close

Вход

Забыли?

вход по аккаунту

?

On the Mechanism of Action of Urocanase Observation of the Enzyme-Bound NAD+ -Inhibitor Adduct by 13C NMR Spectroscopy.

код для вставкиСкачать
COMMUNICATIONS
The molecular packing characteristics in this class of bolaamphiphilic materials is currently under investigation for a wide
range of chain lengths.[25]
Received: Septemberl4,1995 [Z 7313 IE]
German version: Angew. Chem. 1995, 107, 707
Keywords: bolaamphiphiles . structure elucidation . thin films
[l] A bolaamphiphile contains a hydrophobic chain with hydrophilic groups at
both ends. See G. H. Escamilla, G. R. Newkome, Angew. Chem. 1995, 106,
2016; Angew. Chem. Int. Ed. Engl. 1995, 33, 1937.
[2] R. Popovitz-Biro, J. Majewski, L. Margulis, S. Cohen, L. Leiserowitz, M.
Lahav, J. Phys. Chem. 1995, 98,4970.
(31 R. Popovitz-Biro, J. Majewski, L. Margulis, S . Cohen, L. Leiserowitz, M.
Lahav, Adv. Muter. 1995, 6, 956.
[4] a) S. Weinbach, K. Kjaer, W. G. Bouwman, G . Griibel, G. Legrand, J. AlsNielsen, M. Lahav, L. Leiserowitz, Science 1995,264,1566; b) S. Weinbach, K.
Kjaer, J. Als-Nielsen, M. Lahav, L. Leiserowitz, J. Phys. Chem. 1993, 97, 5200.
[5] J. Als-Nielsen, K. Kjaer in Proceedings of the NATO Advanced Study lnsfitule
Phase Transition in Soft Condensed Malter (Eds.: T. Riste, D. Sherrington),
Plenum, New York, Geilo, Norway, 1989, p. 113.
161 R. Popovitz-Biro, unpublished results.
[7] The GID measurements were performed at the BW1 wiggler beamline at
HASYLAB, DESY, Hamburg, Germany, on a liquid surface diffractometer. A
beam of wavelength 1.35 A was incident at a grazing angle slightly below the
critical angle for total external reflection. This geometrical configuration enhances surface sensitivity. The footprint of the X-ray beam on the water surface
was 5 x 50 mm. The backgound level of scattering was reduced by a He atmosphere inside the trough. Detection of both the horizontal (qJ and vertical (yz)
components of the diffracted beam was done by a vertical oriented positionsensitive detector (4. resolution) mounted behind a Soller collimator giving q,,
resolution. The bolaamphiphiles were spread at room temperature from
5x
M chloroform solution over the Millipore-filtered water contained in
a Langmuir trough mounted on the diffractometer. GID measurements were
performed after cooling of the subphase to 5 "C. More experimental details are
given in Ref. [12].
[8] The GID pattern for 70, 100, and 200% surface coverage were essentially the
same, differing only in intensity and thus in the number of observable peaks.
[9] Two layers are found by GID in contrast to the two to five layers of the C , , diol
observed after deposition on mica, by atomic force microscopy (21. We may
reconcile this difference by bearing in mind that the multilayers are prepared
under dynamic conditions away from equilibrium, and so slight changes in
temperature, pressure, and support may lead to structural differences.
[lo] D. M. Small in Handbook of LipidResearch, Vol. 4, Plenum, New York, 1986.
[ll] The b glide is ruled out since it would lead to a crisscrossed arrangement of
chain axes and so to a poor packing, which has not yet been observed. Moreover, such a packing would not fit the Bragg rod profiles.
[12] F. Leveiller, D. Jacquemain, L. Leiserowitz, K. Kjaer, J. Als-Nielsen, J. Phys.
Chem. 1992. 96, 10380.
[I 31 The inversion centers and twofold axes are crystallographically correct for the
molecular chains, but only statistically correct for the hydroxyl hydrogen
atoms in the hydrogen-bonding chain (Scheme 1 b), which must be disordered
as in the 3D crystal structures of hexagonal ice [lsa], methanol [15b],
C,,H,,OH [15c]. and C,,H,,OH [lSd]. A completely ordered O - H . . . O
hydrogen-bonded system with identical 0 - H ' . 0 repeat along the hydrogenbond chain axis a may be achieved with a twofold screw axis parallel to a. Such
an arrangement would yield the commonly observed orthorhombic space
group Pcab, but was found not to fit the GID data, which requires all the chains
axes to be tilted in the same direction vis-a-vis the a axis.
