Synthesis of bis-18,18′-desmethyl ritterazine N

Article abstract of DOI:10.1021/jo800454w

(Chemical Equation Presented) Bis-18,18′-desmethyl ritterazine N has been prepared in enantiomerically pure form. The synthetic alkaloid, lacking only two of the 52 carbon atoms of the natural product, shows selective activity in the NIH 60 cell panel.

Full text of DOI:10.1021/jo800454w

Synthesis of Bis-18,18′-desmethyl Ritterazine N
Douglass F. Taber* and Jean-Michel Joerger
Department of Chemistry and Biochemistry, UniVersity of Delaware, Newark, Delaware 19716.
taberdf@udel.edu
ReceiVed February 26, 2008
Bis-18,18′-desmethyl ritterazine N has been prepared in enantiomerically pure form. The synthetic alkaloid,
lacking only two of the 52 carbon atoms of the natural product, shows selective activity in the NIH 60
cell panel.
Introduction
The ritterazines (represented by ritterazine N 1), found in
small quantities in the lipophilic extract of the tunicate Ritterella
tokioka, induce apoptosis in apoptosis-resistant malignant cells.1
With the closely related cephalostatins, which show the same
activity, they form a unique class of trisdecacyclic molecules
featuring a pyrazine as the core ring, steroid-related structures,
and spiroketal edge-rings (E and F). Partial syntheses from
steroid precursors of several of the 6-6-6-5 cephalostatins and
derivatives have been accomplished.2 There has been no report
of efforts other than our own toward the 6-6-5-5 ritterazines.
In preceding papers, we described3 the preparation of the
A-B-C carbocyclic core 2 of these ritterazines, and the
preparation4 of the 5/5-spiroketal 3 (rings E and F, rings E′ and
F′). We describe here the assembly of bis-18,18′-desmethyl
ritterazine N 4, and an initial report of the activity of 4 in the
NCI 60 cell panel (Figure 1).
Results and Discussion
FIGURE 1. Structures of ritterazines and precursors.
Preparation of the Enantiomerically Pure Tricyclic
Ketone. The ketone 2 we had previously prepared was
racemic. To prepare the enantiomerically pure ketone 8, we
exposed racemic 2 to mandelic acid and N-bromosuccinimide
in the presence of 2,6-lutidine (Scheme 1). As expected,5
just two diastereomeric bromomandelates were formed, the
product of Br+ complexation to the more accessible face of
the alkene followed by diaxial opening with mandelate anion.
The diastereomeric mandelates were separated by
column chromatography. The structures were assigned by 1H
NMR analysis, following our earlier precedent.5 This as-
signment was confirmed by X-ray analysis of the mesylate
11 (Scheme 2).
(1) For the isolation and activity of the ritterazines and cephalostatins, see:
(a) Fusetani, S.; Fukuzawa, S.; Matsunaga, S. J. Org. Chem. 1997, 62, 4484.
(b) Pettit, G. R.; Tan, R.; Xu, J.; Ichihara, Y.; Williams, M. D.; Boyd, M. R. J.
Nat. Prod. 1998, 61, 955. (c) Komiya, T.; Fusetani, N.; Matsunaga, S.; Kubo,
A.; Kaye, F. J.; Kelley, M. J.; Tamura, K.; Yoshida, M.; Fukuoka, M.; Nakagawa,
K. Cancer Chemother. Pharmacol. 2003, 51, 202. (d) Lopez-Anton, N.; Rudy,
A.; Barth, N.; Schmitz, L. M.; Pettit, G. R.; Schulze-Osthoff, K.; Dirsch, V. M.;
Vollmar, A. M. J. Biol. Chem. 2006, 281, 33078.
(2) For leading references, see: (a) Lee, J. S.; Fuchs, P. L. J. Am. Chem.
Soc. 2005, 127, 13122. (b) Nawasreh, M.; Winterfeldt, E. Curr. Org. Chem.
2003, 7, 649.
