Identification of binding interactions between myeloperoxidase and its antibody using SERS

Surface Enhanced Raman Spectroscopy (SERS) is a widely used spectroscopic method that can dramatically increase the sensitivity of Raman spectroscopy and has demonstrated significant benefit in the identification of biological molecules. We report the use of SERS in differentiating the bound immunocomplex of myeloperoxidase (MPO) and its antibody from the unbound complex and its individual components. The SERS signal was enabled by gold nanoparticles attached to MPO, pAb and their immunocomplex at an excitation wavelength of 785 nm. The obtained SERS spectrum of MPO is in agreement with previous literature. Comparative SERS spectrum analysis of MPO, pAb, and their immunocomplex reveals the significant peak shifts and intensity variations caused by the conformational changes due to the immunocomplex formation. Several key areas have been identified which correspond to specific amino acids being shielded from undergoing resonance while new amino acid residues are made visible in the SERS spectrum of the immunocomplex and could be a result of conformational binding. Our work demonstrates the capability of SERS to identify binding events and differentiate an immunocomplex from its unbound components with direct applications in biosensors.

Surface-enhanced Raman spectroscopy (SERS) provides a solution to this challenge by enhancing the signal intensity of Raman scattering of molecules adsorbed to roughened metal surfaces by as much as 10 15 [1]. Fleischman et al. [2] was the first to note this enhanced effect in 1973 while observing the adsorption of pyridine on a silver electrode. Later two groups, DOI: 10.5101/nml.v2i2.p74-82 http://www.nmletters.org Jeanmarie et al. [3] and Albrecht et al. [4] independently realized this as a unique phenomenon and proposed both an electromagnetic mechanism and a chemical mechanism, as theoretical explanations for the enhanced signal. Gold and silver nanoparticles are commonly used as SERS substrates with gold particles being used more frequently due to their high sensitivity and stability compared to silver compounds [5][6][7][8]. It must be noted that the enhanced Raman signal enabled by the plasmon resonance of gold and silver nanoparticles gave rise to new biosensors that have been used for identifying the binding region in proteins [9], and for studying binding affinity of immunoreactions [10]. Signal analysis of SERS was used to identify the native constituents of live epithelial cells employing endocytosed 60 nm gold nanoparticles [11]. SERS has also been used to differentiate bacteria from bacteriophages by conjugating them to 60nm gold nanoparticles [12] and in single molecule detection [13,14]. The combination of plasmon resonance based sensing with real-time SERS analysis could become a novel tool for interrogating the dynamics of protein binding interactions.
Applications of SERS in immunosensing include the successful detection of the thyroid stimulating hormone (TSH) [15], monitoring the immunocomplex formation between mouse IgG and goat anti-mouse IgG [16], detection of membrane bound enzymes within cells and correlation of prostaglandin-H-synthase (PGHS) antigen levels [17], and detection of conformational binding of anti-mouse IgG (bound to gold nanoparticles of 29.7 nm diameter) to mouse IgG antigen [18].
Myeloperoxidase (MPO) is a lysosomal protein found in neutrophilic granulocytes, often overexpressed in inflammatory diseases [19][20][21]  This baseline correction applied to all data sets helped smooth the data and allow an appropriate comparison. Following the baseline correction individual spectra were normalized and averaged. In order to remove the gold nanoparticle (AuNp) signature from the spectra, the average AuNp spectrum set was subtracted from averaged spectra of Ab, MPO and MPO-pAb.
The sum of Ab and MPO spectra was obtained by simply averaging the two spectra in Origin Pro. Origin 8.0's peak analyzer module was used to fit peaks to the averaged spectra.

RESULTS AND DISCUSSION
Raw SERS spectra of AuNp, Ab, MPO, and MPO-pAb are shown in Fig. 1a-1d, respectively. Individual spectra in Fig. 1 were sorted by their average intensity for easy visualization. The background spectra i.e. spectra of AuNp (see Fig. 1a we compared the spectra of AbMPO and the sum of spectra of pAb and MPO as shown in Fig. 3. Table 2 the peak positions of AbMPO along with the peaks that were found to be distinct for their immunocomplex. The peak positions that are unique for the immunocomplex are shown with the respective error bars (average of 80 spectra).
The SERS spectrum of MPO (see Fig. 2b) exhibits strong similarities to the Raman signals reported in previous studies [23,[28][29][30][31]. It is relatively weak below 1100 cm -1 , besides three small peaks between 646 cm -1 (ν 48/ ν 25 ) to 684 cm -1 (ν 7 ), at 835 cm -1 , and between 992 cm -1 (γ(CH)) and 1007 cm -1 (ν 45 ). Table 1 summarizes the MPO Raman peaks published by various studies.   (excitation wavelength of 660nm); h, from [23] (excitation wavelength of 454.5 nm); i, from [35], strong (excitation wavelength of 496.5 nm); j, from [35], observed or very strong in the spectrum (excitation wavelength of 568.2 nm); k, from [35], strong (excitation wavelength of 406.7 nm); l, from [35], strong (excitation wavelength of 514.5 nm); m, medium; s, strong; sh, sharp, w, weak; vw, very weak.  [23,[28][29][30][31]. Therefore the results obtained suggest that the key identifying MPO peaks are independent of the excitation wavelength, or the Raman method (RRS, CARS or SERS) that was used to collect the data. It is encouraging to also see that the SERS spectrum of the pAb has characteristic peaks similar to previously obtained Raman signals of an IgG. This is not surprising given the structural similarities of the antibodies. The complete list of peaks from Fig. 2 is presented in Table 2. Figure 2 compares the SERS spectra of MPO, pAb, and the immunocomplex of MPO bound to pAb and Table 2 lists and compares the peaks of each spectrum. Beginning at 500 cm -1 and working towards 1650 cm -1 we observe distinct peak shifts, altered intensities, and unique peaks when comparing MPO, pAb, and their immunocomplex. Signal intensity alone may not be sufficient to differentiate between bimolecular especially in the SERS mode (amplification factors and specific binding may alter the signal intensity significantly) [33]. It may rather be a combination of signal intensity and peak position that can provide a more reliable means for identifying a given sample.
The differences shown in Fig. 2  and MPO) as normalized average (see Fig. 3). Figure 3 compares the composite and immunocomplex spectra, and for ease of visualization the spectra was divided into 3 sections and presented as Fig. 3a, b,

CONCLUSIONS
Our results demonstrate that binding between an antigen (MPO) and its antibody gives rise to unique peaks absent in the  In summary, we investigated the potential application of SERS in differentiating the bound immunocomplex of an antigen and its antibody from the unbound complex and its components using myeloperoxidase as the model antigen. Obtained results indicate that the SERS spectrum of the immunocomplex is different from that of its parent antigen or antibody, and it is possible to identify conformational changes due to immunocomplex formation. Furthermore, the smallest RSE (18%, n=80) at 1465 cm -1 in the SERS spectra of the immunocomplex supports the notion that further investigation at this particular vibrational mode could provide valuable information towards application of SERS in understanding the changes that occur at a molecular level during binding interactions in biomolecules.