[14] The best fit of the calculated Bragg rod intensity profiles to the observed data
was obtained for an interlayer offset along the a axis of 3.3 8, and an interlayer
separation of 30 A. Furthermore, modeling with one or three layers yielded
calculated Bragg rods with shapes distinctly different from those of the observed data.
[15] a) D. W. Peterson, H. A. Levy, Acta Cryslallogr. 1957, 10, 70; K. Shimaoko, J.
Phys. SOC.Jpn. 1960, 15,106; b) K. J. Tauer, W. N. Lipscomb, Acla Crystallogr. 1952,5,606; c) S. Abrahamsson, G. Larsson, E. von Sydow, ibid. 1960,13,
770; d) J.-L. Wang, F. Leveiller, D. Jacquemain, K. Kjaer, J. Als-Nielsen, M.
Lahav, L. Leiserowitz, J. Am. Chem. Soc. 1995, 116, 1192.
1161 D. Jacquemain, S. Grayer Wolf, F. Leveiller, K. Kjaer, M. Deutsh, J. AlsNielsen, M. Lahav, L. Leiserowitz, Angew. Chem. 1992, 104, 134; Angew.
Chem. Inl. Ed. Engl. 1992, 31, 130.
[17] Model structures made up of one or three layers did not tit the observed Bragg
rod data satisfactorily. We also note that for the given grazing incidence geometry the intensity of the X-ray beam falls off exponentially with depth of penetration of a value I,/e (where I, is the intensity of the incident beam) for the
depth A z 90 A. If the thickness of the multilayer crystallites is less than 2 A ,
the shape of the Bragg rod is barely affected by the limited penetration depth.
Furthermore the shape of the two-dimensional contour of the diffracted inten-
652
0 VCH
Verlagsgeselischafi mbH, 0-69451 Weinheim, 1995
sity f(qxy,qJ of the various reflections (not presented here) show that the ab
plane of the multilayer crystallites are primarily parallel to the water surface.
There is, however, some tendency for deviation therefrom. If such a spread of
crystalline misorientation from the plane of water surface is pronounced, it
would give rise to an apparent broadening of the Bragg rods along 4,. which
would have to be treated in a detailed analysis (K. Kjaer, W. G. Bouwman,
unpublished work),
[18] For a chainlike hydrocarbon molecule the limited number of observable peaks
in the GID data does not allow one to differentiate between two structures
differing in orientation by 180° about the long molecular axis unless other
criteria are introduced.
[19] This reduction in crystal symmetry was found to be necessary for a good fit to
the GID data of monolayer structures of several amphiphilic molecules on
water 112, 15dI. In the multilayer C,,-diol structure the lean angles for the first
and second layers were in opposite directions. In the refined structure the
molecular chains are tilted by 1.5" along the a axis and leaned by 1.5" along the
b axis.
[ZO] In this packing the center of inversion and twofold axes relating the two layers
in the two space group are lost. Moreover, the relaxation of the P 2 J a (or A 2 / a )
symmetry is also manifested by the presence of an observed but very weak
symmetry-forbidden { 1,O) reflection.
[21] The results for symmetry relaxed from A2/a are very similar to that for P 2 J a
and so are not shown here.
[22] In addition to ct and p phases a minor y phase exhibiting two very weak Bragg
rods was observed. Their intensities were approximately one hundredth those
of the major phases. The qxV.q, positions of these two Bragg rods indicate a
rectangular unit cell of dimensions a = 5.08 A, b = 8.73 8,. Modeling yielded
multilayer (trilayer) structure with the molecules tilted by about 30" from the
vertical in the b direction.
[23] The intensity profiles of the {0,2} and {2,0} (not shown in Fig. 4a) reflections
of monolayer phase a were used to derive a model of its structure. The calculated intensitycontribution of this model structure was subtracted from the intensity profile of the unresolved {1,1} reflection. The remaining intensity profile
phase was well fitted to the
together with the resolved {0,2} reflection of the /l
calculated structure of the bilayer previously determined (Fig. 3). The unit cell
of the monolayer was found to differ slightly from that of the double layer:
a = 5.0, b =7.43 8,.
[24] A detailed determination of the multilayer structure would also benefit from
atom-atom potential energy calculations as carried out on the amphiphilic
monolayer structures [12].
[25] The structures of the C,,, C,,, and C,, diols will be reported elsewhere.