(3) Taber, D. F.; Taluskie, K. V. J. Org. Chem. 2006, 71, 2797.
(4) (a) Taber, D. F.; Joerger, J.-M. J. Org. Chem. 2007, 72, 3454. (b) While
our work was in press, Shair reported reaching the same conclusion about the
relative configuration of the spiroketal: Phillips, S. T.; Shair, M. D. J. Am. Chem.
Soc. 2007, 129, 6589.
Saponification of the bromomandelate 6 led directly to the
ꢀ (“up”) epoxide 8. We had previously shown3 that the
10.1021/jo800454w CCC: $40.75
Published on Web 05/08/2008
2008 American Chemical Society
J. Org. Chem. 2008, 73, 4155–4159 4155

Taber and Joerger
SCHEME 1
SCHEME 2
SCHEME 3
SCHEME 4
analogous R epoxide, from direct epoxidation of 2, could be
dimerized to 9 by opening with azide, oxidation to the ketone,
and reduction with Te/NaBH4. To our surprise, only traces
of 9 could be found when the same reductive protocol was
applied to the azido ketone prepared from 8. While it might
be possible to find conditions for the dimerization from the
ꢀ epoxide, we chose to convert the ꢀ epoxide to the easily
dimerized R epoxide before proceeding with the synthesis.
Preparation of the Spiroketal Triflate. The spiroketal 3
that we had previously prepared4a was converted to the triflate
10 by desilylation followed by sulfonylation with triflic
anhydride (eq 1). We also prepared the iodide corresponding
to 10, but this proved to be a less efficient alkylating agent
than 10.
Preparation and Alkylation of the Epoxy Ketone. The
epoxide of 8 was inverted by opening5 with 4-methoxyphenol,
followed by mesylation, to give 11. The mesylate 11 gave
crystals that were suitable for X-ray analysis, confirming the
previously assigned absolute configuration. Oxidative removal
of the aryl ether followed by cyclization then delivered the R
epoxide 12.
The alkylation of 12 was challenging. We anticipated that
we could arrive at 13 by kinetic deprotonation of the more
accessible methylene of 12. In the event, the lithium enolate,
prepared by exposing 12 to LDA, was not sufficiently reactive
toward 10, even at room temperature and above. We eventually
found that exposure of 12 to KH, conveniently delivered as KH
4156 J. Org. Chem. Vol. 73, No. 11, 2008

Synthesis of Bis-18,18′-desmethyl Ritterazine N
TABLE 1. Activity of 1 in the National Cancer Institute 60 Cell Panel
in paraffin,6 generated an enolate that reacted nearly quantita-
tively with the triflate 10.
Successfully reacting 12 with 10 was not the end of the
difficulties. The product was a mixture of both regioisomeric
C-alkylation products and also enol ethers from O-alkylation.
It was necessary to develop conditions for acidic hydrolysis of
the O-alkylated byproducts without upsetting the acid-sensitive
spiroketal. We found success by stirring the crude alkylated
mixture with CDCl3 (nonstabilized chloroform) containing a
little bit of aqueous HCl. The regenerated 12 could then be
separated from the alkylated product 13 and from the alkylated
regioisomer by column chromatography.
The Aldol Condensation Fails. We had originally envisoned
(Scheme 3) that the diketone 14 could cyclize to the aldol
J. Org. Chem. Vol. 73, No. 11, 2008 4157

Taber and Joerger
44.7, 39.0, 37.8, 34.6, 33.2, 32.1, 32.0, 29.0, 28.8, d 136.5, 75.0,
53.3, 52.4, 52.0, 50.5, 47.5, 43.4, 36.4, 36.0, 30.1, 28.3, 12.8, 11.2;
IR (cm-1) 2967, 2921, 1739, 999; MS m/z (%) 451 (M + Na, 100),
243 (6); HRMS calcd for C27H40O4Na (M + Na) 451.2824, obsd
451.2815; [R]D +22 (c 0.36, CH2Cl2).
product 1. In the event, through a range of bases and solvents,
we were not able to detect 1 in the crude reaction mixtures.
