Exchange radioiodination produces inversion at C-4 of 1-(4-deoxy-4-iodo-β-D-xylopyranosyl)-2-nitroimidazole

Article abstract of DOI:10.1002/jlcr.596

The title compound, 3a, when exchange labeled with ¹²⁵I results in three new labeled products. The major labeled product (84.1%) is 1-(4-deoxy-4-iodo-β-L-arabinopyranosyl)-2-nitroimidazole, 3b, that could result from inversion of configuration at C-4. Exchange labeling carried out under conditions of kinetic control yielded dramatically different product ratios than thermodynamic equilibrium reactions. Confirmation of these results was established by extensive ¹H NMR spectral analyses. A possible mechanism is presented. Copyright

Full text of DOI:10.1002/jlcr.596

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2002; 45: 665–673.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.596
Research Article
Exchange radioiodination produces inversion
at C-4 of 1-(4-deoxy-4-iodo-b-d-
xylopyranosyl)-2-nitroimidazole
1
R.F. Schneider , W.A. Price , and J.D. Chapman*
2
,1
1
Department of Radiation Oncology, Tumor Biology and Biophysics
Section, Fox Chase Cancer Center, 7701 Burholme Avenue, Philadelphia,
PA 19111, USA
2
Department of Chemistry and Biochemistry, La Salle University,
1
900 West Olney Avenue, Philadelphia, PA 19141,USA
Summary
1
25
The title compound, 3a, when exchange labeled with I results in three new
labeled products. The major labeled product (84.1%) is 1-(4-deoxy-4-iodo-b-l-
arabinopyranosyl)-2-nitroimidazole, 3b, that could result from inversion of
configuration at C-4. Exchange labeling carried out under conditions of kinetic
control yielded dramatically different product ratios than thermodynamic
equilibrium reactions. Confirmation of these results was established by
1
extensive H NMR spectral analyses. A possible mechanism is presented.
Copyright # 2002 John Wiley & Sons, Ltd.
Key Words: radioiodination; exchange radiolabeling; inversion; pyranosides;
1
H NMR spectroscopy; iodine-125
Introduction
1
In a previous study reported in this journal, we described the
preparation of various radioiodinated azomycin substituted pyranosides
*Correspondence to: J. D. Chapman, Department of Radiation Oncology, Fox Chase Cancer
Center, 7701 Burholme avenue, Philadelphia, PA 19111, USA. E-mail: jd chapman@fccc.edu
Contract/grant sponsor: NIH; contract/grant numbers: CAO6927, CA55893.
Copyright # 2002 John Wiley & Sons, Ltd.
Received 4 May 2001
Revised 17 December 2001
Accepted 12 March 2002

6
66
R. F. SCHNEIDER ET AL.
and furanosides and their evaluation as possible markers of tumor
hypoxia. Those data showed that when 4-iodinated azomycin pentose
1
25
pyranosides were exchange labeled with
I, more than one product
was obtained. We have since identified some of these products and have
examined the factors that determine their formation. This information
should be valuable to others working in the field.
Results and discussion
While the focus of this paper is centered on understanding the
mechanisms for the appearance of multiple products arising from the
exchange radioiodination of 3a, there were sufficient problems in
the synthesis of the iodinated sugars used in this study to warrant
special discussion. For example, we were unable to prepare 3b via the
triphenylphosphine route shown in Figure 1.
The synthesis of 3c was included in an effort to gain some information
about the effect of precursor geometry, 2c, on the course of the
subsequent iodination reaction. The syntheses of 3a and 3c and the
attempted synthesis of the 4-iodo epimer of 3a, compound 3b, are
shown in Figure 1. The preparation of 3a was carried out as previously
1
reported. We attempted to synthesize 3b via a route identical to that
used for the preparation of 3a. However, using the standard iodine–
triphenylphosphine–pyridine system, we could not form the epimeric
4
-iodoarabinoside, 3b. This compound was later isolated, purified and
characterized as a product of NaI exchange with 3a. The C-3 epimer of
b, 1-(4-deoxy-4-iodo-b-l-lyxopyranosyl)-2-nitroimidazole, 3c, can be
3
1
prepared by the triphenylphosphine method, although in low yield.
As shown in Figure 1, the only difference between the precursor leading
to 3b and that leading to 3a or 3c is in the geometry of the hydroxyl
groups on C-3 and C-4. In the former, both the C-3 and C-4 hydroxyl
groups are equatorial, while in the latter, one is axial and the other
equatorial. The initial step of the iodination of 2a involves the
formation of an oxytriphenylphosphonium intermediate with the
equatorial C-4 hydroxyl group, followed by its rate-determining
2
,3
backside displacement by attack of iodide at C-4 to yield 3a. It
appears that the differences in precursor geometry have a profound
effect upon the course of the iodination reaction, and further study of
this problem is warranted.
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 665–673

