Dynamics and Mechanism of Laser Trapping-Induced Crystal Growth

---

Article pubs.acs.org/crystal

Dynamics and Mechanism of Laser Trapping-Induced Crystal Growth of Hen Egg White Lysozyme Published as part of the Crystal Growth & Design virtual special issue of selected papers presented at the 11th International Workshop on the Crystal Growth of Organic Materials (CGOM11 Nara, Japan), a joint meeting with Asian Crystallization Technology Symposium (ACTS 2014) Jing-Ru Tu,† Ken-ichi Yuyama,*,† Hiroshi Masuhara,*,† and Teruki Sugiyama*,†,‡ †

Department of Applied Chemistry and Institute of Molecular Science, National Chiao Tung University, Hsinchu 30010, Taiwan Instrument Technology Research Center, National Applied Research Laboratories, Hsinchu 30076, Taiwan

Downloaded by UNIV OF NEBRASKA-LINCOLN on September 14, 2015 | http://pubs.acs.org Publication Date (Web): September 8, 2015 | doi: 10.1021/cg501860k

‡

S Supporting Information *

ABSTRACT: We propose the dynamics and mechanism of laser trapping-induced crystal growth of hen egg-white lysozyme (HEWL). A continuous-wave near-infrared laser beam is used as a trapping light source and focused at a point 10 μm away from a target tetragonal HEWL crystal that is spontaneously generated in solution. Laser trapping of HEWL liquid-like clusters in solution increases local concentration in the focus, where the free motion and orientation of the clusters are strongly restricted, and the clusters show high rigidity and ordering. The cluster association and reorientation at the micrometer-sized focus is evolved to a large highly concentrated domain of the clusters, where the specific target crystal is grown. Initially, the high rigidity and ordering of the clusters strongly suppress the crystal grow rate compared to spontaneous crystal growth. Continuous laser trapping at the focus of the initially formed domain, however, leads to the transition to another domain with different concentration, rigidity, and ordering of the clusters, which surprisingly enhances the crystal growth rate. More interestingly, the clusters in both domains have anisotropic features reflecting the laser polarization direction, which also contributes to the crystal growth.

1. INTRODUCTION Crystallization is an indispensable process in various research and industrial fields for the purification of target compounds and obtaining a high-quality single crystal for X-ray crystallographic analysis. In particular, protein crystal growth has developed into an extensive research topic with numerous applications. For example, pharmaceutical companies can develop new drugs to combat diseases when the crystal structure of the associated proteins is revealed.1,2 A crucial limitation is that the crystal structure of proteins can be determined only after sufficient crystal growth. Protein crystal growth proceeds only under supersaturated conditions, and several experimental parameters including solution temperature, pH, solvent, and additives affect the solubility of proteins in solution. Researchers have demonstrated numerous experimental approaches to achieve supersaturation by optimizing such parameters.3 Furthermore, protein crystals spontaneously precipitated in solution grow according to the given solution conditions; hence, it is either impossible or very difficult to control the growth rate of a specific target protein crystal arbitrarily. Since the first demonstration by Ashkin in 1970,4 laser trapping has been widely used as a high-potential tool for © XXXX American Chemical Society

manipulating, three dimensionally, in solution, and without any mechanical contact, micrometer- and nanometer-sized targets such as polymers,5−7 biological organs,8 inorganic9 and metal particles.10 Toward application of the laser trapping method to crystal chemistry, we shifted the size of the target substances to molecules/clusters, and in 2007 we successfully demonstrated glycine crystallization for the first time by laser trapping with an intense focused continuous-wave (CW) near-infrared (NIR) laser beam.11 Since then, we have intensively studied laser trapping-induced crystallization, which is now called “laser trapping crystallization”. This method has been applied for the nucleation of several amino acids and can successfully control crystal polymorphism and pseudopolymorphism of glycine12 and L-phenylalanine13 by tuning power and polarization of the trapping laser. Furthermore, by laser trapping, we succeeded in controlling the crystal growth of glycine as well as its polymorphism in 2009.14 In that study, a target glycine crystal spontaneously generated and completely grew, and then the trapping laser was Received: December 23, 2014 Revised: August 1, 2015

