S-adenosylhomocysteine analogues with the carbon-5′ and sulfur atoms replaced by a vinyl unit

Article abstract of DOI:10.1021/ol062026m

(Chemical Equation Presented) Cross-metathesis of suitably protected 5′-deoxy-5′-methyleneadenosines with racemic and chiral N-Boc-protected six-carbon amino acids bearing a terminal double bond in the presence of the Hoveyda-Grubbs catalyst gave adenosyl

Full text of DOI:10.1021/ol062026m

ORGANIC
LETTERS
2006
Vol. 8, No. 22
5093-5096
S-Adenosylhomocysteine Analogues
with the Carbon-5 and Sulfur Atoms
′
Replaced by a Vinyl Unit
Daniela Andrei and Stanislaw F. Wnuk*
Department of Chemistry and Biochemistry, Florida International UniVersity,
Miami, Florida 33199
wnuk@fiu.edu
Received August 16, 2006
ABSTRACT
Cross-metathesis of suitably protected 5
bearing a terminal double bond in the presence of the Hoveyda
double bond. Bromination with pyridinium tribromide and dehydrobromination with DBU followed by standard deprotections yielded the
-(bromo)vinyl analogue.
′
-deoxy-5
′
-methyleneadenosines with racemic and chiral N-Boc-protected six-carbon amino acids
−
Grubbs catalyst gave adenosylhomocysteine analogues with the C5′−C6
′
5′
The enzyme S-adenosyl-L-homocysteine (AdoHcy) hydrolase
(EC 3.3.1.1) effects hydrolytic cleavage of AdoHcy to
adenosine (Ado) and L-homocysteine (Hcy).1 The cellular
levels of AdoHcy and Hcy are critical because AdoHcy is a
potent feedback inhibitor of crucial transmethylation en-
zymes.1,2 Also, elevated plasma levels of Hcy in humans have
been shown to be a risk factor in coronary artery disease.3
The X-ray crystal analysis of human AdoHcy hydrolase
inactivated with 9-(dihydroxycyclopentene)adenine4a and
neplanocin A,4b as well as of AdoHcy hydrolase from rat
liver,4c established the presence of a water molecule in the
active site of the enzyme. This observation made it a high
priority to prepare analogues of AdoHcy that closely
resemble the natural substrate that binds tightly to the
enzyme. Such compounds should form “stable” complexes
with the enzyme that would help to identify key binding
groups at the active site of the enzyme that interact with the
Hcy moiety and participate in subsequent elimination and
“hydrolytic” activity steps.
On the basis of the previous finding that AdoHcy hydrolase
is able to add the enzyme-sequestered water molecule across
the 5′,6′-double bond of 5′-deoxy-5′-(halo or dihalomethyl-
ene)adenosines causing covalent binding inhibition,5,6 we
now describe the synthesis of AdoHcy analogues A (X )
H) with the 5′,6′-olefin motif incorporated in place of the
carbon-5′ and sulfur atoms (Figure 1). The analogues A or
B (X ) halogen) should be substrates for the oxidative
activity of the enzyme, and the resulting 3′-keto products
(1) (a) Yuan, C.-S.; Liu, S.; Wnuk, S. F.; Robins, M. J.; Borchardt, R.
T. In AdVances in AntiViral Drug Design; De Clercq, E., Ed.; JAI Press:
Greenwich, 1996; Vol. 2, pp 41-88. (b) Turner, M. A.; Yang, X.; Yin, D.;
Kuczera, K.; Borchardt, R. T.; Howell, P. L. Cell Biochem. Biophys. 2000,
33, 101-125. (c) Wnuk, S. F. Mini-ReV. Med. Chem. 2001, 1, 307-316.
(2) (a) Ueland, P. M. Pharmacol. ReV. 1982, 34, 223-253. (b) Chiang,
P. K. Pharmacol. Ther. 1998, 77, 115-134.
(3) (a) Nehler, M. R.; Taylor, L. M.; Porter, J. M. CardioVasc. Surgery
1997, 559-567. (b) Refsum, H.; Ueland, P. M.; Nygard, O.; Vollset, S. E.
Annu. ReV. Med. 1998, 49, 31-62. (c) Schynder, G.; Roffi, M.; Pin, R.;
Flammer, Y.; Lange, H.; Eberli, F. R.; Meier, B.; Turi, Z. G.; Hess, O. M.
