Evaluation of protein functionality using the LATEX-CRP kit
The evaluation was performed at different CRP concentrations, namely: 8 mg / L; 16 mg / L and 32,5 mg / L. The samples were initially tested (T0) for verification and a strong agglutination was observed at 16 mg / L and 32,5 mg / L and its absence at 8 mg / L (Fig. 3).
The tests revealed that in the sugar free samples, the protein is functional at 40oC for at most 48h, at the highest concentration (32,5 mg / L), after which it suffers massive degradation and loss of immunogenicity. Instead, samples supplemented with 1M sucrose showed superior thermostability, the protein being intact and functional even after 20 days of incubation at 40oC. A series of studies have shown that covalent binding of glycans to amino acid side chains of a protein can give it high thermostability (Shirke et al, 2016; Barb et al., 2010; Culyba et al., 2011).
Some of the CRP-sucrose solutions were sent for irradiation within the National Research-Development Institute for Physics and Nuclear Engineering "Horia Hulubei" and subjected to 3 degrees of gamma irradiation: 2 kGy, 4kGy and 6 kGy.
Although the agglutination was weaker for the higher irradiation doses, the samples subjected to irradiation with 2 kGy proved to be even more stable than the non-irradiated ones. The tests revealed that the irradiated sugar-protein solutions were still functional and stable after 22 days incubation in stress conditions (Fig.4).
Fourier-Transform Infrared Spectroscopy (FTIR)
Fig. 6 compares the blanks’ spectra for the two types of samples, after the spectral subtraction of water. The common bands of the two samples can be observed, such as 3387 cm-1 coming from the stretching vibration (ν) of the O-H group; 2363 cm-1 and 2334 cm-1 generated by the stretching vibrations of CO2, NO2; 2043 cm-1, from sodium azide; 1637 cm-1, specific for ascorbic acid and 1078 cm-1 from the C-O stretching vibration.
Regarding C-reactive protein (CRP) and the positive control “Spinreact” (Fig. 6 and Fig. 7), these are similar at the molecular level, but “Spinreact” has two additional bands at 1411 cm-1 (δCH) and 1332 cm-1 (νC-C / νC-O). As for the common bands, there should be highlighted those from 1545 cm-1 and 1640 cm-1 which are produced due to amide II, respectively amide I and the band from 2051 cm-1 from a thiol group νS-H. Samples’ ATR-FTIR spectra are presented after subtracting the blanks’ spectra (Fig. 5). Common bands with those of the CRP standard (2360, 2340, 1640, 1545 cm-1) can be observed, which attests the presence of the protein in the formulations. It is known from literature that the bands corresponding to amides I and II are the major bands in the IR spectrum of proteins. Amides I and II specific peak values correspond to those specified in the literature, namely, ͂ 1650 cm-1 for amide I and ~1550 cm-1 for amide II (Yang et al., 2015; Tatulian, 2019). In water, the bands corresponding to amide II are those around 1570 – 1540 cm-1 and include a significant contribution of the binding angle of the NH group, together with CN changes in the binding distance and other vibration-induced amide group changes, thus being extremely sensitive to the kitetics of the hydrogen atoms exchange (Tatulian, 2019).
The difference in the amide I and II specific peak amplitude between the CRP standard and the two samples (P1 and P2) reflects a lower protein concentration in the case of the two formulations. There is also a big difference between the bands related to amides I and II of the control “Spinreact” and those obtained in the case of the two formulations (Fig. 8).
CRP could generate the endogenous fluorescence because it contains Tryptophan (Trp) and tyrosine (Tyr) and phenylalanine (Phe) residues. And, the fluorescence character of CRP is mainly produced by Tryptophan residue as the fluorescence intensity ratio of Trp, Tyr and Phe is 100: 9: 0.5 (Han et al., 2012).
The binding mechanism between sucrose and CRP was followed by steady-state and time resolved fluorescence. Results show that the intensity of fluorescence is significantly influenced by the concentration of sugar in the solution (Figure 9A).
The fluorescence intensity of CRP at 339 nm gradually decreased with the addition amounts of sucrose. This revealed that sucrose could quench intrinsic fluorescence of CRP through producing a non-fluorescent complex between CRP and sucrose. Also, fluorescence lifetime slightly decreases when increasing the sucrose concentration (Figure 9B).
The intensity peak of fluorescence decreases with increasing concentrations of sucrose in the environment, so that at 500 mM it reaches about 73 % of the value obtained for the sample not supplemented with sucrose (Figure 10A).
