Molecular Rotational Resonance Spectroscopy - Chiral Analysis without Chromatography

A chiral molecule has a structure that cannot be superimposed upon its mirror image. Many of the physical properties of the mirror images, or enantiomers, are identical. However, the physical interactions between chiral molecules can be strongly influenced by the handedness of the molecules. These effects are of central importance when chemistry takes places in the homochiral environments of biological systems. In pharmaceutical applications this means that enantiomers of a drug can have different effects in the body. A significant fraction of small drug pharmaceuticals are chiral and there is a general goal of producing the final active pharmaceutical ingredient in an enantiopure form to improve drug potency, drug safety, or to extend intellectual property protection through the strategy of chiral switching.

Chiral analysis in pharmaceutical applications presents major challenges to analytical chemistry, especially in the case where the molecule has multiple asymmetric carbons. Each asymmetric carbon has a local geometry that can be either left or right handed, usually indicated using the R/S convention. Therefore, for a molecule with N chiral centres, there are 2N possible stereoisomers. In the most general case, there will be 2N-1 distinct molecular geometries, known as diastereomers, and each of these diastereomers is chiral so that it has a pair of enantiomers. An example is shown in Figure 1 for ohmefentanyl, an opioid analgesic, which has three chiral centres. This drug illustrates the challenges in pharmaceutical chemistry posed by chirality. The potency of the drug and its potential for addiction depend strongly on both the diastereomer geometry and enantiomeric form [1]. For example, the diastereomers labelled F9202 and F9204 are both highly potent, 1500 and 6400 times more potent than morphine, respectively. However, F9202 is found to be about 600 times less addictive in mice. The mirror images, or enantiomers, of these compounds are 2000 and 100 times less potent.
The ideal analytical technique for chiral analysis would be able to perform the following three measurements: 1) quantitative measurement of the yields of diastereomers produced in the synthesis of a molecule with multiple chiral centres, 2) quantitative measurement of the enantiomeric excess in a sample, and 3) accurate determination of the absolute configuration (or handedness) of the enantiomer that is in excess. Furthermore, rapid measurement times and the ability to perform the measurement directly on a complex reaction mixture would be desired. Currently there is no single technique that can meet this set of challenges and physical separation of the diastereomers and enantiomers by chromatography is still a generally necessary step of the analysis process.

In a recent landmark paper, Patterson, Schnell, and Doyle introduced a new enantiomer-selective measurement technique that uses molecular rotational resonance spectroscopy [1]. Subsequent experiments by these researchers have shown that three wave mixing rotational spectroscopy experiments are a sensitive method for measuring enantiomeric excess [3-6]. The implementation of these techniques in a manner similar to NMR spectroscopy techniques has been described theoretically [7] and was recently implemented experimentally [8]. Molecular rotational resonance spectroscopy, which performs spectroscopy on the quantised energy levels that result from the kinetic energy of overall molecular rotation, has been recognised as a sensitive probe of the molecular three-dimensional geometry for many years [9,10]. The spectral patterns are governed by the principal moments-of-inertia (expressed as the rotational constants: A, B, C [11]) and are, therefore, sensitive to the mass distribution in the molecule. Recent instrument advances have made it possible to study the structure of larger molecules using broadband, chirped-pulse Fourier transform rotational spectroscopy of molecules cooled to temperatures of about 2 K in the adiabatic expansion of a pulsed jet source [12,13]. The spectra recorded in these instruments have exceptionally high spectral resolution so that complex mixtures can be directly analysed. The combination of high-sensitivity, enantiomer-specific measurement capabilities, excellent chemical selectivity, and high spectral resolution for direct mixture analysis offers new opportunities for chiral analysis in chemistry.

Experimental
The results from a recent experiment showing enantiomer-specific spectroscopic signatures in solketal using three wave mixing Fourier transform molecular rotational resonance (FT-MRR) spectroscopy are presented [8]. Solketal was purchased from Sigma-Aldrich in both enantiomerically pure forms and as a racemic mixture. The sample was introduced into the spectrometer vacuum chamber by heating it and entraining the vapour in a flow of neon carrier gas at 1 atm. pressure. A pulsed valve creates a free jet expansion that produces a 2K gas sample. The rotational spectrum of solketal was measured using a CP-FTMW spectrometer as shown in Figure 2. The signal-to-noise ratio in the measurement is higher than 1,000:1 so that it was possible to analyse the spectra of the singly substituted 13C and 18O isotopologues in natural abundance and this information was used to determine the experimental structure of solketal that is shown in Figure 2.

The spectroscopic information from this measurement was used to perform a three wave mixing FT-MRR spectroscopy measurement using the principles introduced by Patterson, Schnell, and Doyle [2-6] and further developed by Grabow [7]. The basic physical principle of this measurement was presented by Giordmaine who showed that three wave mixing (i.e. sum and difference frequency generation) is possible from a bulk sample as long as it has an enantiomeric excess [14]. For rotational spectroscopy, the important physical insight is that a somewhat subtle feature can distinguish enantiomers: the product of the signs of the components of the molecular dipole moment along the principal axis system of molecular rotation is enantiomer-specific. This idea is illustrated in Figure 3.

