5-1. Establishment of activated fluorescent self-assembly AuNPs probes
5-1-1. Characteristic of AuNPs
Fluorescence activatable probes are comprised of the fluorophore (donor, D) and the quencher (acceptor, A) [Lakowicz, 2006]. The donor applied in this probe is FITC, and the quencher is AuNPs. RET occurs between a donor (D) molecule in the excited state and an acceptor (A) molecule in the ground state. As shown in Fig. 4-1, the FITC (D) molecules emit at shorter wavelengths (515 nm) that was almost overlapped by the absorption spectrum of the AuNPs (A).
There is the lowest scattering constant in spherical NPs below 40 nm in diameter and therefore they are with the highest potential to quench fluorescence. An effective distance below 2 nm between the AuNP and fluorophore demonstrates the best
quenching ability [Swierczewska et al., 2011]. Therefore, the AuNPs size for probes is desired to be 15 nm which below 40 nm. In this study, the synthesized AuNPs size is determined by UV-vis spectrum and applied in eq (5) [Haiss et al., 2007]:
𝑑 = exp (𝐵1𝐴𝐴𝑠𝑝𝑟
450− 𝐵2) (5) The numerical data was like given in Fig. 4-2A was then applied in eq (5), which
Aspr is the absorbance at the SPR peak, A450 is the absorbance at 450nm, B1 = 3.00 and B2 = 2.20. The diameter of synthesized AuNPs is about 13.8 ± 1.1 nm. The size of AuNPs was analyzed by TEM, of which the average diameter is about 14.6 ± 2.4 nm as shown in Fig. 4-3. The DLS results also showed that the synthesized AuNPs was monodisperse as shown in Fig. 4-2B.
The extinction coefficient of different size of AuNPs could be applied with eq (3) (Liu et al., 2007):
99
ln 𝜀 = 𝑘𝑙𝑛𝐷 + 𝑎 (3) Where 𝜀 is extinction coefficient in M-1cm-1, D is the core diameter of the NPs, and 𝑘 = 3.32, 𝑎 = 10.8. The 𝜀 is 3.94 108 M-1cm-1 for 15 nm AuNPs in diameter. The concentration of 15 nm AuNPs could determine by Bear’s law (A = εbC) and obtained molar concentration about 5 nM. Furthermore, the number density of the particles (N, NPs/mL) can be determined by eq (6) [Haiss et al., 2007]:
N = 𝐴450×1014
𝑑2[−0.295+1.36𝑒𝑥𝑝(−(𝑑−96.878.2 )2)] (6) The 15 nm AuNPs synthesized in this study has the N = 1. ± 0.2 1011 NPs/mL.
5-1-2. Stability of citrate-capped AuNPs and improvement
The AuNPs owns LSPR spectrum strongly depending on the size of NPs; therefore, the changes of size of AuNPs due to aggregation could be observed in the spectrum showing an absorption red shift [Ghosh and Pal, 2007]. The aggregation level estimated by the absorption spectra of AuNPs which the change of decreasing in A525 and increasing in A625 [Chuang et al., 2010]. Citrate-capped AuNPs are considered less stable of which surrounded by an EDL due to adsorbed citrate and chloride anions [Stankus et al., 2011]. According to DLVO theory, the stability of AuNPs is comprised by two forces: repulsive electrostatic forces and attractive van der Waals forces. As the ion strength is increased in the medium, the thickness of EDL
decreases due to screening of the surface charge. This causes the decrease in VR, and the increases the susceptibility of the dispersed particles to form aggregates [Wu et al., 2011]. The citrate-capped AuNPs conducted with salt stress assays in this study and estimated the
stability under different ion strength. The salt stress assays were performed with sodium chloride as source of salt. As the results in Fig. 4-4A, the adsorption spectra of citrate-capped AuNPs changed due to the addition of salt. The aggregation parameter (A625 /A525) indicates that the salt concentration above 10 mM NaCl caused aggregation, and the A625 /A525 value
100
above 0.3 is considered unacceptable colloid suspension (Fig. 4-4B). The result is agreed with Jans & Huo that citrate-capped AuNPs tend to aggregate at a salt concentration higher than 10 mM NaCl due to the disruption of EDL [Jans et al., 2012].
Moreover, the pH effect to AuNPs also is considered in this study. The synthesized citrate-capped AuNPs is pH 5.6 and the pH was adjusted to different pH by HCl or NaOH properly. As shown in Fig. 4-6A, the adoption spectra of different pH AuNPs showed no red shift and the aggregation parameter (A625 /A525) also proved no significant difference (Fig.
