Synthesis and antitumour activity of new tiazofurin analogues bearing a 2,3-anhydro functionality in the furanose ring

Article abstract of DOI:10.1016/j.bmcl.2007.05.050

This paper describes a divergent de novo synthesis of 2-(2,3-anhydro-β-dribofuranosyl)thiazole-4-carboxamide (2′,3′-anhydro-tiazofurin) and the corresponding α- and β-homo-C-nucleosides, as well as evaluation of their antitumour activities in vitro.

Full text of DOI:10.1016/j.bmcl.2007.05.050

Bioorganic & Medicinal Chemistry Letters 17 (2007) 4123–4127
Synthesis and antitumour activity of new tiazofurin
analogues bearing a 2,3-anhydro functionality in the furanose ring
a
a
b
Mirjana Popsavin,a, Sasˇa Spaic, Milosˇ Svircˇev, Vesna Kojic,
*
´
´
b
a
´
Gordana Bogdanovic and Velimir Popsavin
aDepartment of Chemistry, Faculty of Sciences, University of Novi Sad, Trg D. Obradovic´a 3, 21000 Novi Sad, Serbia
bInstitute of Oncology Sremska Kamenica, Institutski put 4, 21204 Sremska Kamenica, Serbia
Received 17 April 2007; revised 15 May 2007; accepted 17 May 2007
Available online 23 May 2007
Abstract—This paper describes a divergent de novo synthesis of 2-(2,3-anhydro-b-Dribofuranosyl)thiazole-4-carboxamide (20,30-
anhydro-tiazofurin) and the corresponding a- and b-homo-C-nucleosides, as well as evaluation of their antitumour activities
in vitro.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Of the many classes of nucleoside analogues that have
been synthesized for potential clinical applications, one
of the most interesting is the class of C-nucleosides.1
Remarkable among them is tiazofurin (1, Fig. 1), a syn-
thetic C-nucleoside that shows antitumour activity in a
variety of tumour systems.2 In the phase II clinical trials,
it induced haematological responses in patients with
acute myelogenous leukaemia or chronic myeloid leu-
kaemia in blast crisis.3 The biological activity of tiazofu-
rin derives from a combination of cytotoxicity and
maturation-inducing activities.4 Both effects are attrib-
uted to inhibition of inosine 50-monophosphate dehy-
drogenase (IMPDH) by the tiazofurin adenine
dinucleotide, which induces the shutdown of guanylate
synthesis.5 Despite the remarkable efficacy of 1, lack of
Figure 1. Tiazofurin (1) and the targeted analogues 2 and 3.
specificity and occasional toxicity remains a problem
in its clinical use.2,6 In order to provide an access to
configuration at the C-10 position but also the functional
derivatives of reduced toxicity, a number of tiazofurin
analogues have been synthesized and evaluated for their
antitumour activities.7 In the course of our recent pro-
gramme directed towards total syntheses of tiazofurin
analogues with modified sugar moieties, we have re-
cently completed the synthesis of a cytotoxic b-D-ribo-
furanosyl-thiazole via the 2,5-anhydro derivative 4
(Scheme 1).8 Compound 4 has not only the required b-
groups suitable for introduction of diversity into the
nucleoside carbohydrate segment. Therefore, our next
endeavour was focused on further modifications of the
intermediate 4 for synthesis of the hitherto unknown tia-
zofurin derivative 2 (20,30-anhydro-tiazofurin), as well as
its homologue 3 having the 20,30-anhydro function in the
furanose ring. Synthesis and biological activity of nucle-
oside analogues with 20,30-anhydrofuranosyl sugar moi-
eties has been reported.9 Some of the results indicate
that 20,30-anhydro-nucleosides serve as DNA (or
RNA) polymerase termination substrates, that might
be of use for development of new antitumour agents.
Herein, we report on the synthesis of tiazofurin
Keywords: 2,5-Anhydro sugars; C-nucleosides; Antitumour agents;
Homo-C-nucleosides; Tiazofurin analogues; Thiazoles.