Rather, the product appeared to be predominantly the cyclo-
heptenone 15. We are investigating alternative strategies for the
preparation of 1.
Preparation of Bis-18,18′-desmethyl Ritterazine N 4. In
the course of our investigations, we prepared (Scheme 4) the
dimerized pyrazine 17. Diaxial opening of 13 with sodium azide
delivered the alcohol 16. The ketone from the oxidation of 16
was not stable, so we submitted it directly to dimerization
conditions,7 to give 17.
We were pleased to observe that ozonolysis followed by brief
exposure to base led to clean aldol condensation, to deliver bis-
18,18′-desmethyl ritterazine N 4 as a single diastereomer.8 As
an interim step in our investigations, we submitted the synthetic
4 for analysis in the NIH 60-cell screen. In fact, 4 did show
(Table 1) modest differential activity across the several cell lines.
It would be interesting to compare these data to those for
ritterazine N 1 itself. Unfortunately, that substance is not
currently available.
Azido Alcohol 16. In a sealable tube were combined epoxide
13 (20.0 mg, 46.6 µmol), sodium azide (33 mg, 500 µmol), and a
methanol/water solution (8:1, 1 mL). The tube was sealed and
heated. After 5 h at 102 °C, the reaction mixture was cooled, then
partitioned between EtOAc, and sequentially, water and brine. The
organic extract was dried (Na2SO4) and concentrated. The residue
was chromatographed to give the azido alcohol 16 as a colorless
1
oil (19.6 mg, 89%). TLC Rf (MTBE/PE ) 40:60) 0.46; H NMR
δ 0.87 (d, J ) 6.7 Hz, 3H), 1.03 (s, 3H), 1.15 (s, 3H), 1.19-1.44
(m, 7H) includes 1.33 (s, 3H), 1.47-2.14 (m, 17H), 2.20-2.28
(m, 1H), 2.72 (m, 1H), 3.77 (m, 1H), 3.92 (m, 1H), 4.38 (m, 1H),
5.05-5.13 (m, 2H), 5.65 (m, 1H); 13C NMR δ u 219.5, 117.7,
115.3, 81.7, 44.7, 37.8, 36.8, 36.7, 33.2, 32.3, 32.1, 31.6, 28.5, d
136.4, 75.1, 68.4, 61.3, 54.3, 52.0, 47.6, 43.5, 38.7, 35.4, 30.1, 28.3,
12.3, 11.2; IR (cm-1) 3439 (br), 2923, 2102, 1734, 1000; MS m/z
(%) 494 (M + Na, 100); HRMS calcd for C27H41N3O4Na (M +
Na) 494.2995, obsd 494.2985; [R]D -35 (c 0.70, CH2Cl2).
Pyrazine 17. To a solution of azido alcohol 16 (14.0 mg, 39.7
µmol) in CH2Cl2 (1 mL) was added Dess-Martin periodinane (85
mg, 200 µmol) at rt. After 2 h at rt, a 2 mL mixture of saturated
aqueous NaHCO3, saturated aqueous Na2S2O3, and water (1:1:1)
was added to the reaction mixture. The resulting mixture was stirred
for an additional 5 min, then was partitioned between MTBE and
brine. The organic extract was dried (Na2SO4) and concentrated to
crude azido ketone. A solution of NaTeH (approximatively 0.25
M) was prepared by heating powdered tellurium (510 mg, 4 mmol)
and NaBH4 (378 mg, 10 mmol) to 75 °C in ethanol (16 mL) for
1 h. A 0.033 M solution of NaTeH was prepared by adding 4 mL
of the 0.25 M solution to 26.3 mL of ethanol thoroughly degassed
with N2. To the crude azido ketone was added 2 mL (66 µmol) of
the freshly prepared 0.033 M solution of NaTeH, and the resulting
suspension was stirred for 1 h at rt under N2, then overnight at rt
under O2. The mixture was partitioned between CH2Cl2 and brine.
