. Shimazu, H. Seto, J. Antibiot. 1989,
42, 1019–1025; b) S. Fushimi, K. Furihata, H. Seto, J. Antibiot. 1989,
42, 1026–1036.
[4] a) T. Ozasa, K. Suzuki, M. Sasamata, K. Tanaka, M. Kobori, S.
Kadota, K. Nagai, T. Saito, S. Watanabe, M. Iwanami, J. Antibiot.
1989, 42, 1331–1338; b) T. Ozasa, K. Tanaka, M. Sasamata, H.
Kaniwa, M. Shimizu, H. Matsumoto, M. Iwanami, J. Antibiot. 1989,
42, 1339–1343.
[5] a) T. Kohama, R. Enokita, T. Okazaki, H. Miyaoka, A. Torikata, M.
Inukai, L. Kaneko, T. Kagasaki, Y. Sakaida, A. Satoh, A. Shiraishi,
J. Antibiot. 1993, 46, 1503–1511; b) T. Kohama, T. Nakamura, T. Ki-
noshita, L. Kaneko, A. Shiraishi, J. Antibiot. 1993, 46, 1512–1519;
c) T. Shibata, S. Kurihara, K. Yoda, H. Haruyama, Tetrahedron 1995,
51, 11999–12012; d) H. Matsuhashi, K. Shimada, Tetrahedron 2002,
58, 5619–5626.
in the presence of the protected phosphate to deliver the
Z,Z-diene 73 in good yield. The final simultaneous depro-
tection at phosphorous and nitrogen required some optimi-
zation attempts. Pd0 conditions gave a fast deprotection of
the allyl phosphates but only a slow cleavage of the allyl car-
bamate. No Pd0 attack at the C5–C7 allyl substructure was
observed. Finally, using transfer hydrogenation conditions
([PdACHTUNGTRENNUNG
(PPh3)4], HCOOH, Et3N)[12] at 508C for 3 h the target
compound phoslactomycin A could be obtained. The purifi-
cation of the target compound was achieved by reversed-
phase HPLC. The spectra and analytical data of the synthet-
ic sample corresponded to those reported for the natural
material.[3]
[6] a) J. B. Tunac, B. D. Graham, W. E. Dobson, J. Antibiot. 1983, 36,
1595–1600; b) S. S. Stampwala, R. H. Bunge, T. R. Hurley, N. E.
Wilmer, A. J. Brankiewicz, C. E. Steinman, T. A. Smitka, J. C.
French, J. Antibiot. 1983, 36, 1601–1605.
[7] a) D. S. Lewy, C. M. Gauss, D. R. Soenen, D. L. Boger, Curr. Med.
Chem. 2002, 9, 2005–2032; b) M. Shibasaki, M. Kanai, Heterocycles
2005, 66, 727–741.
[8] M. Amemiya, M. Ueno, M. Osono, T. Masuda, N. Kinoshita, C.
Nishida, M. Hamada, M. Ishizuka, T. Takeuchi, J. Antibiot. 1994, 47,
536–540.
[9] H. Ohkuma, N. Nobuaki, Y. Nishiyama, T. Tsuno, J. Antibiot. 1992,
45, 1239–1249.
[10] a) T. Usui, G. Marriott, M. Inagaki, G. Swarup, H. Osada, J. Bio-
chem. 1999, 125, 960–965; b) T. Teruya, S. Simizu, N. Kanoh, H.
Osada, FEBS Lett. 2005, 579, 2463–2468.
[11] S. B. Buck, C. Hardoin, S. Ichikawa, D. R. Soenen, C. M. Gauss, I.
Hwang, M. R. Swingle, K. M. Bonness, R. E. Honkanen, D. L.
Boger, J. Am. Chem. Soc. 2003, 125, 15694–15695.
[12] K. Shimada, Y. Kaburagi, T. Fukuyama, J. Am. Chem. Soc. 2003,
125, 4048–4049.
