Use of Gold Nanoparticles as Additives in Protein Crystallization

Article pubs.acs.org/crystal

Use of Gold Nanoparticles as Additives in Protein Crystallization Diana Ribeiro,† Alina Kulakova,† Pedro Quaresma,‡ Eulália Pereira,‡ Cecília Bonifácio,† Maria Joaõ Romaõ ,† Ricardo Franco,†,* and Ana Luísa Carvalho†,* †

REQUIMTE/CQFB, Departamento de Química, Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa, 2829-516 Caparica, Portugal ‡ REQUIMTE, Departamento de Química e Bioquímica, Faculdade de Ciências da Universidade do Porto, 4169-007 Porto, Portugal ABSTRACT: Gold nanoparticles (AuNPs) exhibit unique properties that have made them a very attractive material for application in biological assays. Given the potentially interesting interactions between AuNPs and biological macromolecules, we investigated AuNPs-induced protein crystal growth. Differently functionalized AuNPs were tested as additives in cocrystallization studies with model proteins (hen egg white lysozyme (HEWL), ribonuclease A (RNase A), and proteinase K) as well as with case studies where there were problems in obtaining well-diffracting crystals. Trials were performed considering different crystallization drawbacks, from total absence of crystals to improvement of crystal morphology, size, twinning, and number of crystals per drop. Improvement of some of these factors was observed in the cases of HEWL, RNase A, phenylalanine hydroxylase (PAH), myoglobin, native aldehyde oxidase (AOH), and human albumin. In these proteins, the presence of the AuNPs promoted an increase in the size and/or better crystal morphology. From the systematic trials and subsequent observations, it can be concluded that the introduction of AuNPs should definitely be considered in crystal optimization trials to improve previously determined crystallization conditions.

■

INTRODUCTION X-ray crystallography is the foremost method to acquire atomic resolution for protein structures, and the limiting step is still the production of protein crystals suitable for structure solution. Therefore, strategies that facilitate the production of wellordered crystals for X-ray diffraction techniques, such as nucleating agents and additives, are highly sought for. Gold occupies a unique position among the elements of the periodic table, and its chemical stability, the useful surface chemistry of the materials it generates, and its distinctive optical properties, have made gold an extremely attractive metal to employ in a variety of technologies.1−3 This is especially true for nanotechnology as a consequence of these unique properties, as it is easier to work with gold at the nanoscale than with any other metal.2,3 Gold nanoparticles (AuNPs) present several advantages, such as the control over size and morphology at the nanometerscale, as well as the ability of functionalization with bioactive materials.4 AuNPs have excellent biocompatibility and display unique structural, electronic, magnetic, optical, and catalytic properties, which have made them a very attractive material for the development of bionanosystems,5 based on coupling AuNPs with biological macromolecules. Given its wide applications in biomaterials and interesting interactions, AuNPs could be potential agents for use in protein crystallization experiments. In 2008, Hodzhaoglu et al. have described AuNPs as effective nucleants for lysozyme (HEWL) crystallization.6 Their study © 2013 American Chemical Society

revealed an increase of the nucleation number of HEWL in the presence of citrate-capped AuNPs and of AuNPs functionalized with alkanethiols presenting a COOH terminal group. Their results suggested that AuNPs induced the crystallization of HEWL and also of ferritin.6 These observations were further explored in the present study, in order to evaluate a possible application to other proteins less prone to crystallization, or to yield well-diffracting crystals.

■

METHODS

AuNPs Synthesis and Functionalization. Spherical AuNPs were synthesized by the Turkevich method with minor modifications,7,8 in which HAuCl4 is chemically reduced by citrate (Table 1). A solution containing 62.5 mL of Milli-Q water and 43 μL of a 30 wt % gold salt solution was heated until boiling under reflux, using a sand-bath, with continuous stirring. At this point 6.25 mL of 36.8 mM sodium citrate solution was quickly added. The solution immediately changed color from the gold complex characteristic yellow color to colorless, then to black, and finally to red. Heating and stirring was continued for another 15 min, and the nanoparticle colloid was cooled down to room temperature. The method proposed by Haiss et al.9 was used to estimate the concentration of AuNPs. The AuNP solution was stored at 4 °C until further use. The average size of citrate-capped AuNPs prepared following the same procedure and under the same conditions Received: September 23, 2013 Revised: November 7, 2013 Published: November 13, 2013 222