*
Corresponding author. Tel.: +381 21 485 27 68; fax: +381 21 454
065; e-mail: mpopsavin@ih.ns.ac.yu
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.05.050

4124
M. Popsavin et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4123–4127
Scheme 1. Reagents and conditions: (a) 4:1, TFA, 6 M HCl, 4 ꢁC, 140 h; (b) NH2OHÆHCl, NaOAc, EtOH, CH2Cl2, rt, 24 h; (c) MsCl, Py, ꢀ15 ꢁC,
0.5 h, then rt, 2 h, 60% from 4; (d) H2S, Py, rt, 4 h, 90%; (e) BrCH2COCO2Et, EtOH, 80 ꢁC, 50 min, 43% of 10, 10% of 9; (f) NH3, MeOH, rt, 8 days,
60%.
analogues 2 and 3 along with their effects on the prolif-
eration of selected human tumour cell lines.
or with KCN in the presence of benzo-15-crown-5 ether
(MeCN, 0 ꢁC), gave the heptononitrile 13 as the major
reaction product (72–73% from 11). The nitrile 13 was
treated with hydrogen sulfide gas under the conditions
similar to those already used for the preparation of 8.
However, the conversion of 13 to 14 required 14 days
to be completed, whereby the desired thioamide 14
was obtained in 78% yield. The Hantzsch reaction of
14 with ethyl bromopyruvate in ethanol gave the thia-
zole 15 (32%), accompanied with a minor amount of
the C-10 epimer 16 (21%). The a-anomer 16 was presum-
ably formed from 15 via a ring opening/ring closure
process promoted by HBr, which was formed as a
by-product in the Hantzsch reaction. Although the
acid-catalysed anomerisation of some a-D-ribofurano-
syl-C-nucleosides has been reported,11 this is the first
example of such a conversion involving a b-D-arabino-
furanosyl-C-nucleoside. Stereochemistry of 15 and 16
was unambiguously resolved by NOE differential 1H
NMR spectroscopy. Designation of the b-anomer 15
was based upon observation of a NOE at H-100 when
H-50 was irradiated. This effect was not observed in
16, presumably the a-anomer. However, this stereoiso-
mer exhibited a strong NOE between H-100 and H-20,
thus implying a spatial vicinity of these protons. Such
an arrangement is only possible if the isomer 16 repre-
sents the a-anomer. Both isomers 15 and 16 upon treat-
ment with saturated ammonia in methanol gave the
expected homo-C-nucleosides 312 and 1713 in 60% and
65% yields, respectively.
The synthesis of 2 started with hydrolytic removal of the
dioxolane protection in 4 that was achieved with 4:1
mixture of trifluoroacetic acid and 6 M hydrochloric
acid at +4 ꢁC. The resulting unstable aldehyde 5 was
not purified, but was rather immediately treated with
hydroxylamine hydrochloride to yield the correspond-
ing oxime 6 as a mixture of the corresponding E- and
Z-isomers. The mixture was not separated but was
rather further treated with mesyl chloride in pyridine
to give the corresponding nitrile 7 in 60% yield (with
respect to starting compound 4). Exposure of 7 to
hydrogen sulfide gas gave the thioamide 8 (90%). The
Hantzsch reaction of 8 with ethyl bromopyruvate affor-
ded the corresponding thiazole 10 (43%), along with a
minor amount of the aromatised product 9 (11%).
Treatment of 10 with methanolic ammonia provided
the target 210 (60%) as a result of successive aminolysis
of ester functions, followed by subsequent epoxide ring
closure process. An efficient two-step hemisynthesis of
20,30-anhydro-ribavirin (a closely related analogue of
2) has recently been described starting from ribavirin.9b
We are looking towards application of this methodol-
ogy for the preparation of 2 starting from tiazofurin.
Regardless of biological activities of both analogues,
compound 2 should have greater pharmacological
potential due to a profound stability of the C-gycosidic
bond in biological environments.
Synthesis of analogue 3 that represents a one-carbon
homologue of 2 is outlined in Scheme 2. Compound 4
was hydrolysed under the same reaction conditions as
described above, and the resulting crude aldehyde 5
was immediately treated with sodium borohydride in
methanol. The corresponding primary alcohol 11 was
thus obtained in 64% overall yield. Reaction of 11 with
triflic anhydride in pyridine and dichloromethane gave
the unstable triflic ester 12, which was used in the next
step immediately after its isolation from the reaction
mixture. Treatment of crude 12 with NaCN (DMF, rt),
Compounds 2, 3 and 17 were evaluated for their in vitro
cytotoxicity towards human myelogenous leukaemia
K562, promyelocytic leukaemia HL-60, human T-cell
leukaemia (Jurkat), human Burkitt’s lymphoma (Raji),
human colon adenocarcinoma HT-29, estrogen receptor
positive breast adenocarcinoma MCF-7 cell line, as well
as normal foetal lung fibroblasts (MRC-5). Cytotoxic
activity was determined by using the standard MTT as-
say, after exposure of cells to the tested compounds for
48 h. Tiazofurin (1) was used as a reference compound.