[13] a) K. Miyashita, T. Tsunemi, T. Hosokawa, M. Ikejiri, T. Imanishi,
Tetrahedron Lett. 2007, 48, 3829–3833; b) K. Miyashita, T. Tsunemi,
T. Hosokawa, M. Ikejiri, T. Imanishi, J. Org. Chem. 2008, 73, 5360–
5370.
Conclusion
The reported total synthesis of phoslactomycin A exempli-
fies several aspects of timing in synthesis. One is the optimal
place for the stereoselective C8–C9 dihydroxylation (first),
the introduction of the C6–C7 double bond (second) and
the introduction of the C25 nitrogen function (third). The
best sequence for the generation of the functional groups in
the core part (first phosphorlyation, second iodo-olefination,
third azide/carbamate conversion) was established in a test
system prior to its application for the synthesis of the target
molecule. The C13–C14 cross-coupling strategy used in the
presence case is comparable to the work of Miyashita/Ima-
nishi and Hatakeyama.[13,15] The strategic decision to accom-
plish the cross-coupling in the presence of the protected
phosphate resulted in a higher convergence of our synthesis
(72+63!73) and in a different optimal sequence for the
partial solutions of the synthetic problem. A comparison of
the approaches by Miyashita/Imanishi, Hatakeyama and the
present one shows the high number of protective group
transformations used in all three cases. Despite the request
for a protective-group free synthesis,[38] a compound with
the high density of different functional groups like phoslac-
tomycin A needs (at the moment) also good timing with re-
spect to the use of protective groups.
[14] Y. G. Wang, R. Takeyama, Y. Kobayashi, Y. Angew. Chem. 2006,
118, 3398–3401; Angew. Chem. Int. Ed. 2006, 45, 3320–3323; Angew.
Chem. Int. Ed. 2006, 45, 3320–3323.
[15] S. Shibahara, S.; M. Fujino, Y. Tashiro, K. Takahashi, J. Ishihara, S.
Hatakeyama, Org. Lett. 2008, 10, 2139–2142; M. Fujino, Y. Tashiro,
K. Takahashi, J. Ishihara, S. Hatakeyama, Org. Lett. 2008, 10, 2139–
2142.
[16] J. Moise, R. P. Sonawane, C. Corsi, S. V. Wendeborn, S. Arseniyadis,
J. Cossy, Synlett 2008, 2617–2620.
[17] V. Druais, M. J. Hall, C. Corsi, S. V. Wendeborn, C. Meyer, J. Cossy,
Org. Lett. 2009, 11, 935–938.
[18] H. Nonaka, N. Maeda, Y. Kobayashi, Tetrahedron Lett. 2007, 48,
5601–5604.
Experimental Section
Full experimental details with complete characterizations of all new com-
pounds are given in the Supporting Information.
[19] a) D. L. Boger, S. Ichikawa, W. Zhang, J. Am. Chem. Soc. 2001, 123,
4161–4167; b) D. E. Chavez, E. N. Jacobsen, Angew. Chem. 2001,
113, 3779–3782; Angew. Chem. Int. Ed. 2001, 40, 3667–3670;
c) Y. K. Reddy, J. R. Falck, Org. Lett. 2002, 4, 969–971; d) K. Miya-
shita, M. Ikejiri, H. Kawasaki, S. Maemura, T. Imanishi, J. Am.
Chem. Soc. 2003, 125, 8238–8243; e) B. M. Trost, M. U. Frederiksen,
J. P. N. Papillon, P. E. Harrington, S. Shin, B. T. Shireman, J. Am.
Chem. Soc. 2005, 127, 3666–3667; f) Y. G. Wang, R. Takeyama, Y.
Kobayashi, Angew. Chem. 2006, 118, 3398–3401; Angew. Chem. Int.
Ed. 2006, 45, 3320–3323; g) Y. Hayashi, H. Yamaguchi, M. Toyoshi-
ma, K. Okado, T. Toyo, M. Shoji, Org. Lett. 2008, 10, 1405–1408;
h) O. Robles, F. E. McDonald, Org. Lett. 2009, 11, 5498–5501;
i) total synthesis of cytostatin: L. Bialy, H. Waldmann, Angew.