dx.doi.org/10.1021/cg4014398 | Cryst. Growth Des. 2014, 14, 222−227

Crystal Growth & Design

Article

Table 1. Schematic Synthesis, Functionalization, and Structure of the Different Capping Agents

as the AuNPs used in this work was 13.9 ± 3.3 nm, as determined by transmission electron microscopy (TEM) (data not shown). These citrate-stabilized AuNPs solutions were then functionalized with four different capping agents: mercaptoundecanoic acid (MUA), pentapeptide CALNN (CASLO Laboratory, Denmark), pentapeptide CALKK (CASLO Laboratory, Denmark), and the thiolated PEG (SPEG) HSC6EG3 (Prochimia Surfaces, Poland). These ligands, presenting thiol groups, have great affinity for the gold surface, easily

replacing the citrate molecules physiosorbed at the nanoparticles surface. The functionalization with MUA and CALNN was carried out by incubating the AuNPs solution with the ligand, for at least 30 min at 4 °C, at an AuNP to ligand molar ratio of 1:5000 using 10 mM MUA stock solution in ethanol; and at a 1:1000 molar ratio from 5 mM CALNN stock solution. Functionalization with SPEG and CALKK was performed using a procedure in which the nanoparticles were first functionalized with SPEG by adding 100 μL of a 10 mM 223



Article

solution of SPEG in ethanol to 10 mL of AuNP colloid. The mixture was incubated for 16 h at room temperature, followed by centrifugation (13400 rpm, 15 min) to remove excess SPEG in the supernatant. For the CALKK functionalized NPs, a batch of AuNPs functionalized with SPEG was used as a starting point. Such procedure intended to avoid aggregation of the AuNPs as observed when CALKK was directly added to the citrate capped NPs. To 10 mL of SPEG-AuNPs, 100 μL of a 10 mM solution of CALKK was added followed by incubation at room temperature for 16 h in order to allow a gradual exchange of the thiol ligands. The colloid was finally centrifuged, andAuNPs were redispersed in an equal volume of Milli-Q water. Model Proteins and “Real” Cases. HEWL, bovine pancreas RNase A, and proteinase K from Tritirachium album, with wellestablished crystallization conditions, were elected as model proteins in order to evaluate changes in the crystals’ morphology induced by the presence of AuNPs as additives in crystallization trials. These test proteins were purchased from Sigma-Aldrich. The remaining proteins were provided by co-workers, as part of their research projects, some with undisclosed crystallization conditions. Crystallization conditions were known in the cases of horse heart myoglobin, human albumin, human transferrin, human phenylalanine hydroxylase (PAH),10 and human aldehyde oxidase (AOH).11 Nevertheless, the crystals obtained were too small and/or with unsuitable morphologies for diffraction experiments. Assays with AuNPs were performed to investigate if optimized crystals could be grown. Crystallization conditions were unknown in the cases of: ypsilon schachtel (YPS);12 inducible NOS (iNOS), which is an homologue of nitric oxide sinthase (NOS);13 the enzyme AraN;14 and human apomalectin, and its truncated form (malectin-trunc).15,16 For malectin, crystallization conditions have been determined for the disaccharideliganded structure; however, crystallization of the apo structure was necessary for structural characterization. Crystallization Trials Using AuNPs as Additives. All the crystallization experiments were performed according to the hanging drop method, using 24-well crystallization boxes, at 20 °C. The crystallization boxes where kept at 20 °C. Observation of the drops was made through an optical microscope (Olympus SZH10 Research Stereo microscope, with an Olympus DF Plan Apo IX objective). These experiments were carried out with AuNP-MUA, AuNPCALNN, AuNP-CALKK, and AuNP-SPEG at a concentration of 3 nM. Solutions of each AuNP and protein were incubated together, overnight at 4 °C. For each condition tested, control drops of protein were always performed. Milli-Q water was used in all aqueous solutions. For each assay, fresh enzymes and proteins were dissolved in the appropriate buffer solution. Drop setups were prepared for each protein accordingly: 15 and 25 mg/mL of HEWL (Sigma-Aldrich) in 50 mM CH3COONa, pH 4.5, using 700 μL of 0.5 M NaCl as precipitant; 50 mg/mL of RNase A (Sigma-Aldrich) in 50 mM CH3COONa, pH 5.5, using 1 mL of a mixture of 3 M NaCl and 1.2 M (NH4)2SO4 as precipitant agent; 20 mg/mL proteinase K in 25 mM Tris-HCl, pH 6.0, using 700 μL of a mixture of 1.2 M (NH4)2SO4 and 0.1 M Na2CO3; 20 and 25 mg/mL of myoglobin (Sigma-Aldrich) in 20 mM Tris-HCl buffer, pH 7.5, using 700 μL of a mixture of 3 M (NH4)2SO4 and 15 mM EDTA as precipitant; 100 mg/mL albumin in 0.1 M HEPES, pH 7.0, using 500 μL of 0.4 M (NH4)2SO4 and 0.11 M CdSO4 as precipitant; 30 mg/mL transferrin in 20 mM de Tris-HCl, pH 8.0, using 700 μL of 20 mM Na2CO3 and 200 mM NaCl. Due to confidentiality reasons, crystallization conditions for PAH and AOH (native and mutant) are kept undisclosed. Drop setups of the incubated AuNPs and protein solutions were prepared using a 1:1 ratio of sample/precipitant, to a total volume of 4 μL in the cases of HEWL, RNase A, and myoglobin, 2 μL for albumin, 6 μL for proteinase K, and 12 μL in the case of transferrin. PAH with AuNPs drops were made using a 2:1 ratio of sample/precipitant, to a total volume of 3 μL and also of 6 μL.