On the Mechanism of Action of Urocanase:
Observation of the Enzyme-Bound
NAD' -Inhibitor Adduct by
13C NMR Spectroscopy**
Carsten Schubert, Herbert Rottele, Manfred Spraul,
and Janos RCtey*
Urocanase (E.C. 4.2.1.49) catalyzes the second step in the
degradation of histidine in most cells (Scheme 1) and contains a
tightly bound NAD' which is essential for catalytic activity.['*
NAD' is supposed to function as an electrophilic catalyst and
forms a transient covalent adduct with the imidazole ring of the
substrate, urocanic acid (1) .I3] According to the mechanism postulated by US,[^-^^ the unusual addition of water leading to
5'-hydroxyimidazolepropionic acid 2 occurs in a chemically
['I Prof. Dr. J. Retey, Dr. C. Schubert, Dr. H. Rottele
Institut fur Organische Chemie der Universitat
Richard-Willstitter-Allee, D-76128 Karlsruhe (Germany)
Telefax: Int. code + (721)608-4823
Dr. M. Spraul
Bruker Analytische MeBtechnik GmbH
D-76287 Rheinstetten (Germany)
[**I Generous financial support by the Deutsche Forscbungsgemeinschaft and the
Fonds der Chemischen Industrie is acknowledged. We thank Ms. J. Herman for
numerous enzyme isolations.
0570-0833/95/0606-06S2$ fO.OO+ .25/0
Angew. Chem. h i . Ed. Engl. 1995, 34, No. 6
COMMUNICATIONS
<’
“+
coon
5’ position, the observation of the enzymebound adduct was confirmed, and the
chemical shift of the C-5’ atom was deterHs, HRe
H,, HRe
Urocanase
mined. In the difference spectrum (Fig. 3 )
A
N v N H
N v N H
7
N V N H
two 13C NMR signals are observed in the
1
H20
2
3
region between 6 = 80 and d = 120, namely
Scheme 1. Stem course of the urocanase reaction. The protons incorporated from the solvent are added
at 6 = 1 1 7 and at 6 = 87.5 for the free and
to the Re face
the bound (5‘-13C) imidazolepropionic
acid, respectively. The large highfield shift
of the C-5’ signal indicates that the imidaplausible manner. Intermediate 2 is in a spontaneous equilibrizole ring in the enzyme-bound adduct is not aromatic; in other
um with imidazolonepropionic acid 3.r6’
words the adduct has structure 4a.
Recently we determined the structure of an NAD+ -inhibitor
adduct, which is formed on urocanase and is most likely
analogous to the transient NAD+-substrate adduct, by mea- a)
suring the one-bond 13C-13Ccoupling constant^.^^' Because of
its instability the double 13C-labeledadduct had to be stabilized
by oxidation while still bound to the enzyme and was subsequently separated from the enzyme.[4.
Here we report on our direct observation of the native adduct
in the enzyme-bound state for which we used difference I3C
NMR spectroscopy. To enhance the relevant I3C signals both
position 4 of the prosthetic NAD’ and position 5’ of the inCOOH
Hs,
pHRe
H,
COOH
o
pH~e
w
’
hibitor were enriched with I3C. In a first experiment we applied
urocanase containing (4-‘ 3C)NAD+ (99 % 13C). In the
150 MHz 13C NMR spectrum the enzyme-bound (4-13C)NAD’ gives rise to a broad signal at 6 = 146 (Fig. 1). After b)
1
55
I
I
50
45
I
40
I
I
I
I
35
30
25
20
I
I
15
-6
V
h
180
170
160
150
120
130
140
- 6
.
I
180
.
, .
170
I
. , .
160
150
I
,
140
I
130
,
I
.
Fig. 2. 13CNMR difference spectrum of (4-’3C)NADt-containing urocanase (after - before) addition of 9.9 pmol of imidazolepropionic acid. a) High-field portion, b) low-field portion. The broad signal at 6 = 37.7 arises from the enzvmebound(4-”C)NADt-imidazolepropionateadduct.
120
-6
Fig. 1. Low-field portionofthe 150MH~”CNMRspectrumofthe(4-’~C)NAD~containing urocanase. The broad signal at 6 = 146 stems from the C-4 atom of the
nicotinamide moiety (99% ”C). Approximately 120 mg (0.99 pmol) of urocanase
in 2 mL of SO mM K,HPO, buffer, pH 7.5, H,O/D,O 8: 1, 10 m m NMR tube, ‘H
broadband decoupled, line broadening = 10 Hz,T = 300 K, 30000 transients.