RADIOIODINATION PRODUCES CHEMICAL INVERSION
667
R2
1
a Arabanopyranoside R ,R =OH
O
N
2
3
Az =
N
R1
1
1
b Xylopyranoside R ,R =OH
1 3
R3
c Ribopyranoside R ,R =OH
1
4
NO2
OH
OH
R4
N
1
. AcOAc, NaOAc 3.
NO2 , Hg(CN)2
NH
2
.HAc, HBr
4
. NaOMe, MeOH
R2
O
Precursor
2
2
2
a R ,R =OH
2 3
R1
b R ,R =OH
Az
1
3
R3
c R ,R =OH
1
4
OH
R4
Ph P, I , Py
3
2
X
2a
2b
2c
H
I
I
O
O
O
I
H
HO
H
Az
Az
Az
HO
OH
OH
OH
OH
3c
3a
3b
Arabinopyranoside
Xylopyranoside
Lyxopyranoside
Figure 1. Synthetic routes to 3a, 3b, 3c
The data presented in Table 1 show the results of the HPLC analysis
125
of the products obtained when 3a was exchange radiolabeled with I.
The total amount of radiolabeled compounds represents only a small
ꢀ4
fraction (approximately 10 ) of the material. In an effort to obtain
amounts of material sufficient for structural analysis, a large-scale
exchange reaction was carried out with equimolar amounts of 3a and
sodium iodide. A complimentary reaction involving the exchange of 3b
with NaI was also studied, and the results of these experiments are also
shown in Table 1. After 23 h, the ratio of products, 3a versus 3b, is
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 665–673

6
68
R. F. SCHNEIDER ET AL.
Table 1. Reaction products from the exchange radioiodination of compound 3a,
and the non-radioactive exchange iodination of substrates 3a or 3b
a
b
c
Expt
Reaction
Time (h)
8
3a (%)
3b (%)
4 (%)
d,e
1.
Radioiodination
[
2.3
84.1
13.6
1
3a>>>> [ I ]
25 ꢀ
2.
Starting with 3a
[3a]=[I ]
Starting with 3a
2
4
91.2
87.7
82.7
82.2
8.8
12.3
17.3
17.8
}
ꢀ
}
ꢀ
[3a]=[I ]
Starting with 3a
7
}
ꢀ
[
3a]=[I ]
Starting with 3a
23
Trace
ꢀ
[
3a]=[I ]
e
3.
Starting with 3b
3a]=[I ]
23
81.9
17.1
Trace
ꢀ
[
a
Xylopyranoside (HPLC retention time=52.2 min).
Arabinopyranoside (RT=26.5).
Unknown (3-iodolyxopyranoside ?, RT=45.6).
b
c
d
Only about 0.32 nmol out of ꢁ2.8 mmol of 3a becomes radioiodinated. Approximately 70% of the
applied radioactivity was recovered in labeled products.
Average of two runs.
e
1
Table 2. 600 MHz, H NMR analysis of 3a and 3b
H4'
I
H5e'
O
H5e'
O
H5
H4
H5
H4
I
H2'
a'
H4'
H2'
5a'
H5
HO
N
N
H
N
N
HO
HO
HO
NO2
NO2
H3'
H1'
H3'
H1'
3
a
3b
δ 7.69(d, 1H, J=1.1)H5
δ 7.17(d, 1H, J=1.1)H4
δ 6.12(d, 1H, J=9.0)H1'
δ 7.66(d, 1H, J=1.1)H5
δ 7.21(d, 1H, J=1.1)H4
δ 6.12(d, 1H, J=8.8)H1'
2
δ 4.25(dd, 1H, J=11.6, J=4.8)H5e'
δ 4.56(m, 1H)H '
4
2
δ 4.03(ddd, 1H, J=11.7, J=10.2,J=4.8)H4'
δ 3.95(dd, 1H, J=11.7, 2J=11.6)H5a'
δ 3.68(dd, 1H, J=10.2, J=8.8)H3'
δ 3.55(dd, 1H, J=8.8, J=8.8)H2'
δ 4.24(dd, 1H, J=13.5, J=1.5)H5e'
2
δ 4.13(dd, 1H, J=13.5, J=1.5)H5a'
δ 3.80(dd, 1H, J=8.8, J=8.8)H2'
δ 2.90(dd, 1H, J=8.8, J=4.0)H3'
essentially the same in both reactions. The two major products arising
from these studies were separated by preparative HPLC.
1
Proton nuclear magnetic resonance spectroscopic ( H NMR) analysis
identified the component with a retention time (RT)=52.2 min
as the starting material, 3a, and that at 26.5 min as its C-4 iodo
epimer, 1-(4-deoxy-4-iodo-b-l-arabinopyranosyl)-2-nitroimidazole, 3b.
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 665–673