A

DOI: 10.1021/cg501860k Cryst. Growth Des. XXXX, XXX, XXX−XXX

Article

Downloaded by UNIV OF NEBRASKA-LINCOLN on September 14, 2015 | http://pubs.acs.org Publication Date (Web): September 8, 2015 | doi: 10.1021/cg501860k

Crystal Growth & Design

molecules should be considered, and we estimated the local temperature elevation to be about 5.0 K, with the assumption that the temperature elevation is simply calculated by the ratio between the absorption coefficient of HEWL solutions at 1064 nm and that of neat D2O. A HEWL stock solution (80 mg/mL) was prepared by dissolving HEWL powder into a D2O acetate buffer solution (100 mM). The pD value was adjusted and corrected by adding 0.4 to the measured value of the pH meter (Beckman Coulter, pHi510).20 We prepared the initial sample solution (HEWL, 40 mg/mL; NaCl, 3.0% (w/v); pD = 5.1) by mixing equal volumes of the stock HEWL and the buffer solutions with 6.0% NaCl. We set the handmade chamber on the stage of a microscope (Olympus, IX 71), poured a small amount (40 μL) of the sample solution into the sample chamber, and then completely sealed it with a 24 × 32 mm borosilicate glass slide (Matsunami, thickness 0.12−0.17 mm) to prevent solvent evaporation. Figure 1

switched on. Immediately after the laser was focused at a position approximately 20 μm away from the crystal edge, the target crystal started growing toward the focal spot, with a growth rate of 1 μm/s. This growth was realized despite the fact that the crystal was not directly irradiated since the focus size was estimated to be approximately 1 μm. Later, we come to explain the unusual growth mechanism by newly proposing the formation of a large highly concentrated domain of the liquidlike clusters around the focus.15,16 In fact, in 2010, our group demonstrated the formation of a dense lens-like liquid droplet of glycine, approximately with 5 mm in diameter, by laser trapping, and found that its concentration of the droplet became twice the concentration to the original solution.17 Recently, we applied the laser trapping-induced crystal growth method to a hen egg-white lysozyme (HEWL) solution.18 We focused a CW NIR trapping laser beam at a position 10 μm away from the edge of a target HEWL crystal generated spontaneously and examined temporal changes in growth rate of {101} and {110} crystal faces. The growth behavior was completely different from that of spontaneous growth, which was explained by the mechanism as involving the formation of a highly concentrated domain of HEWL liquidlike clusters. The laser trapping of HEWL liquid-like clusters increases local HEWL concentration in the focal volume, where the clusters and D2O molecules are closely packed together through hydrogen-bonding interactions. This locally formed dense cluster domain extends from the focal volume to its outside due to mutual molecular/cluster interactions, which is possibly accompanied by various laser-induced phenomena such as convection flow and mass transfer. Consequently, a large highly concentrated cluster domain is formed around the focus. The target HEWL crystal is completely covered by the large domain, and the crystal is grown depending on concentration and rigidity of the clusters in the domain, which differs significantly from those of the initial homogeneous solution. Examination of the growth behavior simply by taking the mean of temporal changes in growth rate, however, showed large standard deviations, on which we proposed the formation of a single large cluster domain. In order to gain further insights into the dynamics and mechanism of the fascinating crystal growth induced by laser trapping, we have systematically examined laser power, polarization direction, and focal position dependences of the growth behavior for a special target HEWL crystal. Through the systematic investigation, we found that more unique processes of HEWL cluster domain formation are involved. In this work, we present new findings that two different kinds of highly concentrated cluster domains determine the crystal growth. The dynamics and mechanism are discussed in view of the formation of two types of domains of HEWL liquid-like clusters with different rigidity, ordering, and anisotropy of the clusters in these domains.

