Transcript presentation

Microwave spectroscopy
of biomimetics molecules
Isabelle KLEINER
Laboratoire Interuniversitaire des Systèmes
Atmosphériques (LISA), Créteil, France
Nice, 15-16 Sept 2009
What do we call « Biomimetic
Molecules » ?
 Small molecules forming the elementary
blocks of biomolecules: amino acids, small
peptides, nucleic acids, sugars…
 Can serve as validation tools
 relatively small molecules are the favourite
candidates for most oral drugs (so-called
« Lipinsky rule »):
-molecular weight of 500 or less,
-not more than 5 hydrogen-bond donor sites,
-not more than 10 hydrogen-bond acceptor
sites
Jorgensen, “Drug Discovery”, Science, 303, 1813
(2004)
Today, what systems will we talk about?
 Proteins are formed by a reservoir of 20 amino acids. Amino acids are
related by peptidic bondings to form polypeptides
Backbone chain
Side chain
Residue 1
Residue 2 Residue3
Peptide link: rigid, planar
Formation of peptide link by
condensation and elimination of
water
Only certain values of the Ramachadran angles
f and Y are possible
Hydrogen Bond
structure Primaire
Secondaire
Tertiaire
Quaternaire
g turns
b feuillets
a helice
Primary structure
Systems between 1 to 5 amino
acids residues (up to a few
hundred Daltons)
Optical spectroscopic techniques
(microwave, millimeter wave,
terahertz, infrared, UV/visible).
Determination of effective neutral
molecular structures
Comparison with quantum
mechanical calculations at
equilibrium.
Advantages : functional-group and
conformational specificity.
Challenge : getting good signal-tonoise
Secondary
5 to 30 amino-acid residues
Tertiary
Quaternary
Above about 30 amino acids
Experimental measurement
of electric dipole moment or
diffusion velocity in a gas
(« ion mobility »)
Mass spectrometry
Such measurements can be
coupled with “hybrid”
calculation methods
Advantages: Many proteins acquire
their secondary or tertiary structures
when they bond.
Advantages: peptidic
« maps » dipoles/mass to
identify proteins
Challenge: Mass spectrometry does
not give structures directly.
For macromolecular systems,
modelling using a classical force
field (AMBER and CHARMM
softwares).
Challenge: need a structure
calculation
Determination of complexation of
the biomolecule by ligands.
Coupling mass spectrometry with spectroscopy (Oomens, Meyer et al, 2005, Kapota , Maïtre, Ohanessian
et al JACS, 2004).IR/UV or UV/UVhole-burning spectroscopy (Mons et al, Zwier et al, Gerhards et al, Simons
et al…)
Hydrogen Bond & Torsion
 Secondary and tertiary structure of proteins
How is Microwave spectroscopy at high
resolution going to contribute ??? :
Internal rotation splittings can be used to obtain the
structure/folding of molecules in gas phase
WITHOUT doing isotopic substitution.
Lavrich et al. JCP 2003
What is internal rotation?
Microwave is a good spectral range to determine very accurately molecular
structures but the size of the molecule is limited
Erot
JKaKc
110
111
101
Rigid rotor (zero order):
Asymmetric top, rotation
structure characterized by the
quantum numbers J, Ka, Kc:
000

B+C
A+C
A+B
Limit size of the
molecule
Detected by most of
FourierTransform
spectrometers
(4-20 GHz) : 250-300
uma
Peptidic bonding and torsion :
a few examples of molecules studied in MW
 Formamide
Astrophysical detection: Rubin et al, ApJ 1971,
Brown et al JMS 1987
 Acetamide
Potential Barrier : V3 = 25 cm-1 ; Ilyushin et al, JMS
2003, 1 top low barrier, Cs frame
Astrophysical detection : « The Largest Interstellar
Molecule with a Peptide Bond », Hollis et al, ApJ 643
2006, L25
 N-methylformamide
Potential Barrier: V3 cis =60 cm-1, V3 trans = 279 cm-1,
Kawashima, et al (Columbus 2002),
Fantoni, Caminati, J. Mol. Struct., 2002
Examples :
Acetamide derivatives
 N-Methylacetamide
V3(1)=36 cm-1, V3(2)=42 cm-1 ;
Ohashi, Hougen et al JMS 2004
2 tops problem, Cs frame
 N-Methylpropionamide
V3(1)=796 cm-1, V3(2)=81 cm-1 ;
Kawashima, Hirota et al JMS 2003
 N-Ethylacetamide
V3= 75 cm-1 ; Kawashima (Dijon 2003)
This talk: dipeptidic derivatives
Collaboration NIST (Gaithersburg, USA), PhLAM (Lille)
 N-acetyl-alanine N’-
methylamide (AAMA)
V3(1)=98 cm-1, V3(2)=81 cm-1
Observé
Lavrich et al. JCP 2003
 Ethyl AcetamidoAcetate
(EAA) or N acetylglycine
ethyl ester
Alanine Dipeptide
Methyl Ester (ADME)
Methylcarbamate
Collaboration with Institute for Radioastronomy of NASU (Ukraine),
PhLAM (Lille), University of Eotvos (Budapest)
METHYLCARBAMATE : isomer
of glycine,
Plausible candidate for an astrophysical detection
because more stable than glycine
Glycine
Rotation-torsion MW spectrum:
Ilyushin et al.,
J. Mol. Spectrosc., 240, 127 (2006).
NH2COOCH3
Good candidate for validation of high level quantum chemical
Calculations: Equilibrium vs. Ground-State Planarity of the CONH Linkage ? Demaison et
al., J. Phys. Chem. A., 111, 2574-2586 (2007).
HOW TO MODEL INTERNAL ROTATION?
For one C3v top, and a frame with a plane of symetry Cs
HRAM = Htor + Hrot + Hd.c + Hint
1) Diagonalization of the torsional part of the Hamiltonian in an
axis system where torsion-rotation coupling is minimal (Rho Axis
Method, RAM), Kirtman et al, Lees and Baker , Herbst et al:
Htor= F (pa - r.Jz)2 + V(a)
F: internal rotation constant
r depends on Itop/Imolecule
Eigenvalues = torsional energies
2) Eigenvectors are used to set up the matrix of the rest of the
Hamiltonian:
Hrot
= ARAMJa2 + BRAMJb2 +CRAMJc2 + Dab(JaJb + JbJa)
Hd.c usual centrifugal distorsion terms
Hint higher order torsional-rotational interactions terms : cos3a et
pa and global rotational operators like Ja, Jb , Jc
Theoretical Model: the global approach
RAM = Rho Axis Method (axis system) for a Cs (plane) frame
HRAM = Hrot + Htor + Hint + Hd.c.
Torsional operators and potential function V(a)
Rotational Operators
Constants
1
1-cos3a p2a Japa
1-cos6a p4a
Jap3a
V3/2
F
r
V6/2
k4
k3
J2 (B+C)/2*
Fv
Gv
Lv
Nv
Mv
k3J
Ja2 A-(B+C)/2*
k5
k2
k1
K2
K1
k3K
Jb2 - Jc2 (B-C)/2*
c2
c1
c4
c11
c3
c12
JaJb+JbJa Dab or Eab
dab
Dab
dab
dab6
DDab
ddab
1
Kirtman et al 1962
Lees and Baker, 1968
Herbst et al 1986
a = angle of torsion, r = couples internal rotation and global rotation, ratio
of the moment of inertia of the top and the moment of inertia of the whole
molecule
Hougen, Kleiner, Godefroid JMS 1994
Internal Rotation Programs
http://info.ifpan.edu.pl/~kisiel/prospe.htm: programs for rotational spectroscopy (Z. Kiesel)
Name authors
what it does?
Method
_______________________________________________________________________
XIAM
Hartwig
up to 3 sym tops
combined RAM-PAM
Maeder
up to one quad.
(based on Woods method)
nucleus
Separate vt fit, sometimes
separate A and E fits
_______________________________________________________________________
ERHAM Groner
one and two
Effective, combined RAM-PAM
internal rotors
Separate vt states fit
of sym.C3v or C2v
J up to 120.
acetone,diMEether
8191 lines max
MeCarbamate
intensities
________________________________________________________________________
BELGI Kleiner
one C3v internal
RAM method
Godefroid,
rotor. Frame can
Global fit of vt states
Hougen
Cs or C1
A and E species fit together
Xu, Ortigoso,
J up to 70
Ilyushin,
vt up to 11
acetaldehyde, acetic acid
Carvajal
intensities
acetamide,MeFormate
1 or 2 different
MeCarbamate, EAA
vibrational states
dipeptide alanine ester
Internal Rotation Programs (suite)
Name authors
what it does?
