Organic molecules in aqueous solution, ranging from contaminants and pollutants to biomedical markers for cancer diagnostics, are nowadays detected with compact, automated laser sensors based on evanescent waves in the visible range, however these molecules are not easily identified by state-of-the-art photonic devices. The original H2020 Key Enabling Technology proposal (KET-Photonics call) "WaterMirror" aimed at solving this issue by introducing infrared vibrational spectroscopy sensors bases on evanescent waves and tunable quantum cascade lasers, capable of identifying molecules by their vibrational infrared spectrum. The present project aims at executing at Sapienza University of Rome the Work Package 1 of the H2020 proposal WaterMirror.
An evanescent-wave sensor for mid-infrared wavelengths (from 5 to 11 micrometers), based on one-dimensional photonic crystals, will be designed, fabricated and tested in our laboratories (located in two collaborating Departments) with the available quantum cascade laser systems and microbolometer detector arrays. The anisotropic conformational changes of proteins of biomedical interest will be detected with the new sensor in an aqueous environment, controlled by an available microfluidics platform. This will bring the Technology Readiness Level of WaterMirror from 3 (proof-of-concept) to 4 (system validation in the laboratory). Other European partners will be involved and kept up-to-date for future H2020 proposals.
The present project will broaden the wavelength range of photonics technologies for a wide variety of future molecule-specific sensing applications in security, health and environmental safety.
The structural analysis of proteins is a field of biochemistry of high relevance for biology and medicine. On one side, protein structures and aminoacid sequences are determined by e.g. X-ray protein crystallography and other well established techniques. On the other side, the spectroscopic characterization "in vitro" of the folding state of the aminoacid chains in their natural aqueous environment is a necessary (and much less developed) complementary technique. In particular, optical spectroscopic techniques, including fluorescence, FRET, Raman and infrared spectroscopy, can monitor the oxidation state of cofactors and the dynamic changes of the protein folding upon physico-chemical perturbations of the environment.
Mid-infrared (MIR) absorption spectroscopy is routinely employed for the study of the secondary structure of proteins in liquid solutions. The amide I band (1600-1740 cm-1, wavelength around 6 microns) and the amide II band (1470-1580 cm-1, wavelengths around 7 microns) are respectively assigned to the C=O stretching and N-H in-phase bending vibrations of the peptide bonds. Different patterns of hydrogen bonding and dipole-dipole interactions within the alpha-helices, beta-sheets, beta-turns and random coil structures are known to result into different wavelength distribution of the C=O and N-H vibrational absorption lineshapes that, in turn, can be correlated with the individual secondary structure of the protein moieties by MIR spectroscopy on purified protein solutions. The state of the art, however, is represented by differential absorption spectroscopy as the main technique to investigate biologically relevant protein conformational changes in realistic biological fluids. The protein functional processes involve only a few aminoacids and cofactors over several thousands, and therefore display variations of the MIR spectrum well below the typical 0,1% absorption threshold of standard MIR spectroscopy in purified solutions. In modern MIR differential spectroscopy experiments performed on short timescales, indeed, only the structural changes are detected, while the constant structural vibration background is instead rejected [for a recent review see e.g. T. Kottke et al. "The Grateful Infrared: Sequential Protein Structural Changes Resolved by Infrared Difference Spectroscopy" 2017 highlights of J. Phys. Chem. B 121, 335-350]. At present, such fine information on conformational changes is obtained only by fluorescent labeling of proteins (obtained by genetic expression of green fluorescent proteins (GFP) or by attachment of dyes) and by detection of the FRET effect, which indirectly correlates with the aminoacid chain position and orientation. In MIR spectroscopy, this can be done in a label-free fashion.
The sensor to be developed in this project goes beyond the state of the art as it will exploit the label-free detection based on the real-time shift of the MIR spectral resonance position of Bloch surface waves (BSW), directly related to the change of the real part of the complex refractive index in the MIR. Sensitivity better than 0.001% relative refractive index change has been demonstrated in the visible in our FP7 project BILOBA, and it is expected in the MIR as well. Also, variations of the resonance width are directly related to changes of the extinction coefficient in the MIR hence to the vibrational absorption lineshape. The sensor capability to sustain simultaneously both TE and TM polarized BSW (the two orthogonally polarized modes of the electromagnetic field at surface), together with the directional immobilization technology based on surface functionalization by antibodies, will permit to explore changes in the protein conformational anisotropy (the relative orientation of alpha-helices, parallel and anti-parallel beta-sheets relative to the sensor surface). Birefringence and dichroic absorption experiments with MIR BSW will represent new ways of performing differential spectroscopy. The timescale of the laser/microbolometer sensor developed in our FP7 project GEMINI will enable the observation of conformation kinetics in the millisecond range. Time-resolved spectroscopy indeed represents the most sophisticated version of differential spectroscopy. In the near future, the rapid development of quantum chemistry computation may render MIR differential spectroscopy a label-free direct probe of the conformational changes, as its spectral assignments are by far easier to calculate than Raman or NMR ones, due to the linear dipole absorption mechanism in the MIR.
In summary, MIR differential spectroscopy has a clearly recognized potential for label-free detection of protein conformational changes, prevented by technical limitations of MIR photonics that, here, we will contribute to surpass.