The results are presented in Table 1.

M. Popsavin et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4123–4127
4125
Scheme 2. Reagents and conditions: (a) NaBH4, MeOH, 0 ꢁC, 40 min, then rt, 40 min, 64% from 4; (b) Tf2O, Py, CH2Cl2, ꢀ10 ꢁC, 0.5 h, then rt,
0.5 h; (c) NaCN, DMF, rt, 1.5 h, 73%; (d) H2S, Py, rt, 14 days, 78%; (e) BrCH2COCO2Et, EtOH, 80 ꢁC, 50 min, 32% of 15, 21% of 16; (f) NH3,
MeOH, rt, 8 days, 60% of 3, 65% of 17.
Table 1. In vitro cytotoxicity of 1, 2, 3 and 17
a
Compound
IC50 (lM)
K562
HL-60
Jurkat
Raji
HT-29
MCF-7
MRC-5
Tiazofurin (1)
2
2.98
0.19
0.49
0.09
1.84
2.08
3.50
>100
0.51
4.84
16.06
0.54
0.07
0.89
>100
0.09
0.53
6.39
0.96
0.49
>100
>100
>100
3
17
2.18
11.01
9.09
10.01
>100
a IC50 is the concentration of compound required to inhibit the cell growth by 50% compared to an untreated control.
Remarkably, all three analogues 2, 3 and 17 exhibit
sub-micromolar cytotoxicity against K562 malignant
cells, with IC50 values ranging from 0.09 to 0.49 lM.
The most active compound against these cells is the
a-homo-C-nucleoside 17, being 33-fold more cytotoxic
than tiazofurin, which was recently approved as an or-
phan drug for treatment of the corresponding malig-
nant disease. Compounds 2 and 3 also efficiently
inhibited the growth of K562 cells, with respective
IC50 values being 16- and 6-fold lower than those ob-
served for the reference compound 1. According to
the data shown in Table 1, analogues 2 and 3, as well
as tiazofurin itself, showed similar and potent antitu-
mour activities towards the HL-60 cells. However, the
analogue 17 was found to be completely inactive
against this cell line. Tiazofurin remains the most po-
tent compound towards the Jurkat T cells and was over
9-fold more potent than the 20,30-anhydro derivative 2.
The homo-C-nucleosides 3 and 17showed a 4- and 22-
fold lower cytotoxicity in the same cell line when com-
pared to tiazofurin, respectively. Compound 2 showed
a potent antiproliferative activity towards the Raji cells
being almost 30-fold more active than tiazofurin. The
b-homo-C-nucleoside 3 exhibited much more pro-
nounced cytotoxicity against this cell line, being
approximately 230-fold more active with respect to
the reference compound 1. In contrast, the a-homo-
C-nucleoside 17 was completely inactive against this
cell line. Compound 2 is devoid of any cytotoxicity
against HT-29 cells, while both homo-C-nucleosides 3
and 17 exhibited almost 10- and 2-fold stronger cyto-
toxicity in the same cell line when compared to 1,
respectively. The analogues 3 and 17 were found to
be somewhat less active than the parent compound 1
against the breast adenocarcinoma MCF-7. However,
the 20,30-anhydro derivative 2 exhibited a stronger cyto-
toxicity towards this cell line, being approximately 7-
fold more active with respect to tiazofurin itself.
Remarkably, all newly synthesized tiazofurin analogues
2, 3 and 17 were found to be completely inactive
against the normal MRC-5 cells. These results do sug-
gest that compounds 2, 3 and 17 are more selective
than tiazofurin, but this should be verified by addi-
tional in vitro experiments with different normal cell
lines. The difference in cytotoxic activity indicates that
analogues 2, 3 and 17 are not acting at the same bio-
logical target (IMPDH) as tiazofurin itself. It is possi-
ble that these analogues act as specific alkylating
agents, and that the observed cytotoxicities originated
from an irreversible covalent binding (‘suicide inhibi-
tion’). Furthermore, the difference in cytotoxicity of
compounds 2, 3 and 17 (including their inactivity
against the normal MRC-5 cell line) may well be ex-
plained by a specificity difference in the hosts’ DNA
polymerases or nucleoside kinases, but further studies
will be needed to address this hypothesis.