Figure 1. Optical setup for experiments on laser trapping-induced crystal growth of HEWL. shows an illustration of the handmade chamber used in this study (upper left). Glass cover slides, 24 × 50 mm and 24 × 32 mm in size were washed with detergent, acetone, and 1 M KOH, followed by plasma cleaning (Diener electronic, Atto). We then set the 24 × 50 mm glass slide on a precut silicone gasket with 2.0 mm thick and 20 mm in diameter (Grace Bio-Laboratories, Silicone isolator). Because the glass surface became hydrophilic sufficiently through these treatments, a portion of the sample solution poured into the sample chamber instantly extended and formed a solution thin-film. After a 40−60 min stationary incubation period at room temperature (25 ± 1.0 °C), a few tetragonal HEWL crystals were spontaneously generated. Under these given solution conditions, a tetragonal HEWL crystal was consistently generated. In order to minimize sample-dependent error for the crystal growth experiment, 40 μL of the fresh-mixed solution was added again into the sample solution after the spontaneous crystal generation, when the solution volume was increased to be 80 μL, and the thickness of the solution layer was estimated to be approximately 300 μm. Then, we completely covered the chamber again with cover glass, for which a laser trapping experiment was performed. 2.2. Optical Setup for Laser Trapping Experiment. Figure 1 schematically illustrates the optical setup for our laser trapping experiments. We introduced a 1064 nm CW Nd3+:YVO4 laser (Coherent, Matrix 1064-10-CW) into an inverted microscope and focused it at a point 3 μm above a glass-solution interface with an objective lens (Olympus, 60 magnification, NA = 0.90). We changed the polarization direction of the linearly polarized laser beam by changing an angle of quartz wave plates (Sigma Koki) and measured

2. EXPERIMENTAL PROCEDURES 2.1. Preparation of HEWL Buffer Solution and Sample Chamber. As a target for laser trapping-induced crystal growth, we used HEWL (Wako, for Biochemistry). D2O (Sigma-Aldrich, 99.9 atom %) was selected as a solvent instead of H2O in order to suppress laser heating due to photon absorption in overtone bands from the molecular vibration of solvents. Indeed, the local temperature elevation in a focal volume was estimated to be 22−24 K/W in H2O and 2.0 K/ W in D2O, respectively, when a 1064 nm laser is irradiated into a focal volume of about 1 μm using an objective lens with a high numerical aperture.19 In this study, the light absorption by both HEWL and D2O B

DOI: 10.1021/cg501860k Cryst. Growth Des. XXXX, XXX, XXX−XXX

Article

Crystal Growth & Design

Downloaded by UNIV OF NEBRASKA-LINCOLN on September 14, 2015 | http://pubs.acs.org Publication Date (Web): September 8, 2015 | doi: 10.1021/cg501860k

Figure 2. (a) A photograph of a tetragonal HEWL crystal and definition of its length (L) and width (W) for calculating crystal growth rates of the {101} and {110} faces, G101 and G110. (b) The schematic illustration of all laser irradiation conditions performed in this study and their abbreviations. Red lines represent a {110} face. Each experiment was carried out at different laser powers of 0.6 and 1.1 W. the laser power with a power meter (Newport, 842-PE) throughout the objective lens. The crystal growth behavior of HEWL with and without laser irradiation was observed by using a charge-coupled device (CCD) (Watec, WAT-231S2) equipped with the microscope. A visible 532 nm laser (Onset) was also introduced into the microscope with the same beam path as that of the trapping laser, in order to check the focal position and to measure the thickness of the HEWL solution before and after laser irradiation. 2.3. Determination of Growth Rate of Tetragonal HEWL Crystal and Abbreviations Representing Laser Irradiation Condition. Figure 2a shows appearance of a tetragonal HEWL crystal observed in the direction of its {110} face. There are only two kinds of crystal facesa quadrilateral face {101} and a hexagonal face {110}, so that the crystal shape can be determined just by the competitive balance of the growth rate of the {101} and {110} faces. We here defined the length (L) and width (W) of a tetragonal HEWL crystal as shown in Figure 2a. The numerical crystal growth rates of the {101} and {110} faces, hereafter G101 and G110, are estimated with reference to equations reported by Durbin et al. in 1986 as follows:21

G101 = 0.45 dL /dt , G110 = 0.5 dW /dt

Figure 3. Temporal changes of G101 and G110 for 30 min spontaneous crystal growth, with standard deviations.

