Conformational analysis of a secondary hydroxamic acid in aqueous solution by NOE spectroscopy

Article abstract of DOI:10.1002/mrc.3916

Hydroxamic acids are metal-binding compounds used by micro-organisms and possess applications in medicine and industry. Hydroxamic acids favor two conformations, E and Z; metal binding is limited to the Z conformation. The Z conformation may be identifiab

Full text of DOI:10.1002/mrc.3916

Rapid Communication
Received: 21 June 2012
Revised: 28 November 2012
Accepted: 30 November 2012
Published online in Wiley Online Library: 7 January 2013
(wileyonlinelibrary.com) DOI 10.1002/mrc.3916
Conformational analysis of a secondary
hydroxamic acid in aqueous solution by NOE
spectroscopy
Stefanie P. Sippl and Heather L. Schenck*
Hydroxamic acids are metal-binding compounds used by micro-organisms and possess applications in medicine and industry.
Hydroxamic acids favor two conformations, E and Z; metal binding is limited to the Z conformation. The Z conformation may
be identifiable by NOE spectroscopy, but analysis is complicated by the potential for long-range coupling as well as for relayed
NOEs due to conformational switching. In this report, we re-examine the reported conformational preference of N-methyl
acetohydroxamic acid (NMHA) in D O using NOE spectroscopy. We find that the favored conformation of NMHA in aqueous
2
solution is the E conformation, contrary to an earlier report. NOE build-up curves are proposed as a valuable tool to probe
conformational behavior in similar systems. Copyright © 2013 John Wiley & Sons, Ltd.
1
Keywords: NMR; H; conformations; NOE spectroscopy; hydroxamic acids; NOE build-up curve; long-range coupling; relayed NOE
[
16]
(
2
Scheme 1). Paraformaldehyde (0.616 g, 0.0205 mol) was stirred in
Introduction
0 ml Millipore water to dissolve. O-benzylhydroxylamine (3.683 g,
The hydroxamic acid is a naturally occurring metal ligand with high af-
finity for ferric iron. The relevance of hydroxamic acids to medicine,
microbial iron harvesting, and industrial metal-binding applications
0.0231 mol) was added, and the pH was adjusted to ~7 with 3 M
NaOH. The resulting two-phase mixture was stirred at room temper-
ature for 75 min. The mixture was extracted three times with CH Cl .
2
2
[1–13]
has prompted the development of various model systems.
Hydroxamic acids favor two conformations, the so-called E and Z (anti-
2 2
The CH Cl extracts were pooled and extracted twice with 0.5 M citric
acid, once with water, and once with brine. The organic layer was
dried with sodium sulfate and evaporated to yield 2.54 g (91.6%) of
[13]
periplanar and synperiplanar; Fig. 1). The hydroxamate unit must
exist in the Z conformation in order to function as a bidentate metal
ligand. The conformations differ in stability, with a rotational barrier
that is high enough to permit resolution of both species by NMR.
Assignment of E and Z forms has been made most frequently us-
1
oxime I as a clear oil. H NMR (CDCl ): 5.1 (s, 2H), 6.41 (d, 1H), 7.03
3
13
(d, 1H), 7.27–7.34 (m, 5H); C NMR: 76.1, 78.0, 128.0, 128.3, 128.4,
137.5; IR (neat): 3065, 3032, 2926, 1612, 1455, 1362, 1024 cm ; MS:
À1
m/z 136.0759 (M + H).
[4,13–15]
ing NMR chemical shift trends.
The propensity for hydroxamic
Oven dried glassware was used for the second reaction. Oxime
I (2.37 g, 0.0175 mol) was dissolved in glacial acetic acid (50 ml).
Acetic anhydride (2.0 ml, 0.021 mol) was added, followed by
sodium cyanoborohydride (1.367 g, 0.02175 mol). The reaction
was stirred at ambient temperature for 2 h, placed in a separatory
funnel with 50 ml methylene chloride, and then extracted twice
with 1 M potassium carbonate, twice with water, and once with
brine. The organic layer was dried with sodium sulfate and
evaporated. The product (II) was purified by chromatography
on silica using 9 : 1 hexanes/acetone to yield 0.699 g (19.0%) of
acids to aggregate in nonpolar media selectively stabilizes the Z form
at higher concentrations, which has been used as a basis for assign-
[
4b,7,9]
ment.
have also been used to support assignment of major and minor con-
Computational studies of differential stabilities of E and Z
[5,15]
formations.