4-6B). On the whole, it is clearly that ionic strength plays vital role in inducing AuNPs aggregation but pH has no effect.
To stabilize citrate-capped AuNPs, ligands or capping agents should be applied. The citrate-capped AuNPs can be functionalized by thiol ligands easily that pseudo-covalently bind (~45 kcal/mol) to the AuNPs surface [Bastús et al., 2011]: peptides [Harkness et al., 2012], proteins, DNA and carbohydrate moieties [Housni et al., 2008]. AuNPs stabilized with an inert macromolecule such as BSA, gelatin, or PEG could avoid undesired or non-specific labeling to other components in the biological system [Thobhani et al., 2010]. Therefore, the stabilizers investigated in this study are PEG and BSA. Fig. 4-5A shows that it is need above 2% PEG to stabilize AuNPs under 25 mM NaCl stress and 5% PEG could further tolerance 50 mM NaCl without serious aggregation. PEG provides steric barrier to stabilize AuNPs with increasing ionic strength; however, the PEG solution could be high viscosity with high
content of PEG. Meanwhile, the BSA via salt-bridge to conjugate on AuNPs surface, and thus could cause the steric barrier to protect AuNPs from getting to close with neighboring AuNPs to interact and aggregate [Brewer et al., 2005]. Therefore, as low as 0.1% BSA could stabilize AuNPs under physical condition (150 mM NaCl) as given in Fig. 4-5B.
101
5-1-3. Effect of peptide substrate charges on AuNPs probes
The importance of peptide sequence design as ligands for conjugating on AuNPs is known [Fanun, 2010], but not suitable for protease activity detecting probe. For different target protease, the peptide sequence must limit to possible hydrolysis substrates. The reasonable thought is to consider overall effect instead of particular amino acid residue. It is proposed that peptide-modified AuNPs aggregation can be controlled by the electrostatic charge of peptides [Tullman et al., 2007]. Therefore, the charges of peptide substrates effect are investigated in this thesis. The pI of peptides determines the charges at function of pH (Table 3-1). UV-Vis spectra studies on A/1466p-FITC probe were given in Fig. 4-7. Fig.
4-7A represents the positively charged modification process and it is clear that positive charges caused aggregation but tunable upon to the environment of strong negatively charged (buffer of pH 11.0). Fig. 4-7B shows the neutral charged modification process and it indicates that postnatal environment alter the stability of 5.6A/1466p-FITC probe. In addition the buffer of pH 4.0 caused the adsorption spectra red shift obviously. Fig. 4-7C indicates the negatively charged modification process and it shows more tolerable to postnatal environment change than neutral charged modification probe. The UV-vis spectra show that the AuNPs probe is stable in neutral and negatively charged environments and the spectra also show little red shift in positively charged environment. Fig. 4-7D also represents the negatively charged
modification process; however, the spectra of pH 11.0 modification before removal process gained not smooth curve, which might cause by tense ionic strength to AuNP during adjusting pH previously. Our results show corresponding to the results reported by Tullman et al. [2007];
however, they indicated that the disruption of EDL due to positively charges induced irreversible aggregation which is reversible in this study. The possible reason is that 1466p-FITC is -2 charged at pH 11.0 (Table 3-1) providing enough repulsive force to
increase the EDL and alter aggregation to dispersion in this study. This UV-vis study provides
102
an important rule that during modification the peptide ligands should present in neutral or negatively charges.
Is the influence of peptide charges during conjugation could continuously affect the stability of AuNPs probes in postnatal environment after purification? It is known that neutral or negatively charges of peptide during modification lead to dispersion when positively charges peptide leads to aggregation. Hence, the results in Fig. 4-8A shows for A/1466p-FITC probe that pH 4.0 modification caused serious aggregation. Besides, Fig. 4-9A shows for A/1477p-FITC probe that pH 4.0 and 5.6 modification caused serious aggregation. But after purification, the postnatal environment is pH 5.6 cause effect on pH 7.4 modification of A/1477p-FITC probe as shown in Fig. 4-9B. The results mentioned above indicate that only neutral or negatively charges peptides conjugate to AuNPs would lead to stable AuNPs probe and conquer resulting positively charges postnatal environment.
5-1-4. Conjugation of peptide substrates to AuNPs
How many peptide substrates per AuNPs are there after conjugation? During the modification, the peptides substrates were existed either on the AuNPs or in the discarded supernatant. Because of the fluorophore modified with peptide substrates, the amount of peptide substrates in each possible part could estimate by fluorescence intensity (Table 4-1).