Method
______________________________________________________________________
JB95
Plusquellic
one internal rotor
PAM
Separate vt states, separate A
and E fits
graphical interface
alanine dipeptide
and many other molecules
http://physics.nist.gov/Divisions/Div844/facilities/uvs/jb95userguide.htm
______________________________________________________________________
SPFIT/ Pickett
one or two internal
Combined RAM-PAM
SPCAT
rotors, sym or asym.
Separate vt states, separate A
and E fits
propane, pyruvic acid
acetaldehyde (more recent)
______________________________________________________________________
Results :
Ethyl AcetamidoAcetate
1.
R. J. Lavrich, A. R. Hight Walker, D. F. Plusquellic, I. Kleiner, R. D.
Suenram, J. T. Hougen, and G. T. Fraser, JCP 119 (2003) 5497
 Experimental problems :
Biomolecules Properties
Spectrometer MWFT NIST (9-18 GHz)
Liquid or solid
Injection with reservoir nozzle
Low vapor pressure
Heated reservoir nozzle (135-155°C)
Thermal instability
Injection with inert material
Multi-conformations
Jet at 1K to simplify the spectra
Internal rotation splittings
Large spectral range investigated
Nitrogen quadrupole
Synthesis of 15N isotopomers
Microwave spectra of EAA
T = 150°C
Two conformers
identificated : CI and
CII
CII: « non planar
?
Structures MP2/6-311G(d,p)
CI : « planar »
EAA (15N) : a good case for comparing
the JB95 and BELGI codes
J up to 20, K up to 6
JB95
«High barrier, perturbative approach»
BELGI
« Global approach »
CI
160 A lines, rms = 1.7 kHz
197 E lines, rms = 1.8 kHz
160A+197E lines, rms = 1.8 kHz
CII
165 A lines, rms = 1.4 kHz
203 E lines, rms = 1.3 kHz
165A+203E lines, rms = 1.7 kHz
For the CII conformer (non-planar), a C1 global code was written (JCP 119, 5505 (2003)
EAA: CH3 group orientations in PAS
V3(1) determined ; V3(2) too high, not determined
BELGI
A,B,C
(EAA)
JB95
BELGI
JB95
Comparisons with
ab initio calculations
 do not predict the correct experimentally observed energy ordering for the two
conformers !  problem of data basis/method ? : MP2/6-311G(d,p)
Ab initio qcalc
qcalc-qobs
planar non planar
Alanine Dipeptide Methyl Ester
 I. Kleiner, J. Demaison, D. F. Plusquellic, R. D.
Suenram, R. J. Lavrich, F. J. Lovas, G. T. Fraser, V.
V. Ilyushin, JCP (2006)
 Theoretical problems:
 Develop new models for molecules which has no
plane of symmetry for the frame(1) AND have more
than one methyl internal rotation groups
 Deal with the hyperfine structure
 Deduce structural informations and compare them
with the ab initio calculations results
(1) I. Kleiner and J.T. Hougen, J. Chem. Phys. 119 (2003) 5505, voir EAA.
ADME: 2methyl tops
Fits: for each internal rotor about 120 lines RMS: 2 kHz
 N-methylacetamide: N. Ohashi, J. T. Hougen, R. D. Suenram, F. J. Lovas, Y.
V3(3) high
Kawashima, M. Fujitake, and J. Pyka, JMS
JKaKc
 3 sets of torsional splittings:
V3 = 68 cm-1
D1 = 2 cm-1
 (AA,AE). V3 = 400 cm-1

(AA,EA).
D2 = 0.01 cm-1

Interaction
between the 2 tops: very
small splittings. NOT
TREATED
(AA,EE).
ADME MW spectrum
Experimentally deduced
molecular parameters for ADME


Good agreement between the global and perturbation approaches
Torsional parameters better determined when V3 is smaller
AR / MHz
BR / MHz
CR / MHz
Rot.