One report has used NOE studies to assign conforma-
[10]
tions. Some inconsistencies exist in E and Z assignments, however.
For example, E and Z have each been assigned by different workers
[5,14,15]
as predominant for N-alkyl hydroxamic acids in water.
We report here the conformational preferences of N-methyl
1
acetohydroxamic acid (NMHA; Fig. 1) in D O, based on NOE data at
O-benzyl-N-methyl acetohydroxamic acid as a clear oil. H NMR
2
13
2
81 K. NMHA exists in the expected two states, in a 76 : 24 ratio,
(CDCl ): 2.05 (s, 3H), 3.18 (s, 3H), 4.81 (s, 2H), 7.37 (s, 5H); C NMR:
3
[5]
consistent with the prior report. The earlier report assigned the
major conformation as Z based on computational analysis of primary
hydroxamic acids. NOE data indicate that the favored conformation
is E. This result has implications for conformational characterization
and metal-binding behavior of hydroxamic acids in aqueous solution.
20.2, 33.4, 76.2, 128.7, 129.0, 129.3, 134.5, 172.7; IR (neat): 3033,
À1
2935, 1667, 1456, 1417, 1384 cm ; MS: m/z 180.1027 (M + H).
Acid washed glassware was used for the third reaction and for
all handling of NMHA solutions. O-benzyl-N-methyl acetohy-
droxamic acid II (0.699 g, 0.00390 mol) was placed in a Parr
Experimental
*
Correspondence to: Heather L. Schenck, Department of Chemistry and
Biochemistry, University of Wisconsin–La Crosse, 1725 State St., La Crosse, WI
All reagents were purchased from Thermo Fisher Scientific, Waltham,
MA, USA or from Sigma-Aldrich Corporation, St. Louis, MO, USA. All
reagents were used without further purification. A synthesis of hydro-
xamic acids reported by Lee and Miller was modified to afford NMHA
54601, USA. E-mail: hschenck@uwlax.edu
Department of Chemistry and Biochemistry, University of Wisconsin–La Crosse,
La Crosse, WI, USA
Magn. Reson. Chem. 2013, 51, 72–75
Copyright © 2013 John Wiley & Sons, Ltd.

Conformational analysis of a secondary hydroxamic acid by NOE spectroscopy
0.0035 M by dilution of the first sample. The specimen used for
NOE studies (0.035 M) was septum sealed and degassed using three
freeze/pump/thaw cycles. D
to 25 mM sodium 3-(trimethylsilyl)-1-propanesulfonate in D
2
O spectra were externally referenced
O.
2
NMR temperature was controlled for all data collection. NMR tem-
perature was calibrated using a commercial methanol standard.
One-dimensional proton spectra were collected using 16 scans
and the Bruker ZG30 pulse program. NOE spectra were collected
using 1056 scans and the Bruker SELNOGP pulse program. A
sweep width of 20.0255 ppm and 65 536 data points were used
for both 1D and NOE data collections. Irradiation for NOE spectra
was centered on either the major or the minor N-methyl proton
frequency. Mixing times for NOE spectra ranged from 0.02 to
1.0 s. NOE spectra took approximately 2 h to acquire. Apodization
that provided 0.3 Hz line broadening was applied to NOE data.
NOE spectra were phased to render the irradiation signal a nega-
tive peak. NOE spectra were referenced by applying the spectrum
reference value from the 1D spectrum to each SELNOGP data set.
Integral intensities for the NOE build-up curve were determined
using Topspin 3.1 software. The acetyl region was expanded, and
an integral region was defined starting at 2.120 ppm, continuing
across the acetyl region, and ending at 2.065 ppm. The overall
acetyl integral was then cut at 2.092ppm, the chemical shift at
which baseline resolution between signals for the two conforma-
tions was apparent at the longest mixing times. Integral area at
higher chemical shift was assigned to the major conformation,
and integral area at lower chemical shift was assigned to the minor
conformation. Integral areas were located under the ‘Integrals’ tab
of the main spectral window; the ‘Integral[abs]’ value was recorded.