N. Engl. J. Med. 2001, 345, 1593-1600.
(4) (a) Turner, M. A.; Yuan, C.-S.; Borchardt, R. T.; Hershfield, M. S.;
Smith, G. D.; Howell, P. L. Nat. Struct. Biol. 1998, 5, 369-376. (b) Yang,
X.; Hu, Y.; Yin, D. H.; Turner, M. A.; Wang, M.; Borchardt, R. T.; Howell,
P. L.; Kuczera, K.; Schowen, R. L. Biochemistry 2003, 42, 1900-1909.
(c) Hu, Y.; Komoto, J.; Huang, Y.; Gomi, T.; Ogawa, H.; Takata, Y.;
Fujioka, M.; Takusagawa, F. Biochemistry 1999, 38, 8323-8333.
(5) (a) Wnuk, S. F.; Yuan, C.-S.; Borchardt, R. T.; Balzarini, J.; De
Clercq, E.; Robins, M. J. J. Med. Chem. 1994, 37, 3579-3587. (b) Wnuk,
S. F.; Mao, Y.; Yuan, C.-S.; Borchardt, R. T.; Andrei, J.; Balzarini, J.; De
Clercq, E.; Robins, M. J. J. Med. Chem. 1998, 41, 3078-3083. (c) Yuan,
C.-S.; Wnuk, S. F.; Robins, M. J.; Borchardt, R. T. J. Biol. Chem. 1998,
273, 18191-18197.
10.1021/ol062026m CCC: $33.50
© 2006 American Chemical Society
Published on Web 10/07/2006

Scheme 1
Figure 1. S-Adenosyl-L-homocysteine and analogues with the
sulfur atom replaced by the vinyl unit.
might be substrates for the hydrolytic activity. Enzyme-
mediated addition of water might occur at C5′ or C6′ of A
or B to generate new species with hydroxyl or keto (after
â-elimination of HBr) binding sites within the enzyme. X-ray
structures of such oxidation and/or hydrolytic activity-bound
products might provide important information regarding key
residues in the protein and their interactions with substrates
(Hcy unit) and/or the sequestered water molecule.
Retrosynthetic analysis indicates that AdoHcy analogue
A (X ) H) can be prepared by construction of a new C5′-
C6′ double bond via Wittig or metathesis reactions. For
example, condensation of adenosine 5′-aldehyde with a
Wittig-type reagent or cross-metathesis between 5′-deoxy-
5′-methyleneadenosine and the appropriate amino acid-
derived terminal alkenes should give A. Because nucleoside
5′-aldehydes are unstable in the presence of strong bases
required for the generation of nonstabilized phosphorane-
Wittig reagents,7 we decided to target an AdoHcy analogue
of type A via the cross-metathesis reaction. Another pos-
sibility is Pd-catalyzed cross-coupling approaches between
sp2 and sp3 hybridized carbons to form a new C6′-C7′ single
bond as a key step.8,9
Alkylation of protected glycine 1 with 4-bromo-1-butene
followed by hydrolysis of the resulting Schiff base deriva-
tive10 2 yielded racemic 2-amino-5-hexenoate 3 (Scheme 1).
Attempted cross-metathesis11 between 5′-deoxy-2′,3′-O-iso-
propylidene-5′-methyleneadenosine 9a5a,7 and N-benzoyl 4
or N-Boc 5 protected amino acids bearing a terminal double
bond in the presence of first and second (2-imidazolidin-
ylidene-Ru) generation Grubbs catalysts11c,e failed to give
the desired products 10a or 11a (Scheme 2). Also, metathesis
of the 6-N-benzoyl adenosine substrate 9b with 4 or 5 was
unsuccessful. It is noteworthy that metathesis between 5′-
deoxy-2′,3′-O-isopropylidene-5′-methyleneuridine12 and 4
(CH2Cl2/second generation Grubbs catalyst) afforded the
desired product of type 10 (i.e., B ) U; 62%)13 in addition
to two dimers resulting from the self-metathesis of nucleo-
side14 and amino acid15 (e.g., 17) substrates.
(6) For other examples on the hydrolytic activity of AdoHcy hydrolase,
see: (a) Ref 1c. (b) Yang, X.; Yin, D.; Wnuk, S. F.; Robins, M. J.;
Borchardt, R.T. Biochemistry 2000, 39, 15234-15241. (4′-Haloacetylene
adenosine analogues). (c) Guillerm, G.; Guillerm, D.; Vandenplas-
Witkowski, C.; Roginaux, H.; Carte, N.; Leize, E.; Van Dorsselaer, A.; De
Clercq, E.; Lambert, C. J. Med. Chem. 2001, 44, 2743-2752. (5′-S-Allenyl-
5′-thioadenosine). (d) Jeong, L. S.; Yoo, S. J.; Lee, K. M.; Koo, M. J.;
Choi, W. J.; Kim, H. O.; Moon, H. R.; Lee, M. Y.; Park, J. G.; Lee, S. K.;
Chun, M. W. J. Med. Chem. 2003, 46, 201-203. (Fluoro-neplanocin A).