4126
M. Popsavin et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4123–4127
In summary, three novel tiazofurin derivatives, 20,30-
anhydro-tiazofurin (2) and the corresponding b-(3)
and a-(17) homo-C-nucleosides, have been synthesized
and evaluated for their in vitro antitumour activity
against a number of human neoplastic cell lines. Com-
pound 2 exhibited the most pronounced cytotoxic activ-
ity against Raji cells, being almost 30-fold more potent
than the reference compound, tiazofurin (1). Compound
3 showed even more potent cytotoxicity towards these
cells, being approximately 230-fold more active with re-
spect to the reference compound 1. The most powerful
antitumour activity of compound 17 was recorded in
the K562 cell line, being 33-fold more active than tia-
zofurin. Moreover, none of the synthesized analogues
showed any significant cytotoxicity towards the normal
foetal lung fibroblasts. Finally, to the best of our knowl-
edge compounds 2, 3 and 17 represent the first biologi-
cally active tiazofurin analogues bearing a 2,3-anhydro
ribofuranosyl moiety, while the analogues 3 and 17 are
the first homo-C-tiazofurin derivatives that demonstrate
antiproliferative activity.
A.; Saita, M. G.; Iannazzo, D.; Romeo, G.; Mates, J. A.;
Tejero, T.; Merino, P. J. Org. Chem. 2005, 70, 8991; (e)
Chun, M. W.; Kim, M. J.; Shin, J. H.; Jeong, L. S.
Nucleosides Nucleotides Nucleic Acids 2005, 24, 975; (f)
Cai, D.-M.; Lin, K. H.; Li, M.-Z.; Wen, J. W.; Li, H.-Y.;
Jou, T.-P. Chin. J. Chem. 2005, 22, 1425; (g) Nair, V.;
Wenzel, T. ARKIVOC 2004, 14, 128; (h) Navarre, J.-M.;
Guianvarc’h, D.; Farese-Di Giorgio, A.; Condom, R.;
Benhida, R. Tetrahedron Lett. 2003, 44, 2199; (i) Liang, C.
W.; Kim, M. J.; Jeong, L. S.; Chun, M. W. Nucleosides
Nucleotides Nucleic Acids 2003, 22, 2039; (j) Popsavin, M.;
´
´
´
´
Torovic, Lj.; Kojic, V.; Bogdanovic, G.; Spaic, S.;
Popsavin, V. Bioorg. Med. Chem. Lett. 2003, 13, 3167;
(k) Cappellacci, L.; Barboni, G.; Franchetti, P.; Martini,
C.; Jayaram, H. N.; Grifantini, M. Nucleosides Nucleotides
Nucleic Acids 2003, 22, 869; (l) Franchetti, P.; Marchetti,
S.; Cappellacci, L.; Yalowitz, J. A.; Jayaram, H. N.;
Goldstein, B. M.; Grifantini, M. Bioorg. Med. Chem. Lett.
2001, 11, 67; (m) Franchetti, P.; Marchetti, S.; Cappell-
acci, L.; Jayaram, H. N.; Yalowitz, J. A.; Goldstein, B.
M.; Barascut, J.-L.; Dukhan, D.; Imbach, J.-L.; Grifan-
tini, M. J. Med. Chem. 2000, 43, 1264; (n) Franchetti, P.;
Marchetti, S.; Cappellacci, L.; Grifantini, M.; Goldstein,
B. M.; Dukhan, D.; Barascut, J.-L.; Imbach, J.-L.
Nucleosides Nucleotides 1999, 18, 679; (o) Zhang, H. Y.;
Yu, H. W.; Ma, L. T.; Min, J. M.; Zhang, L. H.
Tetrahedron: Asymmetry 1998, 9, 141; (p) Franchetti, P.;
Cappellacci, L.; Abu Seikha, G.; Jayaram, H. N.;
Gurudutt, V. V.; Sint, T.; Schneider, B. P.; Jones, W.