addition of unlabeled imidazolepropionic acid in tenfold
excess (fivefold excess with respect to the active sites) the
signal at 6 =I46 largely disappears, while at the same
time a new, likewise broad signal at 6 = 37.7 arises. This
indicates that practically all active sites are occupied by the
inhibitor and is most clearly seen in the difference spectrum
(Fig. 2). Besides the sharp I3CNMR signals of the excess unbound imidazolepropionic acid, one observes the signal of the
I3C-4 atom of the enzyme-bound (4-”C)NAD+ -imidazolepropionate adduct (4 or 4a; Scheme 2) at 6 = 37.7. For comparison: The C-4 atom of NADH 5 resonates at 6 = 21.9,
whereas that of C-4 phosphorylated dihydropyridine derivative 6 appears at 6 =78.7.[*’
In a further experiment, in which instead of NAD+ now the
inhibitor, irnidazolepropionic acid, was labeled with
Angen’ Chem Int. Ed Engl. 1995, 34, N o . 6
0 VCH
13C
in the
CONH,
I
4
4
a
R
CONH,
S
I
R
I
6
R
Scheme 2. Possible structures of the enzyme-bound NAD’ -imidazolepropionate
adduct 4 and 4 a and structures of NADH 5 and of the NAD-Cphosphate adduct 6. R = adenosine diphosphate ribosyl.
The preference of 4 a over 4 at the active site can be explained
by an ionic interaction of a negatively charged group of the
enzyme and the positively charged nitrogen atom of the imida-
Verlugsgesellschufi mbH, 0-69451 Wernherm, 1995
0570-0833/95/0606-0653
$10 00+ 2510
653
COMMUNICATIONS
of toluene molecule, despite their large separation (2.18 A), and
a significant fraction of the positive charge of Et,Si+ is transferred to the aromatic ring. Reed et al.[51recognize the existence
of a covalent interaction between the Et3Si group and the
toluene molecule and assign the adduct the structure of a
complex with an sp2 carbon (2), rather than of the cr complex 1
with an sp3 carbon.
130
110
120
100
90
80
70
BSiR,
-6
Fig. 3. 100 MHz 13C NMR difference spectrum of unlabeled urocanase (after - before) addition of 2 pmol of (5'-'3C)imidazolepropionic acid. The broad
signal at 6 = 87.5 arises from the NAD' -(5'-13C)imidazolepropionateadduct; the
sharp signal at 6 = 117 corresponds to free (5'-'3C)imidazolepropionic aa d . Approximately 150 mg (1.22 pmol) of urocanase in 2 mL of 50 mM K,HPO, buffer,
pH 7.5. H,O/D,O 8:1,10 mm NMR tube, 'H-broadband decoupled, NOE unsuppressed, line broadening = 10 Hz, T = 297 K, 52000 scans.
zole ring. Evidence for an equilibrium between 4 a and 4 is
provided by the previously observed urocanase-catalyzed exchange of the H-5 atom of imidazole propionate with solvent
protons.16, "1
'3
Received: November 4, 1995 [Z 7452 IE]
German version: Angew. Chem. 1995, 107, 719
Keywords: 13CNMR spectroscopy . enzyme mechanisms . urocanase
111 R. M. Egan, A. T. Phillips, J. Biul. Chem. 1977. 252, 5701 - 5707.
[2] V. Keul. E Kaeppeli, C. Ghosh, T Krebs, J. A. Robinson, J. Retey, J. Bid.
Chem. 1979, 254, 843-851.
[3] I. Ritey, Arch. Biochem. Biophys. 1995, 314, 1-16.
[41 J. Klepp, A. Fallert-Muller. K. Grimm, W E. Hull, J. Retey, Eur. J Bmchem.
1990, 192,669-676.
I51 C. Schubert, Y Zhao, J.-H. Shin, J. RCtey, Angew. Chem. 1995, i06,1331-1332;
Angew. Chem. In!. Ed. Engl. 1995.33, 1279-1280.
I61 E. Gerlinger, W. E. Hull, J. Retey, Eur. J. Biochem. 1981, 117, 629-634.
[71 L. H. Matherly. C. W De Brosse, A. T. Phillips, Biochemistry 1982,21. 27892794.
[XI J.-F. Biellmann, C. Lapinte, E. Haid, G. Weimann. Biochemistry 1979, 18,
1212-1217.
[91 R. M. Egan, L. H. Matherly, A. T. Phillips, Biochemistry 1981, 20, 132-137.
[lo] S. Sawada, K. Endo, M. Ushida, N. Esaki, K. Soda, Bull. Kyoto Unrv. Educulion Ser. B 1981, 51, 11-17.