Thus, the shape of the growing crystal was consistently kept, indicating that HEWL concentration remained nearly unchanged for 30 min. One notable result on spontaneous crystal growth is that G101 was always larger than G110 for all our samples. According to experimental results reported by Feher et al. in 1986, G101 is consistently larger than G110 under HEWL solution concentrations ranging from 1.5 to 7.0% (w/v), while the vice versa is true under the concentrations over 7.0% (w/ v).21 This means that considerably high HEWL concentration is necessary for appearance of a crossover point between G101 and G110. 3.2. Two Different Behaviors in Laser TrappingInduced Crystal Growth of HEWL. We systematically examined polarization direction and focal position dependences of G101 and G110 under four different irradiation conditions, as shown in Figure 2b. Furthermore, in order to examine their laser power dependence, the input laser power was tuned to 0.6 or 1.1 W for each condition. As we did for spontaneous crystal growth, G101 and G110 was estimated by measuring L and W of the growing crystal every 5 min. Figure 4 shows temporal changes of mean G101 and G110, which were estimated by repeating these experiments 10 times under each irradiation condition. The black and red horizontal dashed lines in the figures represent G101 and G110 on spontaneous crystal growth as a reference (G101: 3.3 nm/s, G110: 2.4 nm/s). The SD values on each data were also calculated as labeled in the figures, and those on 0.6 W- and 1.1 W-irradiation ranged from 0.3 to 2.2 and from 0.2 to 4.6 nm/s, respectively. For all irradiation conditions, we observed large suppression of G101 and G110 in the early stage of the irradiation and their subsequent

(1)

In this study, only a tetragonal HEWL crystal of 10−20 μm in length and with an L/W ratio of 1.5 ± 0.1 was selected as a starting crystal for our laser trapping experiments, whose experimental conditions are stricter than those in our previous study.18 The images of the growing crystal during laser irradiation was captured for every 5 min by a CCD camera and then were analyzed for calculating G101 and G110 with ImageJ software.22 For laser irradiation conditions, we set the laser focus to a point 10 μm away from {101} or {110} HEWL crystal faces, which is hereafter abbreviated as “{101} irradiation” or “{110} irradiation”, respectively. Figure 2b shows a schematic illustration of all laser irradiation conditions carried out in this study and their abbreviations. For example, when the trapping laser beam at 0.6 W is focused at a point of 10 μm away from the {101} face and the polarization direction is parallel to the {110} face, we here abbreviate the irradiation condition as “{101}-(V)-0.6”.

3. RESULTS AND DISCUSSION 3.1. Spontaneous Crystal Growth of HEWL. Figure 3 shows comparative experimental results of temporal changes of G101 and G110 on spontaneous crystal growth of HEWL without laser irradiation. A series of the experiments was carried out for 10 samples, and the mean G101 and G110 calculated by measuring L and W values of the growing crystal every 5 min are plotted in Figure 3. During 30 min observation time, both G101 and G110 showed almost constant with a standard deviation (SD) of less than 0.87 nm/s, when the mean G101 and G110 can be estimated to be 3.3 and 2.4 nm/s, respectively. C

DOI: 10.1021/cg501860k Cryst. Growth Des. XXXX, XXX, XXX−XXX

Article

Downloaded by UNIV OF NEBRASKA-LINCOLN on September 14, 2015 | http://pubs.acs.org Publication Date (Web): September 8, 2015 | doi: 10.1021/cg501860k

Crystal Growth & Design

Figure 4. Temporal changes of mean G101 and G110 for 10 samples under different irradiation conditions, with their standard deviations.

Figure 5. Representative temporal changes of G101 and G110 in (a) Case 1 and (b) Case 2 on {110}-(V)-1.1. Panels (c) and (d) show schematics of their temporal change in Case 1 and Case 2, respectively. The black and red dashed lines represent the G101 and G110 on spontaneous crystal growth as a reference.

classified as either two typical cases for all samples. Figure 5a,b shows two representative cases of the temporal change of G101 and G110 on {110}-(V)-1.1, which we here call Case 1 and Case 2, respectively. In Case 1, both G101 and G110 consistently showed almost constant and were strongly suppressed particularly at high laser power, compared to those on spontaneous growth. On the other hand, Case 2 also exhibited strong suppression of both G101 and G110 in the early stage of the irradiation, similar to that of Case 1, while their subsequent linear increase was observed after a certain irradiation time.