Since no preliminary crystallization conditions had been found in the case of apo-malectin, iNOS, YPS, and AraN, a screening of crystallization conditions was performed in these cases, using an inhouse prepared sparse matrix screen17 of 80 conditions. X-ray Diffraction and Data Processing. Diffraction data from all suitable crystals were collected, in different occasions, at 100 K in the European Synchrotron Radiation Facility, in Grenoble, France. Crystals were cryo-protected by the introduction of glycerol in the harvesting conditions prior to flash-freezing in liquid nitrogen. Data were processed with Mosflm,18 and selected parameters are summarized in Figure 2 for the cases of HEWL+AuNP-CALNN, RNase A+AuNP-SPEG, and albumin+AuNP-CALKK.

■

RESULTS AND DISCUSSION AuNPs as Additives in the Optimization of Protein Crystallization Conditions. The effect of the AuNPs in the growth of protein crystals was studied by the differences observed in comparison with control drops in the absence of AuNPs. The more relevant results are presented in Table 2. Table 2. Summary of the Main Results Obtained from the Crystallization Trials with AuNPs Functionalized with Different Ligandsa proteins/ AuNPs

AuNPMUA

AuNPCALNN

AuNPCALKK

AuNPSPEG

HEWL RNase A proteinase K

+ − 0

+ 0 0

++ 0 0

++ + 0

myoglobin malectin malectin-trunc PAH AOH AOH-ΔHis AOH-Y885M AOH-Y885F albumin transferrin

++ 0 0 + + 0 0 0 + 0

++ 0 0 ++ 0 − 0 0 + 0

+ 0 0 + 0 − + 0 ++ 0

+ 0 0 0 0 − + 0 + 0

a Symbols represent the effect on crystals grown in the presence of AuNPs comparative to the control drops: 0 indicates no observable effect (crystals are similar to the control, in size, shape, multiplicity, and number of crystals per drop); − indicates negative effect (worse morphology or absence of crystals); and + indicates positive effect (either bigger crystals, less multiplicity, or less crowded drops).

Positive results corresponded to the cases of improvement of the crystals’ size and morphology in comparison to the control drop (symbolized by “+” or “++”). Negative results corresponded to the absence of crystals or crystals with worse morphology in the presence of the AuNPs (symbolized by “−“). When the observations in the drops with AuNPs were the same as in the control drops, the effect was considered to be null (symbolized by a “0”). For several of the proteins in this study, improvement of crystal size and morphology was observed in the presence of the AuNPs. More specifically, this improvement was evident in HEWL, RNase A, myoglobin, PAH, AOH, and albumin. From the three model-proteins, HEWL, RNase A, and proteinase K, only the latter failed to show any changes in its crystal growth in the presence of AuNPs. After overnight incubation with AuNP-CALKK and AuNP-SPEG, respectively, HEWL crystals grew in two days after drop setup, while, in the presence of AuNP-MUA and AuNP-CALNN, HEWL yielded 224