Every individual integral value was normalized to (divided by)
the integral area of the minor conformation at 0.6 ms mixing
time; the latter integral had the largest value of the entire
series (Fig. 4).
Figure 1. N-methyl acetohydroxamic acid in (a) Z conformation and (b) E
conformation.
O
H2O
pH 7
+
O
H
H
H3N
Cl
1
. acetic anhydride,
acetic acid
O
N
2. NaBH CN
3
I
O
O
H2
OH
O
Pd/C
N
N
II
III
Scheme 1. Synthesis of N-methyl acetohydroxamic acid.
hydrogenation flask with 5% Pd on C (0.099 g) and methanol
2
(10 ml). The mixture was shaken under 30 psi of H for 45 min.
The methanol mixture containing the catalyst and NMHA (III)
was centrifuged in acid washed capped centrifuge tubes to pellet
the catalyst. The supernatant was removed, and the Parr vessel
and catalyst were rinsed with methanol, after which the suspen- Results and Discussion
sion was centrifuged and separated as before. The rinse and
1
H NMR spectra of 0.035 and 0.0035 M NMHA showed no discern-
centrifugation were repeated four times, and the combined
methanol supernatant fractions were evaporated under dry
nitrogen to yield 0.230g (66.3%) of NMHA as a faintly yellow oil.
able difference in E : Z ratio. These data were interpreted to mean
that aggregation was not occurring at either concentration. The
higher concentration sample was therefore used for NOE studies.
All signals appeared to be singlets. H NMR peaks for hydroxamic
acids are often broad around 300 K because of conformational
1
13
H NMR (D O): see Table 1; C NMR: see Table 1; IR (neat):
2
1
À1
3
1
400 (br), 3181 (br), 2918, 1621, 1427, 1392, 1202 cm ; MS: m/z
79.1035 (2M + H).
[
5,7,10]
exchange.
Data were therefore collected at 281 K to obtain
NMR studies used a Bruker Avance III 400 MHz spectrometer run-
maximum signal resolution and minimize exchange.
ning Topspin 3.1 software (Bruker BioSpin Corporation, Billerica,
MA, USA). All liquid handling was carried out with auto-pipettes.
NMR tubes were acid washed and oven dried. NMR samples were
Signals from the N-methyl group were better resolved than
the acetyl methyl signals. Accordingly, for NOE experiments,
irradiation was centered on the major or minor N-methyl
frequency. The proximity of acetyl and N-methyl protons in the
Z conformation suggested that an observable NOE in the acetyl
signal would be unique to the Z form.
prepared under N
solved in low-paramagnetic D
2
and were then parafilmed. NMHA was dis-
O (Cambridge Isotope Laboratories,
2
Cambridge, MA, USA; used without further purification) to a
concentration of 0.035 M, and a second sample was prepared at
Analysis of hydroxamic acids by NOE spectroscopy is compli-
cated by conformational interconversion. Polarization created in
one form can be carried into the other form, with the result that
[
10]
chemical exchange NOEs may be observed. Because chemical
exchange NOEs would grow in at later times than direct NOEs,
build-up curves were prepared to differentiate direct from
chemical exchange NOEs. Each N-methyl resonance was irradi-
ated across a range of mixing times, and the response in the
acetyl region was recorded.
Table 1. Chemical shifts of protons and carbons of NMHA in E and Z
conformations
N–CH
ppm)
3
N–CH
(ppm)
3
CH
(ppm)
3
C(O)
CH
(ppm)
3
C(O)
C(O)
(ppm)
(
E
3.22
3.36
32.41
35.78
2.11
2.09
15.60
16.37
170.42
166.24
Z
Irradiation of the minor N-methyl signal (3.36 ppm) generated a
response in the minor acetyl signal (2.09 ppm) in a time-dependent
Magn. Reson. Chem. 2013, 51, 72–75
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/mrc

S. P. Sippl and H. L. Schenck
fashion (Fig. 2). At longer mixing times (≥0.2 s), a second response
was observed in the major acetyl signal (2.11 ppm). The later
appearance of the signal at 2.11 suggested that this response was
a relayed NOE, a result of conformational switching from Z to E.