enhancement. We here should point out that the trend of growth rate change shown in Figure 4 is fully consistent with that in our previous work,18 and these results are completely different than those on spontaneous growth with constant growth rate. Since such a considerably large SD for the mean G101 and G110 was still observed in this work, we carefully checked their temporal changes for each sample individually. Consequently, we found the fascinating and critical fact that the temporal changes of G101 and G110 during laser irradiation can be D

DOI: 10.1021/cg501860k Cryst. Growth Des. XXXX, XXX, XXX−XXX

Article

Crystal Growth & Design Table 1. All Calculated Values for Mean G110 on Basis of Figure 5c,d (1.1 W) Case 1 in Figure 5c a irradiation condition

event number

{110}-(V)-1.1 {110}-(P)-1.1 {101}-(V)-1.1 {101}-(P)-1.1

7 6 6 6

a

Case 2 in Figure 5d

(1) b

mean G110 (nm/s)

event number

± ± ± ±

3 4 4 4

0.24 0.37 0.32 0.39

0.20 0.16 0.13 0.21

b

(2)

(3)

(4)

mean G110 before TT (nm/s)

TT (min)

mean slope of G110 (nm/s·min−1)

0.22 0.36 0.28 0.57

± ± ± ±

0.16 0.25 0.11 0.31

11 12 17 11

± ± ± ±

1.6 3.3 2.7 3.1

0.29 0.40 0.36 0.48

± ± ± ±

0.05 0.06 0.09 0.15

(1)−(4) in the table correspond to the labels in Figure 5c,d. bEvent number = Case 1 + Case 2 = 10.

Table 2. All Calculated Values for Mean G110 on Basis of Figure 5c,d (0.6 W) Case 1 in Figure 5c

Downloaded by UNIV OF NEBRASKA-LINCOLN on September 14, 2015 | http://pubs.acs.org Publication Date (Web): September 8, 2015 | doi: 10.1021/cg501860k

a

(1)

irradiation condition

event numberb

{110}-(V)-0.6 {110}-(P)-0.6 {101}-(V)-0.6 {101}-(P)-0.6

8 3 6 5

a

Case 2 in Figure 5d

mean G110 (nm/s)

event numberb

± ± ± ±

2 7 4 5

0.68 0.69 0.52 0.79

0.17 0.19 0.26 0.23

(2)

(3)

(4)

mean G110 before TT (nm/s)

TT (min)

mean slope of G110 (nm/s·min−1)

0.90 0.98 0.68 0.96

± ± ± ±

0.34 0.55 0.22 0.26

7.6 10 14 14

± ± ± ±

0.2 5.7 1.8 5.2

0.17 0.20 0.18 0.25

± ± ± ±

0.07 0.07 0.06 0.21

(1)−(4) in the table correspond to the labels in Figure 5c,d. bEvent number = Case 1 + Case 2 = 10.

Notably, the crossover point of the lines of G101 and G110 due to a large increase in G110 was sometimes observed. We here define the certain irradiation time leading to linear increase in G101 and G110 as “transition time (TT)”. Simply, Case 1 and Case 2 can be classified as whether or not TT is observed within 30 min of laser irradiation. The TT values for G101 were not easily estimated compared to that for G110, since TT for G101 consistently appeared faster than that of G110 and was often observed even before the initial 5 min irradiation, and the slope of G101 after TT was not large compared to that of G110. Furthermore, it is reported that G110 on spontaneous crystal growth almost linearly increases with HEWL concentration, while G101 gradually saturates with the concentration.21,23 Therefore, we paid particular attention to TT for G110 in this study. In order to elucidate the dynamics and mechanism of the unique crystal growth of HEWL induced by laser trapping, we conducted qualitative analysis for temporal changes of G101 and G110 individually for all samples. Figure 5c,d shows schematic illustrations of Case 1 and Case 2, respectively. For samples showing Case 1, we calculated the mean values of constant G110 (Figure 5c (1)). The details of our calculation method are shown in Supporting Information 1 (SI 1). For samples showing Case 2, we performed a linear least-squares fitting of G110 as a function of laser irradiation time and then calculated the mean values of the slopes of G101 and G110 for all samples under each condition (Figure 5d (4)). In addition, we calculated the mean values of constant G110 before TT (Figure 5d (2)) and determined TT values of G110 as an intersection point of the constant lines and the fitted lines (Figure 5d (3)). All calculated results are summarized in Table 1 (1.1 W) and Table 2 (0.6 W), with their SDs. The numbers in parentheses in these tables correspond to the labels in Figure 5c,d. After our novel classification of Case 1 and Case 2, these SD values in Tables 1 and 2 are significantly lower than those in Figure 4, indicating that our classification yields much better fitting to the experimental results. 3.3. Two Different Highly Concentrated Domain of HEWL clusters Formed by Laser Trapping. We here consider two different highly concentrated domains of HEWL