(e) Wnuk, S. F.; Lewandowska, E.; Sacasa, P. R.; Crain, L. N.; Zhang, J.;
Borchardt, R. T.; De Clercq, E. J. Med. Chem. 2004, 47, 5251-5257. (4′-
Enyne adenosine analogues). (f) Guillerm, G.; Muzard, M.; Glapski, C.
Bioorg. Med. Chem. Lett. 2004, 14, 5799-5802. (Haloethyl esters of
homoadenosine-6′-carboxylic acid). (g) Guillerm, G.; Muzard, M.; Glapski,
C.; Pilard, S.; De Clercq, E. J. Med. Chem. 2006, 49, 1223-1226. [5′-
Deoxy-5′-(cyanomethylene)adenosine].
(7) Wnuk, S. F.; Robins, M. J. Can. J. Chem. 1991, 69, 334-338.
(8) For example couplings between 5′-deoxy-5′-(iodomethylene)-
adenosine5a and suitable alkylzinc bromide produced analogues of type A,
see: Wnuk, S. F.; Lalama, J.; Andrei, D.; Garmendia, C.; Robert, J.
S-Adenosylhomocysteine and S-ribosylhomocysteine analogues with sulfur
atom replaced by the vinyl unit. Abstracts of Papers, Carbohydrate DiVision,
229th National Meeting of the American Chemical Society, San Diego,
CA, March 13-17, 2005; American Chemical Society: Washington, DC,
2005; CARB-035.
We found however that treatment of 9b with 4 in the
presence of the Hoveyda-Grubbs catalyst16 (o-isopro-
poxyphenylmethylene-Ru) led to the formation of metathesis
product 10b (51%) in addition to dimer 17 (11%), and self-
metathesis of 9b was not observed. Metathesis of the 6-N,N-
(10) (a) O’Donnell, M. J.; Wojciechowski, K. Synthesis 1984, 313-315.
(b) O’Donnell, M. J.; Polt, R. L. J. Org. Chem. 1982, 47, 2663-2666.
(11) (a) Grubbs, R. H.; Chang, S.; Tetrahedron 1998, 54, 4413-4450.
(b) Fu¨rstner, A. Angew. Chem., Int. Ed. 2000, 39, 3012-3043. (c) Trnka,
T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18-29. (d) Chatterjee, A.
K.; Choi, T.-L.; Sanders, D. P.; Grubbs, R. H. J. Am. Chem. Soc. 2003,
125, 11360-11370. (e) Nicolaou, K. C.; Bulger, P. G.; Sarlah, D. Angew.
Chem., Int. Ed. 2005, 44, 4490-4527.
(9) (a) Pd-catalyzed alkylation of the 5′-deoxy-5′-(dihalomethylene)-
adenosine5b precursors employing recently reported selective monoalkylation
of the unactivated 1,1-dichloro-1-alkenes9b or 1-fluoro-1-halo-1-alkenes9c
might give direct access to analogues B. (b) Tan, Z.; Negishi, E.-I. Angew.
Chem., Int. Ed. 2006, 45, 762-765. (c) Andrei, D.; Wnuk, S. F. J. Org.
Chem. 2006, 71, 405-408.
(12) Wnuk, S. F.; Robins, M. J. Can. J. Chem. 1993, 71, 192-198.
(13) Pablo R. Sacasa, M. Sc. Thesis, Florida International University,
2003.
5094
Org. Lett., Vol. 8, No. 22, 2006

1:1 mixture of saturated (at ∼0 °C) methanolic ammonia
solution and methanol for 48 h at ∼5 °C removed the 6-N-
benzoyl group(s) and produced a partially separable mixture
of methyl 12 and ethyl 13 esters (∼3:2, ∼92% total yield).