The major N-methyl resonance at 3.22 ppm was also irradiated
in a series of SELNOGP experiments with incremented mixing
times (Fig. 3). Interpretation was complicated by the presence
of weak long-range coupling between acetyl and N-methyl
protons, because COSY artifacts can appear in NOESY spectra.
This phenomenon appears in Fig. 3 as an antiphase component
at 2.11 ppm at shorter mixing times (up to 0.4 s). Presence of
coupling was confirmed by 2D COSY spectroscopy (not shown).
An all-positive response in the minor acetyl frequency
(2.09 ppm) does appear around 0.2 s (Fig. 3), at the same time
when we believe a relayed NOE is seen at 2.11 ppm in irradiation
of the minor N-methyl resonance (Fig. 2). As stated previously, we
believe this result is consistent with relay of polarization via
conformational exchange. We believe the same relay phenome-
non is responsible for the growth of positive intensity in the
major acetyl frequency at 2.11 ppm in Fig. 3. This phenomenon
is not predominant until mixing times exceed 0.4 s, which we
infer is correlated with polarization relayed back from the minor
acetyl (the initial relay in this series of experiments).
Growth curves of the responses to irradiation of the minor
N-methyl resonance (3.36 ppm; data shown in Fig. 2) are shown
in Fig. 4. The intensity of the response at 2.09 ppm (the minor
Figure 3. Result of irradiation of major N-methyl resonance at 3.22 ppm
(
acetyl region shown); from bottom: 1D proton spectrum; SELNOGP
spectra, mixing times: 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, and 1.0 s.
1
0
0
0
0
.8
.6
.4
.2
0
minor
major
0
0.5
1
mixing time (s)
Figure 2. Result of irradiation of minor N-methyl resonance at 3.36 ppm
Figure 4. NOE growth curves for responses at 2.09 and 2.11 ppm
following irradiation at 3.36 ppm. Minor conformer, 2.09 ppm; major
conformer, 2.11 ppm.
(acetyl region shown); from bottom: 1D proton spectrum; SELNOGP
spectra, mixing times: 0.02, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, and 1.0 s.
wileyonlinelibrary.com/journal/mrc
Copyright © 2013 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2013, 51, 72–75

Conformational analysis of a secondary hydroxamic acid by NOE spectroscopy
conformation) grows rapidly in the shortest mixing times, indicat-
ing a direct NOE between the minor N-methyl and minor acetyl
signals. This result is expected to be unique to the Z conformation,
because the methyl groups are too far separated in the E isomer
to show Overhauser enhancement. In contrast, the signal at
as by a University of Wisconsin-La Crosse Undergraduate Grant
for Research and Creativity to S.S. Mass spectra were provided
by the University of Wisconsin-Madison’s Mass Spectrometry Fa-
cility, the contributions of which are gratefully acknowledged.
Helpful discussions with Eric Johnson of Bruker BioSpin Corpora-
tion are also gratefully acknowledged.
2.11 ppm (the major conformation) shows only slowly increasing
NOE enhancement at early timepoints, which we believe is due
to relayed polarization. The rate of increase of peak intensity at
2
.11 ppm increases with mixing time, consistent with increased References
likelihood of conformational switching over time. Conversely,
the growth rate of NOE intensity at 2.09 ppm decreases with
mixing time, which is also consistent with relayed polarization
due to conformational switching. At the longest mixing time
[
[
1] M. J. Miller, Chem. Rev. 1989, 89, 1563–1579.
2] D. S. Kalinowski, D. R. Richardson, Pharmacol. Rev. 2005, 57, 547–583.
[3] J. C. Renshaw, G. D. Robson, A. P. J. Trinci, M. G. Wiebe, F. R. Livens,
D. Collison, R. J. Taylor, Mycol. Res. 2002, 106, 1123–1142.
[
4] (a) D. A. Brown, W. K. Glass, R. Mageswaran, B. Girmay, Mag.