RADIOIODINATION PRODUCES CHEMICAL INVERSION
1
A summary of the H NMR data leading to these assignments is
669
1
presented in Table 2. A key feature in the H NMR analysis of 3b can be
found at d4.13 (H5e
0
) and d4.24 (H5a
due to the very similar dihedral angles.
The H NMR spectrum for 3a, however, clearly shows strong coupling
11.7 Hz) between H5a and H4 , whereas J =4.8 Hz. The strong
coupling between H1 and H2
establishes these protons in a diaxial orientation. Peak assignments were
0
). Both H5e
0
and H5a couple
0
identically to the equatorial H4
0
1
(
0
0
0
0
5e ꢀ4
0
0
in both 3a and 3b (9.0 and 8.8 Hz)
1
1
unequivocally verified using correlated spectroscopy ( H, H COSY,
double quantum filtered, 600 MHz).
The conversion of 3a to a mixture of 3a/3b in the presence of NaI was
also studied as a function of time. The data are also presented in Table
1
. The reaction appears to reach equilibrium in about 7 h. Also inferred
from the data in Table 1 is that 3a is more stable than its epimer, 3b,
since it is the major equilibration product. It had been reported that the
radioiodination of 3a gave the corresponding radiolabeled version of 3a,
4
with no inversion at C-4. Unfortunately, the conditions reported for
that ‘iodine exchange reaction’ were not detailed enough to allow a
direct comparison with our data. Table 1 indicates that the radio-
iodination of 3a by standard procedures produces only 2.3% of
radiolabeled 3a, with the majority (84.1%) of the radiolabeled material
being a new component, 3b. A third component, 4 (13.6%), was also
detected.
The production of 3b from 3a can be explained by an S 2-type
N
reaction involving the backside attack of iodide at C-4. This is
illustrated in Figure 2. In this case iodide would also be the leaving
H
H
OH
O
I-
I-
O
I
Anchiomeric Assistance
O
Az
Az
I-
I-
HO
HO
H
I
I
O
4
I
Az
HO
HO
I
I-
I
3
a
O
O
I-
I
I-
I-
SN2
Az
Az
HO
HO
HO
HO
3b
Figure 2. Proposed mechanism for iodine exchange reactions
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 665–673