Although 1482p-FITC shows significant low conjugation rate in supernatant supplement method, the DTT direct method shows that all of the peptide-FITC own about 50 ~ 60 % conjugation rate. Table 4-2 shows that peptide substrates in supernatant and DTT discard could not match the initial loading amount, especially 1466p-FITC and 1477p-FITC. The disagreement indicates the loss of peptide substrates during purification and less strong pseudo covalent to AuNPs. Therefore, the DTT discard direct method was chosen to estimate the conjugation ratio.
103
Lévy [2004] reported that 12 nm AuNPs - CALNN conjugated ratio is 791
peptides/AuNPs or 1.67 peptides/nm2; for another method they gained 919 peptides/AuNPs or 1.93 peptides/nm2. Olmedo et al. [2008] reported that 12.5 nm AuNPs - CLPFFD, CLPDFF, and CDLPFF conjugated ratios are 460, 420, and 203 peptides/AuNPs, respectively. Guerrero et al. [2012] reported that 12 nm AuNPs-CK conjugated ratio is 122 peptides/AuNPs and 12 nm AuNPs-CLPFFD is 75 peptides/AuNPs [Guerrero et al., 2012]. Our results show that 15 nm AuNPs to 1466p-FITC, 1477p-FITC and 1482p-FITC conjugated ratios are 1789.2, 1494.2, 1065.5 peptides/AuNPs or 5.4 4.5 3.2 peptide/nm2, respectively (Table 4-3).
Compared with others work, our results show very high conjugation ratio of peptide on the AuNP surface; in other words, the synthesized probes is compact packed monolayers by peptides. The advantages of dense packed monolayers have two: (1) more loading peptides mean higher chances for proteases to identify and activate the AuNPs probes, (2) compact coverage usually make AuNPs probes more stable compared with citrate-capped AuNPs.
The zeta potential indicates the degree of repulsion between adjacent, similarly charged particles in dispersion. Colloids with high zeta potential (negative or positive) are electrically stabilized while colloids with low zeta potentials tend to coagulate or flocculate. A value of 25 mV (positive or negative) can be taken as the arbitrary value that separates low-charged surfaces from highly-charged surfaces [Greenwood and Kendall, 1999]. Zeta potential of citrate-capped AuNPs and AuNPs probes were analyzed (Table 4-4). The dense package of AuNPs probes show improvement in stability which have larger negative zeta potentials (-39
~ -52 mV) compared with citrate-capped AuNPs (-34.7 mV). The results indicate higher conjugation rate of peptide substrates provide steric stabilization effect. However, the values of zeta potential are neither coordinated to pI of peptide substrates nor conjugation rates of which the lower pI expected to have larger negative zeta potential and the denser the conjugated peptide substrates forming relatively stable particles. AuNPs process with
104
purification could make extraordinary difference that 7.4A/1466p-FITC and
10.0A/1477p-FITC probes show smaller negative (-15 ~ -17 mV). The 5.6A/1482p-FITC probe remain approximately the same level of zeta potential (-39.2 to -34.3 mV). The results could distinguish 5.6A/1482p-FITC probe of three AuNPs probes that has the best stability;
although the 5.6A/1482p-FITC probe has lowest conjugation rate, the peptide charges seem to provide strong exclusive force leading to dispersion stable.
Besides, the size of citrate-capped AuNPs also determinate the fate of modification, which is the size larger than 25 nm AuNP is hard to self-assembly with peptide substrates without aggregation even processed under the principle mentioned above.
5-1-5. Optimize the sensitivity of AuNPs probes to proteases
Proteinase K is a very high specific activity protease; therefore, proteinase K was used as tools to optimize the experiment conditions. Although the synthesized probes form dense packed monolayers, the stability was still concerned. The purified 10.0A/1477p-FITC probes are poor to tolerance ionic strength change unless process with stabilization. The possible reasons are the pI of 1477p-FITC is high leading to positively charges which easily to induce aggregation and the purification leads to ionic environment changes of which citrate solution to nonionic solution. For improving stability of AuNPs probes, the stabilizers such as PEG and BSA are used; however, do these stabilizers cause negative effect to the sensitivity of AuNPs probe to proteinase? The purification process consist of two times centrifugation, thus there would be the need of wash buffer and suspended buffer. The wash buffer determined to use PEG solution; hence the shorty immersed in PEG solution affection is confirmed in Fig.