Tors
Eab / cm-1 
Ebc / cm-1
Eac / cm-1
Global Fit :BELGI
(AA,EA) States (AA,AE) States
LOW Barrier
HIGH Barrier
2998.7(2)
669.23(6)
596.97(2)
-0.0163809(2)
0.000528077(6)
0.fixed
-0.00714(9)
0.00031(6)
0.fixed
ρ
V3 / cm-1
F / cm-1
0.013375(1)
64.96(4)
5.341(2)
0.01768(2)
396.45(7)
5.30fixed
θa / °
θb / °
θc / °
44.86(1)
46.75(1)
80.36(1)
22.5(1.5)
67.7(1.0)
87.0(3.5)
PERTUR. JB95
(AA,EA) States (AA,AE) States
LOW Barrier
HIGH Barrier
2998.1(7)
670.05(8)
596.29(2)
-0.0147019(1)
0.0019754(1)
0.00717fixed
0.01336(5)
66.35(5)
5.30fixed
44.94(5)
46.59(5)
80.58(5)
0.fixed
0.fixed
0.fixed
0.01719(1)
402(4)
5.30fixed
25.0(7.7)
65.2(8.5)
86.4(1.2)
Conformational searches, Structure and hydrogen bond
 13 stable conformers of ADME located, full geometry optimisations with
B3LYP/6-31G(d) et G3MP2B3
 Comparison of ab initio structure for AAMA (alanine dipeptide) et ADME
(N-acetyl alanine methyl ester)
AAMA
ADME
φ
ψ
C5
C7
Ramachandran angles
Ψ 75°
φ -82°
Similar to a g-turn structure
Ramachandran angles
Ψ 171°
φ -159°
Similar to a b-sheet structure
Ab initio calculations : structural
comparisons of ADME
φ
ψ
MP2 et B3LYP: base cc-p-VTZ,
Gaussian03 ; PW91 et HCTH: double
numerical basis, DMol
Expt.
MP2
B3LYP
PW91
HCTH
A / MHz
2998.4(3)
-16.1
+54.6
+114.0
+138.6
B / MHz
669.6(4)
-5.6
+2.4
+3.1
+5.2
C / MHz
596.6(3)
-2.9
+1.6
+1.3
-1.8
θa / °
d
44.9(1)
+1.9
+2.1
+1.9
+3.1
θb / °
d
46.7(1)
-1.4
-1.8
-1.4
-2.9
θc / °
d
80.5(1)
-1.4
-1.0
-1.5
-1
V3 / cm
θa / °
θb / °
d
θc / °
d
d
-1
V3 / cm
φ/°
ψ/°
rN-H---O=C
θN-H---O=C
V3 / cm-1
65.6(7)
-2.7
+20.1
23.7(1.2)
66.5(1.2)
-2.8
+3.0
-2.0
+2.2
-2.9
+3.1
86.7(3)
-1.6
-3.1
-3.2
399.2(3.0)
+0.3
-159.4
171.1
2.218
105.4
+145.6
-155.5
169.8
2.239
105.2
-22.2
+166.3
-0.4
e
-52.6
e
-1.2
+1.4
-2.2
e
-153.3
169.2
2.234
105.7
+99.9
e
-150.8
166.9
2.291
103.9
1229.7
DFT (B3LYP) gives rotational constants too small and MP2 too big.
DFT overestimates the structure, MP2 underestimate it !
Methylcarbamate
Equilibrium structure of Methyl carbamate is not planar!
Method
B3LYP B3LYP
MP2_FC
CCSD(T)_AE
Basis
VTZ
AVTZ 6-311
VTZ AVTZ VQZ V(D,T)Z
---------------------------------------------------------------------------------------------------
H9N1C2O3
Method
Basis
13.12 10.18 12.59 17.59 16.02 15.88 16.52
a
b
c
tot
0.163(2)
2.294(9)
0a
2.300(9)
VTZ
0.222
2.412
0.757
2.538
AVTZ
0.204
2.462
0.671
2.560
VQZ
0.238
2.459
0.673
2.560
MP2
6-31G*
0.115
2.089
0.862c
2.263
CCSD(T)
B3LYP
V(T,D)Z
VTZ
0.234
0.347
2.215
2.353
0.710
0.512
2.338
2.433
AVTZ
0.200
2.410
0.374
2.447
Exp.
MP2
Ground state is planar: no out-of-plane terms needed to
fit the spectrum, no c type transitions, c = 0
Ilyushin, Alekseev, Demaison, Kleiner JMS 2006
J up to 20, Ka up to 10
Methyl Carbamate
Syn configuration
Equilibrium vs. Ground-State Planarity of the CONH Linkage ?
Jean Demaison, Attila G. Császár, Isabelle Kleiner, and Harald Møllendald
Formamide (X = Y = H), carbamic acid (X = OH, Y = H), urea (X = NH2, Y = H),
acetamide (X = CH3, Y = H), and methyl carbamate (MC, X = OCH3, Y = H): all except
formamide have a pyramidalized N at equilibrium with a very small inversion barrier !
The effective structure (ground state) (determined by experimental microwave work)
is however planar
ALL ab initio optimizations indicate that the amide group is non planar
(difference between planar and non planar is 53 cm-1 CCSD(T)/V(T,D)Z
in apparent contradiction with experimental results (c is zero)
WHAT’s GOING ON?