Three wave mixing spectroscopy is used to create an emission signal that is proportional to this product so that the sign, or phase, of the emitted light identifies the enantiomer. This works because the molecular rotational spectrum is actually composed of three distinct spectra with intensities controlled by the three dipole moment vector components denoted ma, mb, and mc. These are called the a-type, b-type, and c-type spectra [11]. In order to generate the chiral emission signal, the electric field vectors of the two excitation pulses need to by orthogonal and the electric field polarisation of the coherent emission signal will be mutually orthogonal to both excitation pulse polarisations. The way that these measurement concepts are implemented is shown in Figures 4 and 5. First, the experiment requires a transition ‘cycle’ between three of the quantised rotational kinetic energy levels. The intensity of the transitions between each pair of energy levels, Figure 4, is governed by a different dipole moment component. The transition frequencies fall in the microwave region of the electromagnetic spectrum, and polarised excitation pulses can be created using horn antennae as shown in Figure 5.

The electric field polarisations of the two excitation pulses are orthogonal and the coherent signal generated through the three wave mixing process has an electric field that is orthogonal to both excitation pulses. The frequency of the emitted wave is the difference of the excitation frequencies in the level diagram of Figure 4. The coherent emission signal for the two different enantiopure samples of solketal are shown in Figure 6 where it is seen that they have opposite sign, or phase, as expected from the sign of the product of the dipole moment components. The Fourier transform of the coherent signals is also shown in Figure 6. This figure also shows the measurement using the racemic sample where no signal is generated.

Results and Discussion
The spectroscopic results demonstrate that three wave mixing rotational spectroscopy produces an enantiomer-specific signature through the phase of coherent emission signal. This signal is produced at the sum or difference frequency of the two, resonant microwave pulses used to excite the sample and has a similar magnitude to the normal, ‘achiral’ signal.

This result is an important difference between three wave mixing FT-MRR spectroscopy and better known techniques like vibrational circular dichroism (VCD) [15]. These optical rotation methods rely on the interference of electrical and magnetic dipole moment contributions to the spectral intensity and these signals are generally 10-4 of the intensity of the ‘achiral’ vibrational spectrum due to the small size of the magnetic dipole moment. In the FT-MRR measurement, detection at the sum or difference frequency occurs against zero background giving high sensitivity even for small enantiomeric excesses.

There is an important limitation in the current measurement: it has not yet been possible to achieve absolute phase calibration of the signal so that the absolute configuration, or handedness, of the enantiomer in excess cannot be directly determined. However, the current instrument does give a measurement phase that is reproducible day-to-day and this indicates that it will be possible to achieve absolute phase calibration in future instrument designs. There may also be possibilities to determine the absolute configuration using internal molecular standards [5].

This field of spectroscopy is in its infancy and the early experiments have selected molecular targets that do not push the limits of Fourier transform molecular rotational resonance spectroscopy. For the technique to have an impact in applications related to small drug discovery it will be necessary to perform chiral analysis on larger molecules. The general limitations on molecules for successful orally-delivered drugs summarised by the well-known Rule of Five of Lipinski [16] and augmented to consider molecular flexibility by Veber [17] are favourable to the technique. In general, small molecule drugs have molecular masses less than 500 Da. They are fairly polar, as conveyed through the general result that the octanol-water partition coefficient is not larger than 5. There is a general preference for molecules with limited numbers of flexible bonds thereby limiting the conformational complexity of the molecule.

As an illustration of what might be possible as this technique develops, we consider simulations of the molecular rotational resonance spectra of ohmefentanyl (Figure 1). A summary of the spectroscopic properties of the 8 stereoisomers that are obtained from quantum chemistry calculations [18] is given in Table 1. The rotational constants, related to the moments-of-inertia for the rotational motion, are identical for enantiomers (giving the same energy level structure) but differ for the diastereomers that have distinct geometries. The electric dipole moment varies for the diastereomers, especially in the vector components in the principal axis system of the molecular rotation. These results further reinforce the key physical insight that the sign of the product of the dipole moment components uniquely defines the enantiomer. Simulations of the spectra of the four diastereomers show that they would be easily identified in a measurement with a key feature being the change in the relative intensities of the three different a-, b-, and c-type spectra as shown in Figure 7. Furthermore, FT-MRR spectroscopy can fully resolve the spectra of all four diastereomers. In principle, quantitative diastereomer yields and enantiomeric excess measurements could be performed without the need for chromatography to separate the 8 stereoisomers.