Article

Figure 1. Photographs of the most relevant results obtained in the presence of AuNPs functionalized with different ligands. Crystallization conditions for each protein are as follows: (a) 25 mg/mL in 50 mM CH3COONa, pH 4.5 + 5% NaCl; (b) 50 mg/mL in 50 mM CH3COONa, pH 5.5 + 3 M NaCl + 1.2 M (NH4)2SO4; (c) undisclosed conditions; (d) undisclosed conditions; (e) 20 mg/mL in 20 mM Tris-HCl, pH 7.5 + 3 M (NH4)2SO4 + 15 mM EDTA; (f) 100 mg/mL in 0.1 M HEPES, pH 7.0 + 0.4 M (NH4)2SO4 + 0.11 M CdSO4. All proteins were incubated in the presence of 3 nM of the respective AuNPs solutions. NA: not available.

of RNase A) seemed to have a similar positive effect in inducing the growth of HEWL and RNase A single crystals in a low number of crystals per drop. Furthermore, as shown in Figure 2, the presence of the AuNPs did not disrupt HEWL and RNase A usual crystal lattices, and the diffraction patterns and cell parameters of the crystals where comparable to those of native crystals. The potential of the AuNPs-induced crystal growth was then explored by applying the method to other proteins that had proved difficult to crystallize or from which no well-diffracting crystals could be obtained. In total, six distinct proteins were tested, and relevant results were obtained for PAH, myoglobin, native AOH, its mutant AOH-Y885 M, and albumin.

crystals after four days. This is a similar observation to the control drops, with the exception that, in the control drop a higher number of smaller crystals were detected (Figure 1a). RNase A crystals in the control drops grew within three days, as observed for a crystal grown in the presence of the AuNPs. For the drops of RNase A with AuNP-CALNN and AuNP-CALKK, the number and size of the crystals obtained were similar to those in the control drop. Nevertheless, in the presence of AuNP-SPEG, two single crystals were formed (Figure 1b). At this time, no crystals were detected in the drops prepared with AuNP-MUA, although one single crystal appeared after almost one month of regular inspection of the setups. AuNP-SPEG and AuNP-MUA (although with a significant delay in the case 225



Article

Figure 2. Example X-ray diffraction patterns and experimental details for the diffraction of selected crystals grown after incubation of the respective proteins with 3 nM of the different AuNPs.

PAH crystals grown in the presence of AuNP-CALNN and AuNP-CALKK showed a significant improvement in crystal size due to fewer crystals per drop, compared with the control drops (Figure 1c). This notorious improvement in crystal morphology (larger and with sharper edges) did not reflect, however, in improvement of diffraction of X-rays (4 Å maximum resolution). Although no cell parameters could be accurately determined, the crystal’s frailty suggests a very high solvent content. AOH crystals in the presence of AuNP-MUA grew to larger sizes and with a significantly lower number of crystals per drop, as compared to the control ones (Figure 1d). AOH-Y885 M crystals obtained in the presence of AuNP-CALKK and AuNP-SPEG grew slightly larger than the control ones, although still exhibiting the same needle morphology. Myoglobin in the presence of AuNPs with any of the functionalization molecules tested yielded crystals before the appearance of any crystals in the control setups. These early crystals obtained in the presence of AuNPs where fewer in number and presented larger sizes than their control ones. Nevertheless, it was still not possible to obtain single isolated crystals (Figure 1e). For albumin it was also observed that crystals grown in the presence of AuNPs were larger and fewer with AuNP-CALKK (Figure 1f). Since these are difficult cases, the fact that consistently larger and fewer crystals per drop were obtained is by itself a good result. Nevertheless, in each case, a strategy for the optimization of these conditions must be further designed. In the case of PAH crystals, we believe that controlled dehydration experiments may help to produce better ordered crystals, which hopefully will generate a useful diffraction pattern. Due to their significantly small dimensions, AOH crystals will have to be tested in a microfocus beamline and maybe no further optimization is needed. Nevertheless, it is worth trying