[R,Si-arenej
+
CT
Complexes in the Gas Phase**
Fulvio Cacace,* Marina Attina, and
Simonetta Fornarini
The report by Lambert et al.['] that an R3Si' ion that is not
coordinated to the anion and only weakly to the solvent exists
in the solid [Et3SiB(C,F5), . toluene] has stirred a controversy,
mainly focused on the role of toluene.
Schleyer et al.,[*I in agreement with Olah et al.,[31argue that
the compound is actually a salt of the arenium ion
[Et,Si . toluene]', that is, a o complex of type 1. According to
Pa~ling,[~I
in the compound studied by Lambert et al.['] a covalent bond does exist between the Si atom and the para C atom
[*I Prof. F. Cacace, Prof. M. Attina, Prof. S. Fornarini
Universita di Roma "La Sapienza"
P. le Aldo Moro 5, 1-00185 Rome (Italy)
Telefax: Int. code + (6349913.881
[**I
This work was supported by Italian Minister0 dell'Universiti e della Ricerca
Scientifica e Tecnologica (MURST) and Consiglio Nazionale delle Ricerche
(CNR). The experimental help of F. Angelelli and A. Ricci is gratefully acknowledged.
654
0 VCH
Verlugsgesellschuft mbH. 0-69451 Weinheim, 1995
2
1
Discrimination between the above structures is difficult, since
it is necessarily based on certain critical parameters such as the
Si-C-C,,,, angle, whose values deduced from crystal structure
analysis happen to be intermediate between those expected for
unperturbed ions 1 and 2. Indeed, the intrinsic structural features of the ions in the crystal lattice are affected by interactions
with neighboring molecules, which detracts from the generality
of the conclusions and may account for their inconsistency with
the results of theoretical calculations on free, unperturbed
species found exclusively in the gas phase.L6I
We have addressed the problem by investigating in the gas
phase the same [R,Si-arene]+ adducts studied in the solid state
by the radiolytic techniqueL7]and by the FT-ICR (ICR = ion
cyclotron resonance) mass spectrometry. The reaction sequence
shown in Scheme 1 in which B denotes a gaseous base or nucleophile was used.
R,Si+
-
+ C6X5RL
[R,Si-C,X,R']+
R,SiB+
+B
[R,Si-C,X,R']+
(1)
+ C,X,R'
(11)
+ BX+
R,SiC,X,R'
(111)
(R = Et, iPr; R ' = X, CX,; X = H, D)
In the radiolytic experiments, the i-C3H: ions from the y
irradiation of gaseous mixtures containing C3H, as the bulk
component react with R,SiH to give R3Si+ ions, which react
with toluene according to (I) to yield [R,Si-C,H,]+ adducts.
The latter, efficiently deactivated by thermalization in the dense
gas, undergo reactions (11) or (111) or both with the bases and
nucleophiles present in the system, yielding neutral products
that are analyzed by GC/MS.[71The composition of typical
systems and the yields of the products, expressed by their G , ,
values (molecules formed per 100 eV), are reported in Table 1 .
In the absence of added bases, only cymenes are formed from
Table 1. Gas-phase trialkylsilylation of toluene [a]
Partial pressures/Torr [b]
C,H, Toluene Additives
R
Product yields (G+J [c]
(me1u:puru) MeC,H,i-Pr
MeC,H,SiR,
iPr
iPr
(20:80)
(21:79)
(20:80)
(24:76)
(34:66)
(31:63)
(25:75)
0.13
0.13
0.14
0.23
0.21
0.34
0.30
(24:76)
0.15
615
650
620
650
630
620
650
1.70
1.63
1.80
1.60
1.68
1.87
2.07
pyridine: 0.85
pyridine: 0.36
pyridine: 1.75
Et,N: 0.78
iPr,EtN: 0.80
r'i
iPr
iPr
iPr
(MeO),PO: 1.01 PI
0.16
0.12
0.16
0.076
0.038
0.012
0.007
630
1.10
pyridine: 1.30
Et
0.07
650
1.40
-
-
670
0.82
Et,N: 0.79
Et
Et
0.088
-
(24:76)
0.19
0.14
[a] All irradiations performed at 80°C. [b] All systems contained R,SiH (10 Torr)
and 0, (10Torr). [c] Standard deviation: + l o % .
0570-0833/95/0606-0654$10.00+ .25/0
Angew. Chem. Int. Ed. Engl. 1995, 34, Nu. 6
Документ
Категория
Без категории
Просмотров
1
Размер файла
365 Кб
Теги
adduct, nad, spectroscopy, nmr, bound, 13c, enzymes, inhibitors, observations, mechanism, action, urocanase
1/--страниц
Пожаловаться на содержимое документа