The fluorescence lifetime decreases from ≈ 7.31 ns, for CRP in buffer, with ≈ 15 % in presence of 500 mM of sucrose, to approximately 6.26 ns (Figure 10B).
To highlight possible quenching mechanisms that occur in the interaction between CRP and sucrose, the fluorescence quenching experiments were carried out, at 25 °C, and an attempt was made to obtain the quenching constant from Stern–Volmer equation (Lakowicz et al., 2006; Valeur et al., 2001; Shi et al., 2014):
where F and F0 are the fluorescence intensities with or without sucrose, respectively, τ and τ0 are the mean fluorescence lifetimes of Tryptophan residues, from CRP structure, with or without sucrose, [Q] is the concentration of sucrose, KSV is the quenching constant (Stern-Volmer constant), kb is the quenching rate constant of protein (bimolecular constant) and τ0 is the average fluorescence lifetime of CRP without sucrose and its value is ≈ 7.31 ns, experimentally determined.
In general, two types of fluorescence quenching can be distinguished: dynamic quenching and static quenching. The static and dynamic quenching can be differentiated by analysis of Stern-Volmer plots described by previous equation and, in our case, represented in Figure 11A for steady state fluorescence and Figure 11B for fluorescence lifetime.
When only dynamic quenching or static quenching occur, the Stern-Volmer plots of steady state fluorescence data should be a straight line with a slope between 0 and 1. Static quenching can be distinguished from dynamic quenching using Stern-Volmer plot of fluorescence lifetimes, in that case (static quenching) the slope of graph being equal to 0 (Lakowicz et al., 2012; Valeur et al., 2001).
In our case, quenching of Tryptophan, from CRP primary structure, by sucrose presented a pronounced downward curvature, this implying two or more classes of fluorophores with different degrees of accessibility for quencher molecules (Figure 3A). We tested, using the obtained data, the method of analysis of fluorescence quenching using the modified Stern-Volmer plots (data not shown) (Lakowicz, 2006), plotting the variation of (F0/F0 – F) dependence by (1/[Q]), but also within this analysis method the obtained graph showed a downward curvature. This implies that there are more than two classes of tryptophan molecules with different degrees of accessibility for sucrose, CRP protein having in his primary structure six Tryptophan residues (Protein Data Bank).
Our results are in opposition to those obtained in another study, in which glucose and glycerol were added to samples with human glucokinase, and glucose binding seems to determine an increase of the fluorescence quantum yield, of almost 2 times (Zelent et. al, 2017). This may be due to factors such as protein type, amino acid composition, but also selected sugar. On the other hand, other researchers studying the interaction between human serum albumin (HAS) and 2–amino-6-hydroxy–4–(4-N, N-dimethylaminophenyl)-pyrimidine-5-carbonitrile (AHDMAPPC), have reported that the fluorescence intensity reduced gradually, as the concentration of AHDMAPPC got more and more higher in the sample. They concluded that the effect of extinguishing fluorescence was due to the formation of a non-fluorescent complex (Suryawanshi et al., 2016).
Other studies have also reported that high concentrations of sugars such us sucrose and glucose can increase thermostability, also mentioning that a stronger stabilization effect was observed in sucrose samples (Oshima and Kinoshita, 2013).
In an attempt to establish the molecular mechanism by which fluorescence is extinguished, we used the 1-click docking software from “mcule”. Both molecules were loaded in the database (fig. 12) and after selecting the proper binding center, a series of docking experiments were done.
The first docking had a score of -3.2 (Fig. 13) and the second one -4.9 (Fig. 14), and revealed that sucrose was bound near TRP5. The more negative the docking score is, the better the match. Since there are a number of tryptophans located on the outside of the molecule (TRP187, TRP524, TRP605, etc.) these simulations suggest that extinguishing fluorescence with increasing sucrose concentration may be correlated with blocking the intrinsic fluorescence of tryptophan and perhaps other fluorescent aminoacids such as tyrosine and phenylalanine.
Compared to TYR and PHE, TRP is the most abundant and complex, exhibitting high sensitivity to the environment and more than two different fluorescent lifetimes (Ghisaidoobe and Chung, 2014). Studies show that proteins containing one TRP undergo multiexponential fluoresence decay. A hypothesis for this decay is the existence of different rotameric conformations of the aminoacid side chain (Millar, 1996).
Although we selected a series of binding centers located near TRP, the best fit appeared to be near the TRP5 region.