macro- or microseeding experiments. In this case, protein availability is an issue. Microseeding experiments should also be considered to obtain thicker needles of myoglobin crystals, hopefully overcoming also the problem of highly aggregated crystals. AuNPs as Promoters of Nucleation. The propensity to promote nucleation by AuNPs was evaluated in the case of malectin, iNOS, YPS, and AraN, all with yet-to-determine crystallization conditions. In these cases, after incubation with AuNPs, a screening for preliminary conditions was performed. Unfortunately, for these proteins all crystallization trials in the presence of the tested AuNPs failed, also in the case of control setups. Due to its particulate nature, it was expected that, at least under some conditions, crystallization could be induced, although these proteins are not straightforward cases in the quest for crystallization conditions. It should be mentioned that, at the time of manuscript preparation, several crystallization conditions were found for apo-malectin, which have in common the presence of organic solvents as precipitants (30% ethanol/isopropanol/methylpentanediol). The conditions were found after performing a microseed matrix screening using seeds of liganded-malectin crystals. Considering that AuNPs could have a nucleant effect and act in a similar way to crystal seeds, the reason for not generating crystals under similar crystallization conditions may be related to the high content of organic solvents required, which may be incompatible with the use of AuNPs. X-ray Diffraction Analysis. X-ray diffraction was tested for all cases where well-shaped crystals could be obtained, and three were selected for presentation in Figure 2. PAH crystals also diffracted, but due to confidentiality reasons, diffraction details are not included. Besides PAH, morphologically acceptable single crystals could be obtained, in a reproducible 226



■

way, for HEWL incubated with all AuNPs (diffracting to a maximum resolution of 1.2 Å), for RNase A incubated with AuNP-MUA and AuNP-SPEG (up to 1.4 Å), for PAH incubated with AuNP-CALNN (up to 4 Å), and for albumin incubated with AuNP-CALKK (up to 2.4 Å). A comparison was possible with the diffraction behavior of control crystals, and in all cases, it was observed that the tested AuNPs caused no significant changes in unit cell parameters, space group, or resolution of the data. Further analysis is underway to evaluate the quality of electron density maps, as well as differences in relation to control crystals.

■

Article

AUTHOR INFORMATION

Corresponding Author

*Phone: (351) 212948300. Fax: (351) 212948550. E-mail: [email protected]; [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors would like to acknowledge Abhik Mukhopadhyay for assistance in data collection and processing of lysozyme crystals, Teresa Santos-Silva, Angelina Palma, Benedita Pinheiro, Márcia Correia, and Catarina Coelho for kindly providing the model proteins used as “real” cases in this study, and Marino Santos for data processing from albumin crystals and assistance in preparation of figures. We also acknowledge assistance in synchrotron data collection at ESRF (Grenoble, France) and SLS (Villigen, Switzerland). The work was supported by the Portuguese Science and Technology Foundation (FCT-MEC) and COMPETE through Grants PTDC/CTM-NAN/112241/2009 to R.F. and Grants PEst-C/ EQB/LA0006/2011 and PEst-C/EQB/LA0006/2013 (to Associate Lab REQUIMTE).

CONCLUSIONS

HEWL, RNase A, and proteinase K are frequently used as model proteins for their trustworthy crystallization and diffraction properties. Therefore, the probability of significantly improving the crystallization or diffraction quality of these proteins was expectedly low. Nevertheless, except for proteinase K, where no observable effect could be detected, it was possible to assess that AuNPs can still have a significant effect on the morphology of crystals, more specifically on crystal size and number of crystals per drop. For all tested AuNPs, this was observed in the case of model protein HEWL. A similar observation was made in the case of RNase A incubated with AuNP-SPEG and AuNP-MUA (although with a slight delay in crystal formation). The existence of proteins corresponding to “difficult” case scenarios is always desirable, and this study included crystallization trials of six distinct proteins incubated with the AuNPs. Significant improvement in crystal size was observed for PAH, AOH, and albumin. In summary, it is remarkable that, in a significant number of cases, fewer, bigger crystals per drop were detected. We can conclude that the results here described reinforce that AuNPs can be used to improve the size and morphology of crystals when crystallization conditions are already known, but they do not appear to induce crystal growth in the first place. Used as additives prior to crystallization setups, the AuNPs seem to prevent excess nucleation, allowing the growth of single, larger crystals. This crystal enlargement effect seems to be more significant at lower pH. Furthermore, X-ray diffraction experiments using these crystals revealed that the presence of the AuNPs did not disrupt the crystals’ lattices or affect the respective diffraction patterns. The response of the proteins crystallization behavior in the presence of each type of AuNPs might be intrinsically related with its overall surface charge toward the AuNPs surface potential, which would then influence the nucleation process. So the nature of the AuNP surface, imparted by functionalization, is critical for crystal growth, for a particular protein with characteristic surface properties. The four types of functionalized AuNPs used in this study were revealed to be weak promoters of nucleation and were more effective as additives, even if this was not reflected in the X-ray diffraction properties of the tested crystals. Although performed for a limited variety of functionalized AuNPs, the important observations in this study will stimulate and inspire the design of distinct capping agents. Their physical and chemical potential should be further tested in the optimization of crystallization conditions of these or other case studies.