4-10A. The concentration of PEG has no effect to proteinase K activity, except 5% PEG is significant lower compared with 0% PEG. It points out that shorty immersion would not cause PEG adhesion on AuNPs probes, but high content of PEG still rise the possibility of PEG remain. The suspension buffer is BSA solution as efficient stabilizer to NPs and BSA
105
hydrodynamic diameter is about 7 nm [Yohannes et al, 2010]. It is clear that the increase of BSA concentrations was with the proteinase K activities decrease (Fig. 4-10B). The results might due to the BSA is a substrates of proteinase K and high content of BSA increase the steric barrier of proteinase K to hydrolysis peptide substrates on AuNPs. Consequently, the 2% PEG solution is for purification and 0.1%BSA solution is for suspension and stabilization for processed AuNPs.
The optimal pH of proteinase K sensitivity to 7.4A/1466p-FITC is pH 9.0 (Fig. 4-11).
The quenching effect of AuNPs to fluorophore (D) is distance dependence; therefore, the concentration of AuNPs probe is important. To identify the optimal concentration of AuNPs probes, two AuNPs were applied and with various concentrations of proteinase K. The 7.4A/1466p-FITC probe detecting proteinase K (100 ng/mL, for 1 hr at 37°C) shows the optimal concentrations are 0.63 ~ 1.25 nM (Fig. 4-12A). Besides, the 10.0A/1477p-FITC probe detecting proteinase K (400 ng/mL, for 1 hr at 37°C) owns the optimal concentrations is 1.25 nM (Fig. 4-12B). Both of the results show that higher or lower concentration lead to lower proteinase K sensitivity responded on fluorescence intensity change; which corresponds to the hypothesis of higher concentration AuNPs probes may lead to negative impact on fluorescence emission by quenching. The lower concentration AuNPs probes has the same phenomenon that could explain by insufficient peptide substrates for proteinase K to perform.
5-1-6. Establishment of proteinase K activity assay by AuNPs probes
There are three AuNPs probes conduct in this thesis, the differences between three of them are the length and charges of peptide substrates (Table 3-1). The first investigated probe is 7.4A/1466p-FITC probes. The linear correlation ranged from 10 to 400 ng/mL proteinase K for 1 hr detection time is confident, of which y = 8.32 x + 433.82 and R² = 0.96 (Fig.
4-13A) ;and the time course of proteinase K activity also given in Fig. 4-13B. Yang et al.
[2011] provided that middle region of peptides contains four alanine (AAAA) could promote
106
peptide assembly into densely packed monolayer on AuNPs. The repeating amino acid reissue also applied by Kim et al. [2008], and that they used four glycine (GGGG) as linker to
cysteine (C). Wang et al. [2010] also used three glycine (GGG) as linker in design. The 10.0A/1477p-FITC probe is designed by the combination idea of mentioned advantages of repeating amino acid reissue and glycine is chosen instead of alanine for the simplest structure which can fit into hydrophilic or hydrophobic environments. The purpose of five glycine (GGGGG) is to play the role of linker and to liberate the steric barrier formed by dense packed monolayer of peptide substrates on AuNPs. It is expected to increase the sensitivity of protease to the 10.0A/1477p-FITC probe. As shown in Fig. 4-14A, both AuNPs probes have great linear correlation between fluorescence intensity change and proteinase K concentration.
The 7.4A/1466p-FITC probe has the linear correlation ranged from 25 to 400 ng/mL proteinase K for 15 min detection time, of which y = 3.14 x + 6.82 and R² = 0.99; while 10.0A/1477p-FITC has the range from 10 to 400 ng/mL proteinase K for 15 min detection time, of which y = 10.56 x + 396.76 and R² = 0.99. Besides, comparison in time course of low concentration proteinase K (25 ng/mL) activity between two AuNPs probes also was given in Fig. 4-14B. The results clearly indicate that proteinase K is more sensitive to
10.0A/1477p-FITC responding on the fluorescence intensity change increases above three folds under the same conditions.