clusters formed by laser trapping on the basis of Tables 1 and 2. We first note strong suppression of mean G110 seen in Case 1, which are estimated to be 0.24−0.39 and 0.52−0.79 nm/s at 1.1 and 0.6 W, respectively ((1) of Tables 1 and 2), compared to spontaneous crystal growth (2.4 nm/s). Since supersaturation is a driving force behind crystal growth according to conventional crystal growth theory in isotropic and homogeneous solution, the strong suppression of G110 will be first considered to be due to the decrease in SS of solution. There are two plausible and considerable causes of the decrease in SS, temperature elevation by laser heating and a decrease in HEWL concentration around the target crystal by laser trapping. For temperature elevation by laser heating, we have already reported that little or no effect of the temperature elevation contributes to the crystal growth rate.18 Furthermore, we also confirmed in this study that temperature elevation of a glass substrate by laser irradiation is negligible for the crystal growth rate. We carried out the same trapping experiment by replacing a glass substrate with a fused quartz slide showing a much lower absorption coefficient at 1064 nm.24 Consequently, the same degree of the decrease in G110 was still observed (see SI 2). This means that the temperature elevation of a glass substrate by laser irradiation is not responsible for the large suppression of the crystal growth rate. On the other hand, we consider that the decrease in HEWL concentration around the target crystal by laser trapping at the focus is also not a main reason for the large suppression of the crystal growth rate, since the decrease in the concentration around the target crystal should be easily and instantly compensated by the diffusion of the surrounding molecules/clusters. Furthermore, it is unlikely for the compensation not to be completed during 30 min irradiation. Thus, we should consider another laser trappinginduced phenomenon to be responsible for the strong suppression. As we previously proposed,18 the strong suppression of G101 and G110 can be explained in terms of the formation of a highly concentrated domain of HEWL clusters by laser trapping. Namely, the clusters in the domain show high rigidity and ordering, and the supplies of clusters from the domain to the target crystal are disrupted, which leads E

DOI: 10.1021/cg501860k Cryst. Growth Des. XXXX, XXX, XXX−XXX

Article

Downloaded by UNIV OF NEBRASKA-LINCOLN on September 14, 2015 | http://pubs.acs.org Publication Date (Web): September 8, 2015 | doi: 10.1021/cg501860k

Crystal Growth & Design

Figure 6. (a−d) Schematic illustrations of the dynamics and mechanism of laser trapping-induced HEWL crystal growth through formation of two different domains of the clusters.

clusters in both Domain 1 and Domain 2 have an anisotropic feature. In addition, (P)-irradiation leads to a preferential orientation of the clusters inside the domains to complementary sites on {110} face, which enhances the crystal growth rate. Several previous reports support our interpretation for the laser polarization dependent crystal growth of HEWL. Many researchers have revealed that HEWL crystal growth proceeds through an adsorption of the cluster units, not single HEWL molecules using relative light scattering25 and electron microscopy.26 Furthermore, in 1996, Nadarajah et al. experimentally and theoretically demonstrated that each HEWL clusters of dimer, tetramer, and octamer have a complementary site to the crystal faces of {101} or {110}.27 In addition, the orientation of HEWL crystals on the spontaneous crystal growth was controllable by electric28,29 and magnetic30,31 fields. 3.4. Dynamics and Mechanism of Laser TrappingInduced Crystal Growth of HEWL. Figure 6 shows schematic illustrations of the proposed mechanism of HEWL crystal growth inside two different highly concentrated domains of the clusters formed by laser trapping. Since SS of HEWL/ D2O buffer solution used in this work was estimated to be over 3.0,32 it can be considered that HEWL molecules in the solution are strongly linked through D2O molecules and themselves and form the large liquid-like clusters.33,34 Therefore, the efficient laser trapping of the clusters should be realized in a focal volume, where the cluster concentration is extremely increased. The increase in the local concentration eventually overcomes an energy barrier leading to formation of the first highly concentrated domain of the clusters, Domain 1, in the focus (see Figure 6a). The free motion and orientation of the clusters in the domain are strongly limited and restricted by laser trapping, and the clusters in the domain show high rigidity and ordering. The locally formed domain with high rigidity and ordering of the clusters at the micrometer-sized focus instantly extends to its outside through mutual intercluster interactions (Figure 6b) and entirely covers a specific target crystal within 5 min (Figure 6c). The growth rate of HEWL crystal inside the domain is strongly suppressed due to the high rigidity and ordering of the clusters. The cluster rigidity and ordering are