Using diluted NH3/MeOH minimized the formation of the
amidation byproducts (∼5%). Acid-catalyzed deprotection
of 12 and 13 with an aqueous solution of trifluoroacetic acid
(TFA) effected the removal of both the Boc and the
isopropylidene protection groups to give 14 and 15 in high
yields. It is important to perform debenzoylation of 11c (or
11b) as the initial deprotection step because treatment of 11c
(or 11b) with TFA/H2O resulted in the substantial cleavage
of the glycosylic bond. Saponification of 14 and 15 with
NaOH in H2O/MeOH solution and purification on RP-HPLC
afforded the sodium salt of 16 [67%; E, 9′R/S (∼1:1)].
Scheme 2
Because the separation of 9′R/S diastereomers in products
10-16 was difficult, we attempted the synthesis of analogue
A with a 9′S configuration employing a chiral amino acid
precursor, e.g., (S)-homoallylglycine. Given that the methods
available for the preparation of enantiomerically pure un-
natural amino acids usually require multistep synthesis,18 we
chose the enantioselective hydrolysis of racemic 5 as a way
to provide chiral (S)-homoallylglycine. Thus, treatment of 5
with R-chymotrypsin in phosphate buffer (24 h, 37 °C)19 gave
the unreacted (R)-ester 5 (∼50%) and (S)-acid 6 (∼50%,
Scheme 1). Enantiomeric purity of 5-R was established using
the Mosher test.20a Thus, treatment of 5-R with TFA/H2O
followed by acylation with (R)-2-methoxy-2-trifluoromethyl-
2-phenylacetyl chloride (MTPA-Cl)20 gave 8-R/S. (Note that
the absolute configuration at the chiral carbon in the Mosher
reagent is the same, but the R/S descriptors change owing
to the change in Cahn-Ingold-Prelog priority.) Analysis
of the 19F NMR spectra [δ -69.16 (s, 0.02F) and -69.55
(s, 0.98F)] established the stereochemistry for 5 as R (ee
96%) in agreement with Mosher’s empirical formula.20a
Because metathesis of the “free” carboxylic acid precursor
6-S with 9b or 9c in the presence of the Hoveyda-Grubbs
catalyst failed, 6-S was converted into the methyl ester 7-S
with diazomethane.
dibenzoyl 9c with 4 gave 10c in 60% yield in addition to
dimer 17 (18%). The protection of the 6-amino group of the
adenine ring seems to be necessary because metathesis
between 9a and 4 or 5 in the presence of the Hoveyda-
Grubbs catalyst did not yield the corresponding product 10a
or 11a.
Metathesis of 9b and 9c with N-Boc-protected 5 gave 11b
(61%) and 11c (76%) in higher isolated yields. Moreover,
byproducts of the self-metathesis of amino acid or nucleoside
substrates were not isolated. The cross-metathesis products
10 and 11 were found to be predominantly trans isomers.17
Purification on a silica gel column afforded 10 and 11 as
5′E isomers of a ∼1:1 mixture of 9′R/S diastereomers. The
1
E stereochemistry for 10 and 11 was established from H
NMR spectra based on the magnitude of JH5′-H6′. For ex-
Cross-metathesis of 9c with 7-S afforded 18-S (77%;
Scheme 3). Sequential deprotections of 18-S with NH3/
MeOH (to give 12-S, 91%) and TFA/H2O gave the enan-
tiomerically pure 14-S (90%) as the single E isomer (Scheme
3). On the other hand, metathesis of 9c with 5-R gave ethyl
ester 11c-R. Contrary to products 10-16 obtained from
racemic homoallylglycine, the 13C NMR spectra for the
products obtained from (S)- and (R)-homoallylglycine sub-
strates showed a single set of peaks.
ample, the 5′ proton in 11c appears at δ 5.58 (dd, JH5′-H4′
7.3 Hz and JH5′-H6′ ) 15.2 Hz) and the 6′ proton resonates
at δ 5.73 (dt, JH6′-H7′/7′′ ) 6.5 Hz and JH5′-H6′ ) 15.2 Hz).
Deprotection of 10 or 11 turned out to be more challenging
than we expected. Thus, treatment of 11c (or 11b) with a
)
(14) For recent reviews on application of metathesis towards synthesis
of nucleoside analogues, see: (a) Agrofoglio, L. A.; Nolan, S. P. Curr.
Top. Med. Chem. 2005, 5, 1541-1558. Amblard, F.; Nolan, S. P.;
Agrofoglio, L. A. Tetrahedron 2005, 61, 7067-7080. (b) For an example
on the self-metathesis reaction of carbohydrate-derived terminal olefins (e.g.,
5,6-dideoxy-1,2-O-isopropylidene-R-D-ribo-hex-5-enofuranose), see: Had-
wiger, P.; Stu¨tz, A. E. Synlett 1999, 1787-1789.