D.; Goldstein, B. M.; Perra, G.; De Montis, A.; Loi, A.;
La Colla, G. P.; Grifantini, M. J. Med. Chem. 1997, 40,
1731; (q) Franchetti, P.; Cappellacci, L.; Grifantini, M.;
Barzi, A.; Nocentini, G.; Yang, H.; O’Conor, A.; Jayaram,
H. N.; Corell, C.; Goldstein, B. M. J. Med. Chem. 1995,
38, 3829.
Acknowledgment
Financial support from the Ministry of Science and
Environment Protection of the Republic of Serbia (Pro-
ject No. 142005) is gratefully acknowledged.
References and notes
1. For recent reviews on C-nucleoside synthesis, see: (a)
Wellington, K. W.; Benner, S. A. Nucleosides Nucleotides
Nucleic Acids 2006, 25, 1309; (b) Pankiewicz, K. W.;
Watanabe, K. A.; Lesiak-Watanabe, K.; Goldstein, B. M.;
Jayaram, H. N. Curr. Med. Chem. 2002, 9, 733; (c)
Lamberth, C. Org. Prep. Proc. Int. 2002, 34, 149.
2. Grifantini, M. Curr. Opin. Invest. Drugs 2000, 1, 257.
3. (a) Malek, K.; Boosalis, M.; Waraska, K.; Mitchell, B. S.;
Wright, D. G. Leukemia Res. 2004, 28, 1125; (b) Wright,
D. G.; Boosalis, M.; Malek, K.; Waraska, K. Leukemia
Res. 2004, 28, 1137.
4. (a) Wright, D. G.; Boosalis, M. S.; Waraska, K.; Oshry, L.
J.; Weintraub, L. R.; Vosburgh, E. Anticancer Res. 1996,
16, 3349; (b) Tricot, G.; Weber, G. Anticancer Res. 1996,
16, 3341; (c) Weber, G.; Nagai, M.; Natsumeda, Y.; Eble,
J. N.; Jayaram, H. N.; Paulik, E.; Zhen, W. N.; Hoffman,
R.; Tricot, G. Cancer Commun. 1991, 3, 61.
´
´
´
8. Popsavin, M.; Torovic, Lj.; Kojic, V.; Bogdanovic, G.;
Popsavin, V. Tetrahedron Lett. 2004, 45, 7125.
9. (a) Takatsuki, K.; Ohgushi, S.; Kohmoto, S.; Kishik-
awa, K.; Yamamoto, M. Nucleosides Nucleotides
Nucleic Acids 2006, 25, 719; (b) Li, Z.; Chen, S.;
Jiang, N.; Cui, G. Nucleosides Nucleotides Nucleic Acids
2003, 22, 419; (c) Dyatkina, N. B.; Atrazheva, E. D.;
Alexandrova, L. A.; Krayevsky, A. A.; von Janta-
Lippinsky, M. Bioorg. Khim. 1988, 14, 815; (d) Webb,
T. R.; Mitsuya, H.; Broder, S. J. Med. Chem. 1988, 31,
1475.
25
10. Selected data for 2: mp 120–121 ꢁC (from MeOH), ½aꢁD
1
+32.8 (c 1.2, MeOH). H NMR (250 MHz, methanol-d4):
d 3.36 (dd, 1H, J5 a;5 b ¼ 11:4 Hz, J4 ;5 a ¼ 6:5 Hz, H-50a),
0
0
0
0
3.45 (dd, 1H, J4 ;5 b ¼ 5:5 Hz, H-50b), 3.75 (d, 1H,
0
0
J2 ;3 ¼ 2:8 Hz, H-30), 4.05 (dd, 1H, H-40), 4.11 (d, 1H,
H-20), 5.10 (s, 1H, H-10), 8.21 (s, 1H, H-5); NOE contact:
H-10 and H-40; 13C NMR (62.9 MHz, methanol-d4): d
59.89 (C-30), 61.02 (C-20), 63.30 (C-50), 78.95 (C-10), 82.12
(C-40), 126.06 (C-5), 151.19 (C-4), 165.50 (C-2), 172.08
(CONH2); CI MS: m/z 243 (MH+).
0
0
5. For a recent review, see: Pankiewicz, K. W.; Patterson, S.
E.; Black, P. L.; Jayaram, H. N.; Risal, D.; Goldstein, B.