The organic extract was dried (Na2SO4) and concentrated. The
residue was chromatographed to give the pyrazine 17 as a white
solid (3.4 mg, 27%). Mp 244-246 °C dec; TLC Rf (MTBE/PE )
Conclusion
We are pleased that we were able to prepare practical
quantities of both the enantiomerically pure ketones 8 and 12,
and the triflate 10, and that we were able to alkylate the ketone
12 with the triflate 10. The capability to dispense scrupulously
dry KH in paraffin in micromole quantities was critical for the
success of this alkylation. For the first time, this makes
derivatives such as 4, having the full ring framework of the
6-6-5-5 ritterazines, available for further evaluation.
Experimental Section
Alkylated Ketone 13. To ketone 12 (234 mg, 1.06 mmol),
azeotropically dried with toluene, was added KH (128 mg of 50%
w/w mixture of KH/paraffin, 1.59 mmol of KH)7 and THF (8 mL).
After 1 h at rt, the triflate 10 (190 mg, 0.530 mmol) in THF (8
mL) was added. After 30 min at rt, the reaction mixture was
partitioned between EtOAc and saturated aqueous NH4Cl. The
organic extract was dried (Na2SO4) and concentrated. To the residue
were added CDCl3 (8 mL) and 3 drops of 0.1 M aqueous HCl.
The mixture was stirred for 1.5 h, then partitioned between EtOAc,
and, sequentially, saturated aqueous NaHCO3 and brine. The organic
extract was dried (Na2SO4) and concentrated. The residue was
chromatographed to give the alkylated product 13 as a colorless
oil (20.0 mg, 9% yield based on starting triflate, 24% yield based
on recovered ketone), recovered ketone 12 (82%), and the spiro
alcohol (43%). 13: TLC Rf (MTBE/PE ) 20:80) 0.36; 1H NMR δ
0.80 (s, 3H), 0.86 (d, J ) 6.7 Hz, 3H), 1.10-1.36 (m, 8H) includes
{1.14 (s, 3H), 1.31 (s, 3H)}, 1.42-2.12 (m, 18H), 2.23 (dd, J )
16.8, 4.9 Hz, 1H), 2.73 (m, 1H), 3.10 (dd, J ) 5.8, 4.1 Hz, 1H),
3.17-3.21 (m, 1H), 4.37 (m, 1H), 5.03-5.13 (m, 2H), 5.65 (dt, J
) 16.9, 9.8 Hz, 1H); 13C NMR δ ꢀ9 u 219.7, 117.6, 115.2, 81.6,
1
50:50) 0.46; H NMR δ 0.83 (s, 6H), 0.87 (d, J ) 6.7 Hz, 6H),
1.14 (s, 6H), 1.23-1.51 (m, 10H) includes 1.33 (s, 6H), 1.64-2.25
(m, 28H), 2.31 (dd, J ) 16.4, J ) 6.5 Hz, 2H), 2.58 (dd, J ) 17.7,
12.3 Hz, 2H), 2.66-2.93 (m, 8H), 4.42 (m, 2H), 5.06-5.15 (m,
4H), 5.70 (m, 2H); 13C NMR δ u 219.4, 148.5, 148.4, 117.8, 115.3,
81.6, 46.1, 44.6, 37.8, 36.4, 35.3, 33.1, 32.0, 31.7, 29.1, d 136.2,
74.9, 53.2, 52.0, 47.7, 43.5, 41.9, 36.0, 30.1, 28.3, 11.7, 11.2; IR
(cm-1) 2967, 2924, 1739, 1002; MS m/z (%) 872 (M + Na), 447
(29), 399 (63); HRMS calcd for C54H76N2O6Na (M + Na) 871.5601,
obsd 871.5596; [R]D +16 (c 0.17, CH2Cl2); UV (MTBE) λmax 289
(ꢀ 11600).