So far, we conclude that increasing length of peptide substrates with glycine is efficient to increase sensitivity by decrease steric barrier. However, the 10.0A/1477p-FITC probe is not easy to synthesize due to its’ high pI value and less stable for easily positively charged. For the purpose of conducting the principle of design peptide substrate for various proteases, the challenge of high pI of peptide substrate is likely to face with. Therefore, the linker of glycine (GGGGG) is replaced by aspartic acid (DDDDD). The function of aspartic acid only
discusses in the aspect of locating at far end from AuNPs could help forming stable packed
107
monolayer [Olmedo et al., 2008]. In this study, the five D residues provide a simple chain like five G but with strong negatively charges. Therefore, the chain of aspartic acid (DDDDD) as the linker not only decreases the steric barrier but also decreases the pI of peptide substrates, and that is 1482p-FITC. The 5.6A/1482p-FITC is more stable due to negatively charge expected increasing the EDL of AuNPs probe. The conjugation ratio also shows lower than 10.0A/1477p-FITC, the explanation is corresponded with Olmedo et al. [2008]. It indicated that D inducing the exclusion of more molecules of absorbed citrate (due to the repulsive interaction between D residues and citrate carboxylates). Because the stability of
5.6A/1482p-FITC probe is improved this probe could suspend without stabilizer and stand the experiment salt stress (data not shown). Hence, the sensitivity of 5.6A/1482p-FITC probe is expected to increase without the BSA interference. The time course of proteinase K activity is given in Fig. 4-15A. The 5.6A/1482p-FITC shows the linear correlation ranged from 0.1 to 12.5 ng/mL proteinase K for 15 min detection time was confident, of which y = 570.36 x + 209.17 and R² = 0.99 (Fig. 4-15B). The detection limit is sharply down to pg/mL level compared with 10.0A/1477p-FITC only for ng/mL level in shorter detection time (15min).
5-2. AuNP and the morphology change analysis 5-2-1. Gel electrophoresis analysis
The morphology change of AuNPs after modification and 5.6A/1482p-FITC probe activated by proteinases could be observed in gel electrophoresis. The migration differences of citrate-capped AuNPs, 5.6A/1482p-FITC and the AuNPs probe activated by chymotrypsin and proteinase K are clearly shown in Fig. 4-16. With the hydrolysis of peptide substrates, the molecular weight of AuNPs probe decreases and shows higher mobility. The UV-light could excite the FITC fluorophore; therefore, the quench phenomenon was confirmed in Fig. 4-17B.
108
The UV-light excited image of 5.6A/1482p-FITC without activated by proteinase still
presented with very weak fluorescence band, which the band were considered the unbounded peptide substrates. Both free and bound peptide substrate were cleaved by proteinase were given. Besides, the bands of hydrolyzed peptide substrates has lower mobility could explained by the loss of five Asp (D) which have very strong negatively charges.
5-3. Establishment of chymotrypsin activated fluorescent self-assembly AuNPs probes
5-3-1. Establishment of chymotrypsin activity assay by AuNPs probes
The AuNPs probes conducted in further also could apply in detecting chymotrypsin activity. The optimal pH of chymotrypsin sensitivity to 10.0A/1477p-FITC is pH 8.0 , as shown in Fig. 4-18 [Norris et al., 1970; Wilcox, 1970]. Although three probes have same cleavage sites to chymotrypsin, Fig. 4-19A still indicates that 7.4A/1466p-FITC probe has very low sensitivity to chymotrypsin responding on the low fluorescence intensity change. In the meantime, the 10.0A/1477p-FITC probe acquires confident linear correlation ranged from 25 to 500 ng/mL chymotrypsin for 1 hr detection time, of which it y = 6.75 x + 87.34 and R² = 0.96. Fig. 4-19B gives the results of corresponding to the concentration results; that the
10.0A/1477p-FITC probe shows time correlated to delta fluorescence intensity, while 7.4A/1466p-FITC probe shows very low increase instead. Again, increasing the length of peptide substrates shows improvement of chymotrypsin to activate the AuNPs probe. The 10.0A/1477p-FITC probe could reduce the detection time from 60 min to 30 min with a perfect linear correlation range from 25 to 300 ng/mL chymotrypsin, of which y = 5.07 x - 105.87, R² = 1 (Fig. 4-20). The specificity of 10.0A/1477p-FITC probe also investigated, and that serine protease like trypsin was applied. Trypsin has one cleavage site in 1477p-FITC.
Although the results showed that the linear correlation ranged from 200 to 600 ng/mL trypsin
109
is confident, of which y = 0.62 x - 76.48 and R² = 0.99, but relatively low about 10 fold fluorescence intensity change compared with chymotrypsin (Fig. 4-21).
To lower the detection limits and in short detection time, the 5.6A/1482p-FITC applied.
To lower the detection limits and in short detection time, the 5.6A/1482p-FITC applied.