(18) (a) Dunn, M. J.; Jackson, R. F. W.; Pietruszka, J.; Turner, D. J.
Org. Chem. 1995, 60, 2210-2215. (b) Waelchli, R.; Beerli, C.; Meigel,
H.; Revesz, L. Bioorg. Med. Chem. Lett. 1997, 7, 2831-2836. (c) Lo¨hr,
B.; Orlich, S.; Kunz, H. Synlett 1999, 1139-1141. (d) Bachmann, S.;
Knudsen, K. R.; Jorgensen, K. A. Org. Biomol. Chem. 2004, 2, 2044-
2049.
(19) Schricker, B.; Thirring, K.; Berner, H. Bioorg. Med. Chem. Lett.
1992, 2, 387-390.
(20) (a) Sullivan, G. R.; Dale, J. A.; Mosher, H. S. J. Org. Chem. 1973,
38, 2143-2147. (b) Oh, S. S.; Butler, W. M.; Koreeda, M. J. Org. Chem.
1989, 54, 4499-4503.
(15) (a) Gibson, S. E.; Gibson, V. C.; Keen, S. P. Chem. Commun. 1997,
1107-1108. (b) Biagini, S. C. G.; Gibson, S. E.; Keen, S. P. J. Chem.
Soc., Perkin Trans. 1 1998, 2485-2499. (c) Vasbinder, M. M.; Miller, S.
J. J. Org. Chem. 2002, 67, 6240-6242.
(16) (a) Garber, S. B.; Kingsbury, J. S.; Gray, B. L.; Hoveyda, A. H. J.
Am. Chem. Soc. 2000, 122, 8168-8179. (b) Gessler, S.; Randl, S.; Blechert,
S. Tetrahedron Lett. 2000, 41, 9973-9976.
(17) 1H NMR analysis of the crude reaction mixtures showed the presence
of other isomers in variable quantities of ∼2-8%.
Org. Lett., Vol. 8, No. 22, 2006
5095

Scheme 3
Scheme 4
In summary, we have developed a synthesis of AdoHcy
analogues in which the carbon-5′ and sulfur atoms are
replaced by a vinyl unit utilizing cross-metathesis reactions
between 5′-deoxy-5′-methyleneadenosine analogues and
homoallylglycine in the presence of the Hoveyda-Grubbs
catalyst. The 5′-(bromo)vinyl AdoHcy analogue has been
prepared via the bromination-dehydrobromination strategy.
Enzymatic studies with AdoHcy hydrolase and our attempts
to synthesize 6′-(halo)vinyl analogues B via cross-coupling
approaches will be published elsewhere.
Finally, we attempted the synthesis of bromovinyl ana-
logue B by the bromination-dehydrobromination strategy.
Treatment of 11c with pyridinium tribromide21 gave the 5′,6′-
dibromo diastereomers 19 which were dehydrobrominated
with 1,8-diazobicyclo[5.4.0]undec-7-ene (DBU) to yield 20
as a single isomer (one of the 6-N-benzoyl protective groups
was also partially cleaved) in 70% yield (Scheme 4).
Standard deprotections with NH3/MeOH and TFA/H2O
followed by saponification with NaOH and HPLC purifica-
tion gave 23 (E, 54% overall).
The regioselectivity of HBr elimination and the position
of bromine (at 5′) were assigned on the basis of the presence
of a triplet signal for the olefinic hydrogen (H6′) in 1H NMR
spectra [δ 6.40 (t, J6′-7′/7′′ ) 7.6 Hz) for 23]. This assignment
was also supported by a COSY experiment. The product E
configuration is expected from a specific antiaddition in the
pyridinium tribromide bromination of the E alkene 11c
followed by an E2 (antielimination) process. This was also
supported by NOESY analysis of 23 in which the cross-
peaks between H4′ and H7′/7′′ were observed.
Acknowledgment. We thank the NIH/NIGMS program
(S06 GM08205) for supporting this research, US ARO
(W911NF-04-1-0022) for financial help to purchase a 600
MHz NMR spectrometer, and Materia Company for a gift
of the Hoveyda-Grubbs catalyst.
Supporting Information Available: Experimental pro-
cedures and characterization data for all compounds (PDF).
This material is available free of charge via the Internet at
http://pubs.acs.org.
(21) Husstedt, U.; Scha¨fer, H. J. Tetrahedron Lett. 1981, 623-624.
OL062026M
5096
Org. Lett., Vol. 8, No. 22, 2006