Bis-18,18′-desmethylritterazine N 4. To bisalkene 14 (3.4 mg,
4.0 µmol) was added 0.1 mg of Sudan-III and 2 mL of CH2Cl2,
and ozone was passed through the solution at -78 °C until the
color turned from red to yellow. Excess ozone was removed by
bubbling nitrogen through the solution. PPh3 (10 mg, 40 µmol)
was added, the solution was warmed to rt, and the solvent was
evaporated. To the residue was added 2 mL of the upper phase
from a mixture of {THF (3 volumes) + ethanol (3 volumes) + 10
wt % aqueous NaOH (1 volume)}. The mixture was stirred at rt
for 1.5 h, then partitioned between EtOAc, and sequentially, half-
saturated aqueous NH4Cl and brine. The organic extract was dried
(Na2SO4) and concentrated. The residue was chromatographed to
give the desmethylritterazine N 4 as a colorless oil (3.3 mg, 97%).
(5) Taber, D. F.; Liang, J. J. Org. Chem. 2007, 72, 4313.
(6) For the advantages of KH in paraffin, see: Taber, D. F.; Nelson, C. G. J.
Org. Chem. 2006, 71, 8973.
(7) (a) Suzuki, H.; Kawaguchi, T.; Takaoka, K. Bull. Chem. Soc. Jpn. 1986,
59, 665. (b) Jeong, J. U.; Sutton, S. C.; Kim, S.; Fuchs, P. L. J. Am. Chem. Soc.
1995, 117, 10157.
(8) We have not yet had sufficient material for X-ray analysis of the synthetic
4. The structure is assigned on the basis of aldol addition to the more open face
of the cyclopentanone, to give the diastereomer of the secondary alcohol that
can hydrogen bond to the ketone.
1
TLC Rf (EtOAc) 0.45; H NMR δ 0.99 (s, 6H), 1.10 (d, J ) 6.7
Hz, 6H), 1.18 (s, 6H), 1.22-1.38 (m, 8H) includes 1.34 (s, 6H),
1.38-1.52 (m, 2H), 1.52-1.63 (m, 2H), 1.66-2.17 (m, 26H),
2.42-2.70 (m, 6H), 2.86 (dd, J ) 18.0, 5.5 Hz, 2H), 2.96 (d, J )
(9) 13C multiplicities were determined with the aid of a JVERT pulse
sequence, differentiating the signals for methyl and methine carbons as “d” from
methylene and quaternary carbons as “u”.
4158 J. Org. Chem. Vol. 73, No. 11, 2008

Synthesis of Bis-18,18′-desmethyl Ritterazine N
16.1 Hz, 2H), 3.06 (d, J ) 16.2, 2H), 4.19 (m, 2H), 4.72 (m, 2H);
13C NMR δ u 220.6, 148.3, 148.3, 120.0, 82.0, 68.2, 45.7, 45.7,
37.5, 37.3, 35.4, 34.9, 33.3, 32.8, 28.7, d 84.3, 77.9, 56.4, 55.5,
47.2, 42.9, 33.6, 30.0, 28.5, 13.6, 13.4; IR (cm-1) 3411 (br), 2966,
2925, 1731, 1400, 999; MS m/z (%) 876 (M + Na, 30), 579 (100);
HRMS calcd for C52H72N2O8Na (M + Na) 875.5186, obsd
875.5186; [R]D +52 (c 0.16, CH2Cl2); UV (MTBE) λmax 289.5
(ꢀ 12800).
assistance, Glenn Yap for the X-ray structures, and the NIH
(GM 60287) for financial support of this work. We thank John
A. Beutler of NCI for the 60 cell results.
Supporting Information Available: General experimental
procedures, experimental procedures,1H and 13C spectra, and
X-ray data for11. This material is available free of charge via
the Internet at http://pubs.acs.org.
Acknowledgment. We thank John Dykins for recording mass
spectra (supported by NSF 0541775), Steve Bai for NMR
JO800454W
J. Org. Chem. Vol. 73, No. 11, 2008 4159