to the strong suppression. We here call such a domain seen in Case 1 “Domain 1”. On the other hand, Case 2 showed two different laser trapping-induced crystal growth behaviors: strong suppression of G101 and G110 similar to that in Case 1 and subsequent linear increase after TT (Figure 5d). Note here that the irradiation time in this work was fixed for 30 min in order to avoid an edge of the growing crystal overlapping with the laser focus. However, we confirmed that TT and the following linear increase of G110 are consistently observed for all samples, when laser irradiation for longer than 30 min is conducted. This result strongly supports that Case 1 is consistently followed by Case 2. Therefore, the strong suppression of G101 and G110 seen in both cases ((1) and (2) of Tables 1 and 2) is considered to be due to the high rigidity and ordering of the clusters in Domain 1. On the other hand, as seen in (4) of Tables 1 and 2, the mean slopes of G110 at 1.1 W were almost twice those at 0.6 W under corresponding irradiation conditions, namely, the mean slopes after TT strongly depended on input laser power. This laser power dependence seems to be opposite to that seen in Domain 1, where laser irradiation at high laser power suppresses G101 and G110 to a greater degree. This opposite laser power dependence strongly supports the formation of other domain with physical and/or chemical properties. Namely, subsequent irradiating at the focus of Domain 1 induces the transition to different domain. We call such a domain after the transition “Domain 2” in this study, and the formation of Domain 2 should be responsible for the linear increase of G110 seen in Case 2. More interestingly, the crystal growth rate depended on the laser polarization direction. For examples, as seen in (1) of Tables 1 and 2, the mean G110 on {101}-(P) or {110}-(P) is consistently larger than that on {101}-(V) or {110}-(V) at same laser power, respectively, although that on {110}-(V)-0.6 and {110}-(P)-0.6 are close to the same. The similar trend is also observed for the mean slopes of G110 after TT as shown in (4) of Tables 1 and 2. Thus, (P)-irradiation always yields higher G110 compared to (V)-irradiation, independent of laser power and focal position. These results clearly indicate that the F

DOI: 10.1021/cg501860k Cryst. Growth Des. XXXX, XXX, XXX−XXX

Downloaded by UNIV OF NEBRASKA-LINCOLN on September 14, 2015 | http://pubs.acs.org Publication Date (Web): September 8, 2015 | doi: 10.1021/cg501860k

Crystal Growth & Design

■

considered to become high with input laser power, and in fact, laser trapping at high laser power led to strong suppression of crystal growth rate. In addition, the clusters in the domain show anisotropic features reflecting laser polarization direction, which causes laser polarization-dependent crystal growth. Such an anisotropic feature is never observable in homogeneous solution. Subsequent laser irradiation at the focus of the initially formed domain, Domain 1, continuously traps and collects the clusters, and further local concentration increase is achieved. Consequently, the clusters at the focus of Domain 1 become more tightly packed, and at a certain time (TT), the increase in the inner stress between the clusters may cause reorganization of HEWL hydrogen-bonding network due to conformational change in the molecular and/or cluster structure, which triggers transition to another cluster domain (Domain 2) with different concentration, rigidity, and ordering (Figure 6c). Because of the reorganization of HEWL hydrogen-bonding network, the rigidity and ordering of the clusters in Domain 2 possibly become lower compared to those in Domain 1. Consequently, HEWL concentration becomes a relatively dominant factor, and the crystal growth rate is enhanced (Figure 6d). The clusters in Domain 2 are still anisotropically oriented, which leads to the laser polarization-dependent crystal growth.