Acknowledgements
The Deutsche Forschungsgemeinschaft and the Fonds der Chemischen
Industrie are gratefully acknowledged for financial support.
[1] a) The Logic of Chemical Synthesis (Eds.: E. J. Corey, X. M. Cheng),
Wiley, New York, 1989; b) Elements of Synthesis Planning (Ed.:
R. W. Hoffmann), Springer, Berlin, 2009.
5940
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5934 – 5941

Phoslactomycin A
FULL PAPER
Chem. 2002, 114, 1819–1822; Angew. Chem. Int. Ed. 2002, 41, 1748–
1751.
[30] Y. Hayakawa, S. Wakabayashi, T. Nobori, R. Noyori, Tetrahedron
Lett. 1987, 28, 2259–2262.
[31] G. Stork, K. Zhao, Tetrahedron Lett. 1989, 30, 2173–2174.
[32] A. Barbero, P. Cuadrado, I. Fleming, A. M. Gonsalez, F. J. Pulido,
R. Rubio, J. Chem. Soc. Perkin Trans. 1 1993, 1657–1662.
[33] G. Quinkert, T. Mꢃller, A. Kçniger, O. Schultheis, B. Sickenberger,
G. Dꢃrner, Tetrahedron Lett. 1992, 33, 3469–3472.
[34] a) T. Hayashi, Synlett 2001, 0879–0887; b) T. Hayashi, K. Yamasaki,
Chem. Rev. 2003, 103, 2829–2844.
[35] J. K. Stille, Angew. Chem. 1986, 98, 504–519; Angew. Chem. Int. Ed.
Engl. 1986, 25, 508–524.
[20] C. M. Kçnig, B. Gebhardt, C. Schleth, M. Dauber, U. Koert, Org.
Lett. 2009, 11, 2728–2731.
[21] M. J. Gaunt, D. F. Hook, H. R. Tanner, S. V. Ley, Org. Lett. 2003, 5,
4815–4818.
[22] J. E. Baldwin, M. G. Moloney, A. F. Parsons, Tetrahedron 1992, 48,
9373–9384.
[23] M. C. Viaud, P. Rollin, Synthesis 1990, 130–132.
[24] H. C. Kolb, M. S. VanNieuwenzhe, K. B. Sharpless, Chem. Rev. 1994,
94, 2483–2547.
[36] G. D. Allred, L. S. Liebeskind, J. Am. Chem. Soc. 1996, 118, 2748–
2749.
[25] D. A. Evans, J. V. Nelson, E. Vogel, T. R. Taber, J. Am. Chem. Soc.
1981, 103, 3099–3111.
[37] a) H. Wehlan, M. Dauber, M. T. Mujica Fernaud, J. Schuppan, M. E.
Juarez Garcia, U. Koert, Angew. Chem. 2004, 116, 4698–4702;
Angew. Chem. Int. Ed. 2004, 43, 4597–4601; b) I. Paterson, R. Brit-
ton, O. Delgado, A. Meyer, K. G. Poullennec, Angew. Chem. 2004,
116, 4729–4733; Angew. Chem. Int. Ed. 2004, 43, 4629–4633.
[38] R. W. Hoffmann, Synthesis 2006, 3531–3541.
[26] D. A. Evans, B. W. Trotter, B. Cote, P. J. Coleman, Angew. Chem.
1997, 109, 2954–2957; Angew. Chem. Int. Ed. Engl. 1997, 36, 2741–
2744.
[27] D. A. Evans, S. L. Bender, J. Morris, J. Am. Chem. Soc. 1988, 110,
2506–2526.
[28] K. Ando, J. Org. Chem. 1997, 62, 1934–1939.
[29] A. S. Pilcher, D. K. Hill, S. J. Shimshock, R. E. Waltermire, P. De-
Shong, J. Org. Chem. 1992, 57, 2492–2495.
Received: January 15, 2010
Published online: April 15, 2010
Chem. Eur. J. 2010, 16, 5934 – 5941
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5941