M.; Stuyver, L. J.; Schinazi, R. F. Curr. Med. Chem. 2004,
11, 887.
´
6. (a) Vranic-Mandusic, V.; Subota, V.; Savovski, K.; Medic,
L.; Dramicanin, T.; Jozanov-Stankov, O.; Popov-Celek-
ˇ ´
´
11. Cupps, T. L.; Wise, D. S.; Townsend, L. B. J. Org. Chem.
1986, 51, 1058.
ˇ
´
25
´
´
´
etic, D.; Jokanovic, M.; Dimitrijevic, B. Toxicol. Lett.
2004, 146, 275; (b) Tricot, G.; Jayaram, H. N.; Weber, G.;
Hoffman, R. Int. J. Cell Cloning 1990, 8, 161.
12. Selected data for 3: mp 141–141.5 ꢁC (from MeOH), ½aꢁD
+53.3 (c 0.45, MeOH). 1H NMR (250 MHz, methanol-d4):
d 3.35 (d, 2H, J1 ;1 ¼ 7:0 Hz, 2· H-100), 3.60 (dd, 1H,
0
00
J4 ;5 a ¼ 5:2 Hz, J5 a;5 b ¼ 11:8 Hz, H-50a), 3.68 (dd, 1H,
´
7. (a) Popsavin, M.; Spaic, S.; Svircev, M.; Kojic, V.;
Bogdanovic, G.; Popsavin, V. Bioorg. Med. Chem. Lett.
ˇ
´
0
0
0
0
J4 ;5 b ¼ 4:1 Hz, H-50b), 3.90 (d, 1H, J2 ;3 ¼ 2:8 Hz, H-30),
3.96 (d, 1 H, H-20), 4.10 (t, 1 H, H-40), 4.44 (t, 1 H, H-10),
8.13 (s, 1 H, H-5); NOE contact: H-100 and H-50, H-100 and
´
0
0
0
0
´
ˇ
2006, 16, 5317; (b) Popsavin, M.; Torovic, Lj.; Svircev,
´
´
M.; Kojic, V.; Bogdanovic, G.; Popsavin, V. Bioorg. Med.
Chem. Lett. 2006, 16, 2773; (c) Merino, P.; Tejero, T.;
Unzurrunzaga, F. J.; Franco, S.; Chiacchio, U.; Saita, M.
G.; Iannazzo, D.; Pipernoc, A.; Romeo, G. Tetrahedron:
Asymmetry 2005, 16, 3865; (d) Chiacchio, U.; Rescifina,
13
H-20; C NMR (62.9 MHz, methanol-d4): d 37.33 (C-100),
59.83 (C-30), 60.88 (C-20), 63.28 (C-50), 78.93 (C-10), 81.32
(C-40), 125.98 (C-5), 150.21 (C-4), 159.22 (C-2), 168.77
(CONH2); CI MS: m/z 257 (MH+).

M. Popsavin et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4123–4127
4127
13. Selected spectral data for 17: mp 200 ꢁC (from MeOH–
with D2O, OH), 7.52 and 7.71 (2· br s, 2H,
exchangeable with D2O, NH2), 8.10 (s, 1H, H-5);
NOE contact: H-50 and H-30; 13C NMR (62.9 MHz,
DMSO-d6): d 34.59 (C-100), 57.67 (C-20), 58.04 (C-30),
61.51 (C-50), 76.32 (C-10), 79.08 (C-40), 124.73 (C-5),
149.72 (C-4), 162.77 (C-2), 167.25 (CONH2); CI MS:
m/z 257 (MH+).
25
1
CHCl3), ½aꢁD +4.0 (c 0.5, DMSO). H NMR (250 MHz,
00
00
DMSO-d6):
0
d
3.15 (dd, 1H, J1 a;1 b ¼ 14:9 Hz,
J1 ;1 a ¼ 7:3 Hz, H-100a), 3.28 (dd, 1H, J1 ;1 b ¼ 5:7 Hz,
00
0 00
H-100b), 3.45 (m, 2H, 2 · H-50), 3.79 (d, 1H,
J2 ;3 ¼ 3:0 Hz, H-30), 3.96 (d, 1H, H-20), 3.99 (t, 1H,
0
0
H-40), 4.37 (dd, 1H, H-10), 4.94 (br t, 1H, exchangeable