6
70
R. F. SCHNEIDER ET AL.
group. The resulting Walden inversion generates the C-4 iodo epimer of
5
a, the arabinoside 3b. In the case where the leaving group is identical
3
to the attacking nucleophile, the ratio of the products formed is
determined by their relative stability in the ground state, which leads to
5
the conclusion that 3a is more stable than 3b. Comparison of the data
presented in Table 1 reveals that in the radiolabeling procedure, when
4
there is a 10 -fold excess of 3a over radioiodide, radioiodinated 3b is the
major product. In the non-radioactive exchange reaction where
equimolar amounts of 3a and iodide are reacted, 3a is the major
product.
In the radiolabeling procedure that was performed under non-
equilibrating conditions, the product of kinetic control, radioiodinated
3
b, predominates. The reaction involving equivalent amounts of 3a and
iodide results in an equilibrium being established and represents
thermodynamic control of the product. A significant amount of product
4
resulted only from the radiolabeling procedure carried out under
kinetic control. This is a classic example of product distribution being
6
determined by conditions of kinetic versus thermodynamic control. In
Figure 2, we present a possible structure for 4, as well as a mechanism
for its formation. The synthetic procedures described in this paper did
not yield amounts of 4 sufficient for its definitive characterization.
Experimental
Chemicals and reagents were purchased from Aldrich Chemical Co.,
Milwaukee, Wisconsin or Sigma Chemical Co., St Louis, Missouri and
used as received. Preparative flash chromatography was carried out
using E. Merck silica gel 60 (200–400 mesh). TLC was performed on E.
Merck silica gel 60 fluorescent plates using a CHCl /MeOH (77.5:22.5)
3
125
elution system. I (NaI, 1 mCi, carrier free, nominally 2180 Ci/mmol)
was obtained in a pH 8.0–9.0 solution from NEN, Boston, MA. Melting
points were determined with a Laboratory Devices Mel-Temp II and are
uncorrected. Infrared spectra were obtained using KBr pellets with a
Nicolet 205 FTIR spectrometer. HPLC analyses were carried out on a
Beckman System Gold equipped with a 3.9 ꢂ 300 mm Millipore Delta
Pak 15 mm C-18 column and a BioScan Flow-Count g detection system.
The solvent system consisted of methanol/water (25:75) at 0.7 ml/min.
1
1
The uv/vis detector was set at 310 nm. Proton NMR and H, H
correlated spectroscopic studies (COSY) were carried out on a Bruker
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 665–673

RADIOIODINATION PRODUCES CHEMICAL INVERSION
671
6
00 DMX spectrometer using CDCl (99.8% deuterated) as solvent and
3
internal standard (d7.24). Gamma counting was carried out with a
1
25
Packard Cobra II Auto Gamma spectrometer. For
I, a 15–75 KeV
window was used with an efficiency of 81.5% and 4p geometry.
1
-(4-Deoxy-4-iodo-b-D-xylopyranosyl)-2-nitroimidazole, 3a
This material was newly synthesized according to a published
1
.
procedure. Briefly, l-arabinose was converted into tetra-O-acetyl-
a-l-arabinopyranoside by treatment with anhydrous NaOAc and
acetic anhydride under reflux. Treatment of the tetra-acetate with
30–32% HBr in acetic acid produced 2,3,4-tri-O-acetyl-a-l-arabinopy-
ranosyl bromide. The bromide was treated with 2-nitroimidazole and
Hg(CN)2 in anhydrous CH CN at ambient temperature to yield
3
1–(2,3,4-tri-O-acetyl-b-l-arabinopyranosyl)-2-nitroimidazole. Deacyla-
tion in anhydrous methanol in the presence of a catalytic amount
of NaOCH3 gave 1-(b-l-arabinopyranosyl)-2-nitroimidazole (2a in
Figure 1). Treatment of 2a with triphenylphosphine and iodine in fresh
anhydrous pyridine at 608C for 22 h yielded the title compound, 3a,
0
.235 g.
1
-(4-Deoxy-4-iodo-b-L-arabinopyranosyl)-2-nitroimidazole, 3b
An attempt was made to synthesize this material, the C-4 epimer of 3a,
by the route used to prepare 3a, but with the substitution of d-xylose for
l-arabinose as the starting sugar.. The final iodination of 2b to 3b was
unsuccessful. This compound was later isolated, purified and character-
ized as a product of 3a exchange with NaI (see large-scale conversion
experiment, below).
1-(4-Deoxy-4-iodo-b-L-lyxopyranosyl)-2-nitroimidazole, 3c
This material was synthesized by the route used to prepare 3a, but with
the substitution of tetra-O-acetylribose for l-arabinose.
Exchange radioiodization of 3a
1
25
Na I (1 mCi, carrier free, nominally 2,180 Ci/mmol) was transferred to
a 0.1 ml. reactivial and evaporated to dryness under a stream of dry
nitrogen. Anhydrous ethanol (100 ml) was added and re-evaporated to
dryness. A solution consisting of 1.0 mg of 3a dissolved in 20 ml
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 665–673