Conclusions
A fundamental advance in physical chemistry has produced a new spectroscopy technique that can perform quantitative chiral analysis directly on complex chemical mixtures with the potential to make the measurements on time scales of minutes or less. As pointed out by Nafie in a companion article to the original discovery [19], this new field faces several challenges to create a useful tool for analytical chemistry but there is reason for optimism. Techniques will be required to volatilise large molecules, while retaining the ability to cool them through adiabatic expansion, so that they can be measured in the gas phase [20, 21]. When faced with a similar challenge the field of mass spectrometry produced volatilisation methods that gave incredible leaps in the power of the technique [22,23]. The rotational spectroscopy community needs to validate that quantum chemistry can accurately calculate the molecular parameters required in the analysis (the rotational constants obtained from the geometry and the electric dipole moment). In this case, however, the computational requirements are vastly lower than for the interpretation of VCD spectra [24]. Finally, like other methods including VCD [15,24] and NMR spectroscopy [25], the measurements and analysis will need to deal with conformational flexibility – an issue omitted from the ohmefentanyl discussion above. Conformational flexibility causes fewer problems in rotational spectroscopy because there is usually some cooling of the conformational energy in the pulsed jet expansion leading to spectral simplification [26] and the conformations are static on the time scale of the measurement so dynamic exchange effects are not present. As the rotational spectroscopy community works to develop this new technique, it welcomes input from the broader analytical chemistry community to help define the problems where the unique features of three wave mixing FT-MRR spectroscopy can solve problems that are unaddressed by current analytical chemistry techniques.

Acknowledgements
Measurements performed at the University of Virginia were supported by the National Science Foundation through grant CHE-0960074.

References
1. G.W. Guo, Y. He, W.Q. Jin, Y. Zou, Y.C. Zhu, and Z.Q. Chi, Life Sciences 67 (2000) 113.
2. D. Patterson, M. Schnell, and J.M. Doyle, Nature 497 (2013) 475.
3. D. Patterson and J.M. Doyle, Phys. Rev. Lett. 111 (2013) 023008.
4. V.A. Shubert, D. Schmitz, and M. Schnell, J. Mol. Spectrosc. 300 (2014) 31.
5. V.A. Shubert, D. Schmitz, D. Patterson, J.M. Doyle, and M. Schnell, Angew. Chem. Int. Ed. 53 (2014) 1152.
6. D. Patterson and M. Schnell, Phys. Chem. Chem. Phys. 16 (2014) 11114.
7. J.U. Grabow, Angew. Chem. Int. Ed. 52 (2013) 11698.
8. S. Lobsiger, C. Perez, L. Evangelisti , K. K. Lehmann, and B. H. Pate, J. Phys. Chem. Lett. 6 (2015) 196.
9. E. B. Wilson Jr, Science 162 (1968) 59.
10. B.H. Pate, Science 333 (2011) 947.
11. W. Gordy and R.L. Cook, R.L., Microwave Molecular Spectra, 3rd ed.; Wiley: New York, 1984.
12. G.G. Brown, B.C. Dian, K.O. Douglass, S.M. Geyer, and B.H. Pate, Rev. Sci. Instrum. 79 (2008) 053103
13. C. Pérez, S. Lobsiger, N.A. Seifert, D.P. Zaleski, B. Temelso, G.C. Shields, Z. Kisiel, and B.H. Pate, Chem. Phys. Lett. 571 (2013) 1.
14. J.A. Giordmaine, Phys. Rev. A. 138 (1965) 1599.
15. Y. He, B. Wang, R.K. Dukor, and L.A. Nafie, Applied Spectroscopy 65 (2011) 699.
16. C.A. Lipinski, F. Lombardo, B.W. Dominy, and P.J. Feeney, Adv. Drug Delivery Rev. 46, (2001) 3.
17. D.F. Veber, S.R. Johnson, H.-Y. Cheng, B.R. Smith, K.W. ward, and K.D. Kopple, J. Med. Chem. 45 (2002) 2615.
18. Gaussian 09, Revision D.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.
19. L.A. Nafie, Nature 497 (2013) 446.
20. F. Gamez, B. Martinez-Haya, S. Blanco, J.C. Lopez, and J.L. Alonso, J. Phys. Chem. Lett. 3 (2012) 482.
21. J.L. Alonso, M.A. Lozoya, I. Pena, J.C. Lopez, C. Cabezas, S. Mata, and S. Blanco, Chem. Sci. 5 (2014) 515.
22. J.B. Fenn, M. Mann,C.K. Meng, S.F. Wong, C.M. Whitehouse, C. M. Science 246 (1989) 64.
23. K. Tanaka, H. Waki,Y. Ido, S. Akita, Y. Yoshida, T. Yoshida, Rapid Commun. Mass Spectrom. 2 (1998) 151.
24. J. Shen, Y. Li, R. Vaz, and H. Izumi, J. Chem. Theory Comput. 8 (2102) 2762.
25. T.J. Wenzel and C.D. Chisholm, Prog. in NMR Spectrosc. 59, (2011) 1.
26. S. T. Shipman, J. L. Neill, R. D. Suenram, M. T. Muckle, and B. H. Pate, J. Phys. Chem. Lett. 2 (2011) 443.