■

REFERENCES

(1) Blaber, M. G.; Ford, M. J.; Cortie, M. B. In Gold, Science and Applications; Corti, C., Holliday, R., Eds.; Taylor and Francis Group, LLC: 2010; Chapter 2. (2) Cortie, M.; McDonagh, A. In Gold Chemistry, Applications and Future Directions in the Life Sciences; Mohr, F., Ed.; WILEY-VCH Verlag GmbH & Co. KGaA: Weinheim, 2009; Chapter 7. (3) Schwerdtfeger, P.; Lein, M. In Gold Chemistry, Applications and Future Directions in Life Sciences; Mohr, F., Ed.; WILEY-VCH Verlag GmbH & Co. KGaA: Weinheim, 2009; Chapter 4. (4) Kim, J.; Grate, J. W.; Wang, P. Trends Biotechnol. 2008, 26, 639− 46. (5) Kurniawan, F. New Analytical Applications of Gold Nanoparticles. University of Regensburg, Germany, 2008, pp 2−50. (6) Hodzhaoglu, F.; Kurniawan, F.; Mirsky, V.; Nanev, C. Cryst. Res. Technol. 2008, 593, 588−593. (7) Turkevich, J.; Stevenson, P. C.; Hillier, J. Discuss. Faraday Soc. 1951, 11, 55−75. (8) Kimling, J.; Maier, M.; Okenve, B.; Kotaidis, V.; Ballot, H.; Plech, A. J. Phys. Chem. 2006, 110, 15700−15707. (9) Haiss, W.; Thanh, N. T. K.; Aveyard, J.; Fernig, D. G. Anal. Chem. 2007, 79, 4215−21. (10) Fitzpatrick, P. F. Annu. Rev. Biochem. 1999, 68, 355−381. (11) Garattini, E.; Terao, M. Expert Opin. Drug Metab. Toxicol. 2012, 4, 487−503. (12) Mansfield, J. H.; Wilhelm, J. E.; Hazelrigg, T. Development 2002, 129, 197−209. (13) Fischmann, T.; Hruza, A.; Niu, X. Da; Fossetta, J.; Lunn, C.; Dolphin, E.; Prongay, A.; Reichert, P. Nat. Struct. Biol. 1999, 6, 233. (14) Ferreira, M.; Nogueira, I. J. Bacteriol. 2010, 20, 5312. (15) Galli, C.; Bernasconi, R.; T. Soldà, Calanca, B.MolinariM. PLoS One 2011, 10.1371/journal.pone.0016304. (16) Chen, Y.; Hu, D.; Yabe, R.; Tateno, H.; Qin, S. Y.; Matsumoto, N.; Hirabayashi, J.; Yamamoto, K. Mol. Biol. Cell 2011, 19, 3559. (17) Jancarik, J.; Kim, S. H. J. Appl. Crystallogr. 1991, 24, 409−411. (18) Leslie, A. G. W. CCP4 ESF-EAMCB Newsl. Protein Crystallogr. 1992, 26. (19) Browne, K. P.; Grzybowski, B. A. Langmuir 2011, 27, 1246−50. (20) ExPASy, http://web.expasy.org/cgi-bin/compute_pi.

227


Use of Gold Nanoparticles as Additives in Protein Crystallization

Recommend Documents