MC behaves like other molecules containing the amino group:
small barrier between planar and non planar and the ground torsional state
is above this barrier.
Kydd and Rauk,
J. Mol. Struct.
1981
Conclusions : EAA and ADME
 The internal rotation splittings in vt = 0 from different
peptide mimetics containing one or more CH3 groups
have been analyzed with two different theoretical
methods : “perturbative” and “global ” .
 Spectroscopic results were compared to quantum
chemical calculations.



Very good agreement for the internal rotor with a low
potential barrier (larger splittings)
Care for conclusions concerning the CH3 with a high
barrier as no excited torsional states measured
(small spittings, thus spectroscopic parameters less
well determined). Higher order terms not taken into
account
Ab initio calculations relatively more precise for higher
barriers; the choice of methods/bases must be
pertinent.
Conclusions: validation of ab initio
calculations
 Torsional barriers at the MP2/cc-pVTZ level are in good agreement
with experimental values. DFT barriers are 8 to 80% off!
 DFT overestimates the structure, MP2 underestimates (same
discrepancy found with crystalline peptides : trialanine, THz
absorption spectrum agrees with X-ray but not with DFT
calculations, Siegrist et al, JACS, 128, 5764, 2006)
Ab initio calculations at high level are very useful for
Spectroscopists, since they can calculate precisely internal
rotation parameters
High resolution spectroscopy can be used to guide the
choice/optimization of ab initio calculations!
Conclusions : methyl carbamate
 formamide should not be considered as a
general model of the amide linkage !
several molecules containing the CONH linkage
seem to have a pyramidalized nitrogen at
equilibrium and a double-minimum inversion
potential with a very small inversion barrier
allowing for an effectively planar groundstate structure
Acetamide or methyl carbamate : good model
for this
UNDER COURSE : Trans and gauche conformer of
ethyl acetate.
Collaboration with Institute of Physical Chemistry, RWTH Aachen (Germany)
W. Stahl, L. Nguyen, D. Jelisavac, L. Sutikdja, D. Cortés Gómez , H. Mouhib
gauche conformer
trans conformer
Very few esters (even simple) have been studied so far by MW spectroscopy:
- many atoms for isotopic substitution
- Large internal rotation splittings
- Different conformers
Jelisavac et al. JMS 2009
Under course :Microwave Study of Phenyl Alanine Methyl
Ester: Reducing the Complexity of Confomational Searches
Douglass, Roe, Plusquellic, Pratt and Pate
Previous works: IR-R2PI spectroscopy
and DFT ab initio (Gerharts et al)
γLg+
E
βL(g-)
βL(a)
δd(g+)
βL(g+)
Lowest energy conformers MP2/6-311++G**
Now:
- mini-FTMW (NIST) :12-18 GHz
- Semi-Confocal Chirped-Pulse FTMW :
12.6-18 GHz, makes possible the
recording of the complete microwave
spectrum of a gas phase sample
using a single 1 μs pulse.
-assignment of overlapping sub-bands :
genetic algorithms (L. Meerts)
Perspectives:
towards larger biomimetic molecules?
 Experimental challenge:
-nondestructively vaporizing fragile biomimetics: laser
ablation
 Theoretical challenge :
-extend present modeling using effective Hamiltonians and
codes to describe more complicated system (containing
two or more internal rotors CH3).
Methyl acetate CH3COOCH3 : collaboration with Jon Hougen
Sonia Melandri (Bologna), Lilian Sutikdja
….
-transfer the information obtained by gas phase MW
high resolution spectroscopy to biomolecules in a
cell environnement!
National Institute For
Standards And Technology
(NIST, USA)
Jon Hougen
David Plusquellic
Richard Lavrich
Richard Suenram
Frank Lovas
Gerald Fraser
Angela Hight Walker
Laboratory of Molecular Spectroscopy
(Budapest, Hungary)
Attila G. Császár
Laboratoire de Physique des
Lasers, Atomes, et Molécules
(Lille, France)
Jean Demaison, L. Margulès, Th. Huet,
R. Motyenko, M. Tudorie
Institute of Radio Astronomy of NASU
(Kharkov, Ukraine)
Vadim Ilyushin
Physical Chemistry, RWTH Aachen (Germany)
Eugene Alekseev
W. Stahl
L. Nguyen
D. Jelisavac
L. Sutikdja
D. Cortés Gómez , H. Mouhib