Reson. Chem. 1988, 26, 970–973; (b) D. A. Brown, W. K. Glass,
P. Mageswaran, S. Ali Mohammed, Magn. Reson. Chem. 1991, 29,
40–45; (c) D. A. Brown, R. A. Coogan, N. J. Fitzpatrick, W. K. Glass,
D. E. Abukshima, L. Shiels, M. Ahlgren, K. Smolander, T. T. Pakkanen,
T. A. Pakkanen, M. Perakyla, J. Chem. Soc. Perkin Trans. 2 1996,
(1s), relative intensities of peaks at 2.09 and 2.11 ppm have
reversed, with the major species at that timepoint being the
major conformation in solution.
Taken in total, these data are all supportive of a favored confor-
mation of E for NMHA in D
as the Z conformation.
2
O. The minor species is thus assigned
2673–2678; (d) D. A. Brown, K. M. Herlihy, S. K. O’Shea, Inorg. Chem.
1
999, 38, 5198–5202; (e) D. A. Brown, L. P. Cuffe, G. M. Fitzpatrick,
N. J. Fitzpatrick, W. K. Glass, K. M. Herlihy, Collect. Czech. Chem.
Commun. 2001, 66, 99–108.
[5] M. T. Caudle, A. L. Crumbliss, Inorg. Chem. 1994, 33, 4077–4085.
6] C. P. Brink, A. L. Crumbliss, Inorg. Chem. 1984, 23, 4708–4718.
Conclusions
[
Results of NOE studies indicate that the E conformation is favored
[7] B. Garcia, S. Ibeas, A. Munoz, J. M. Leal, C. Ghinami, F. Secco,
M. Venturini, Inorg. Chem. 2003, 42, 5434–5441.
for NMHA in D O. The presence of long-range coupling in at least
2
[
[
8] G. A. Hope, R. Woods, A. N. Buckley, J. M. White, J. McLean. Inorg.
the Z conformation is observed. Relayed NOEs due to conforma-
tional switching are also evident at longer mixing times. Data
from a range of mixing times clarify the structural predilections
of NMHA, which has now been shown to favor the E conforma-
Chim. Acta 2010, 363, 935–943.
9] V. N. Kalinin, V. M. Yurchenko, J. Org. Chem. U.S.S.R. 1982, 1982,
1267–1271.
[10] W. Przychodzen, J. Chojnacki, Struct. Chem. 2008, 19, 637–644.
[11] R. Yamasaki, A. Tanatani, I. Azumaya, H. Masu, K. Yamaguchi,
H. Kagechika, Cryst Growth Des 2006, 6, 2007–2010.
[
4b]
tion in all solvents except dimethyl sulfoxide.
NOE build-up
curves have proven to be an effective way to differentiate the
effects of direct and relayed NOEs. Additional conformational
analysis of NMHA and homologs will aid in the design of effective
hydroxamic acids for medical and commercial applications.
[
12] E. Lipczynska-Kochany, H. J. Iwamura, J. Org. Chem. 1982, 47,
277–5282.
5
[13] J. Schraml, M. Kvicalova, V. Blechta, L. Soukupova, O. Exner,
H.-M. Boldhaus, F. Erdt, C. Bliefert, Magn. Reson. Chem. 2000,
3
8, 795–801.
14] M. Birus, M. Gabricevic, O. Kronja, B. Klaic, Inorg. Chem. 1995, 34,
110–3113.
[
[
[
Acknowledgements
3
15] M. Birus, M. Gabricevic, O. Kronja, B. Klaic, R. van Eldik, A. Zahl, Inorg.
Chem. 1999, 38, 4064–4069.
16] B. H. Lee, M. J. Miller, C. A. Prody, J. B. Neilands, J. Med. Chem. 1985,
28, 317–323.
This research was supported by an award from Research
Corporation for Science Advancement and by a Faculty Research
Grant to H.S. from the University of Wisconsin–La Crosse, as well
Magn. Reson. Chem. 2013, 51, 72–75
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/mrc