6
72
R. F. SCHNEIDER ET AL.
anhydrous DMF was added to the reactivial. The vial was sealed and
heated at 1008C for 8.0 h, then cooled and the DMF solvent removed
under a stream of dry nitrogen. The labeled 3a was purified by passing
through a mini-column containing (from bottom to top) 40 mg BioRad
AG1 ꢂ 8 anion resin (200–400 mesh, chloride form) and 50 mg AgCl
impregnated Celite (56 mg AgCl/100 mg). The column was washed with
2
.5 ml water. The sample was analyzed via HPLC using a Beckman
System Gold equipped with a 3.9 ꢂ 300 mm Millipore Delta Pak 15 mm
C-18 column. The solvent system consisted of methanol/water (25:75) at
0
.7 ml/min. The uv–vis detector was set at 310 nm.Radioactivity was
detected using a BioScan Flow Count Radiochromatography System set
125
to detect I.
Large-scale conversion of 3a to 3a/3b. (Alternate synthesis of 3b)
A solution of 3a (99.2 mg, 0.279 mmol) and dry sodium iodide (42 mg,
0
1
.279 mmol) in 1.98 ml of anhydrous dimethylformamide was heated at
o
00 C. Samples were withdrawn at 2, 4, 7 and 23 h, and analyzed via
HPLC. The reaction mixture was cooled to room temperature and the
dimethylformamide was removed on a rotary evaporator. The sticky
residue, which was dissolved in methanol, was subjected to preparative
HPLC using a C-18 Magnum column eluted with methanol/water
(25:75) at a flow rate of 0.7 ml/min. The detector was set at 315 nm.
Two fractions were collected at RT=26.5 min (12 mg, 17.8%) and
RT=52.2 min (63 mg, 82.2%).
Conversion of 3b to 3a/3b
A solution of 3b (5.4 mg, 0.015 mmol), and dry sodium iodide
(
2.27 mg, 0.015 mmol), in 108 ml of anhydrous dimethylformamide was
o
heated at 100 C for 23 h. The cooled reaction mixture was diluted with
methanol and analyzed via HPLC, using conditions as previously
described. Two peaks were observed at RT=26.5 min (17.1%) and
RT=52.2 min (81.9%) (see Table 1).
Conclusion
We have shown that the exchange radioiodination of 3a,
when performed under conditions of kinetic control, gives rise to
its epimer, radioiodinated 3b, and not radioiodinated 3a, as was
4
previously reported. We have also shown that the product
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 665–673

RADIOIODINATION PRODUCES CHEMICAL INVERSION
673
mix resulting from exchange iodination is determined by whether the
reaction is carried out under kinetic or thermodynamic control. Finally,
the direct preparation of 3b starting with d–xylose failed due to the
geometrical arrangement of the C-3 and C-4 hydroxyl groups of its
precursor, 2a.
Acknowledgements
F. inancial support for this research was provided by NIH Grants
#
CA06927 and #CA55893 and an appropriation from The Common-
wealth of Pennsylvania. In addition, sabbatical support for Dr William
Price was provided by La Salle University, Philadelphia, Pennsylvania.
The assistance of Cindy Myers in running the preparative HPLC
separations and Pat Bateman in formatting the manuscript is greatly
appreciated.
References
1
. Schneider RF, Engelhardt EL, Stobbe CC, Fenning MC, Chapman JD.
J Label Compd Radiopharm 1997; 39: 541–557.
2. Schaefer JP, Weinberg DS. J Org Chem 1965; 30: 2635–2639.
3. Haynes RK, Holden M. Aust J Chem 1982; 35: 517–524.
4. Mannan RH, Mercer JR, Wiebe LI, Kumar P, Somayaji VV, Chapman JD.
J Nucl Biol Med 1992; 36: 60–67.
5
. Eliel EL. Stereochemistry of Carbon Compounds (1st edn), McGraw-Hill:
New York, 1962; 219–229.
6
. March J. Advanced Organic Chemistry (3rd Edn), Wiley: New York, 1985;
1
87–189.
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 665–673