Article

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/cg501860k. SI 1: calculation method for mean G110 in Case 1; SI 2: laser trapping-induced HEWL crystal growth using a quartz-slide-bottomed chamber (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*(T.S.) Tel.: +886 3 5712121. Fax: +886 3 5723764. E-mail: [email protected]. *(H.M.) Tel.: +886 3 5712121. Fax: +886 3 5723764. E-mail: [email protected]. *(K.Y.) Tel.: +886 3 5712121. Fax: +886 3 5723764. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The present work is partly supported by the Ministry of Science and Technology of Taiwan to T.S. (MOST 104-2113-M-492001-), to H.M. (MOST 103-2113-M-009-003), and to K.Y. (MOST 103-2113-M-009-022-MY2), and the MOE-ATU Project (National Chiao Tung University) of the Ministry of Education, Taiwan, to H.M..

4. CONCLUSION We have studied the dynamics and mechanism of laser trapping-induced crystal growth of HEWL. The laser power, polarization, and focal position dependences of the crystal growth were systematically investigated, and the crystal growth rate was calculated individually for all samples under each irradiation condition. Through this work, we obtained new knowledge that laser trapping produces two kinds of large highly concentrated domains of the liquid-like clusters with different cluster concentration, rigidity, and ordering, which determines the crystal growth rate. Laser trapping of the clusters first leads to the formation of Domain 1, and continuous laser irradiation at the focus of Domain 1 then induces transition to another cluster domain, Domain 2. The clusters in Domain 1 show high rigidity and ordering, which extremely suppressed the crystal growth rate compared that on spontaneous crystal growth. The rigidity and ordering of the clusters after the transition to Domain 2 become low due to high concentration increase of the clusters, which inversely enhances the crystal growth rate. Interestingly, the clusters in both domains exhibited anisotropic features, which cause laser polarization-dependent crystal growth. Concentration, ordering, rigidity, and anisotropy of the clusters in domain are initially achieved in a focal volume by laser trapping and then extend to the outside of the focus. This means that the chemical and physical properties of the domain are controllable by optimizing laser parameters such as power, polarization, and focal position of trapping laser. This crystal growth method is in principle applicable for all proteins, so that we can expect that precision control of crystal growth of a specific protein crystal in solution is made possible by optimizing laser parameters. We are now developing our study on the formation dynamics and mechanism of a highly concentrated domain of protein clusters by using threedimensional imaging and spectroscopic analysis of fluorescent dye-labeled proteins. These results on our work will be reported in the near future.

■

REFERENCES

(1) Shekunov, B. Y.; York, P. J. Cryst. Growth 2000, 211, 122−136. (2) McPherson, A.; Gavira, J. A. Acta Crystallogr., Sect. F: Struct. Biol. Commun. 2014, 70, 2−20. (3) McPherson, A. In Crystallization of Biological Macromolecules; Spring Harbor Laboratory Press: New York, 1999. (4) Ashkin, A. Phys. Rev. Lett. 1970, 24, 156−159. (5) Hofkens, J.; Hotta, J.; Sasaki, K.; Masuhara, H.; Taniguchi, T.; Miyashita, T. J. Am. Chem. Soc. 1997, 119, 2741−2742. (6) Borowicz, P.; Hotta, J.; Sasaki, K.; Masuhara, H. J. Phys. Chem. B 1997, 101, 5900−5904. (7) Masuo, S.; Yoshikawa, H.; Asahi, T.; Masuhara, H.; Sato, T.; Jiang, D.-L.; Aida, T. J. Phys. Chem. B 2002, 106, 905−909. (8) Yang, A. H. J.; Moore, S. D.; Schmidt, B. S.; Klug, M.; Lipson, M.; Erickson, D. Nature 2009, 457, 71−75. (9) Hansen, P. M.; Bhatia, V. K.; Harrit, N.; Oddershede, L. Nano Lett. 2005, 5, 1937−1942. (10) Friese, M. E. J.; Nieminen, T. A.; Heckenberg, N. R.; Rubinsztein-Dunlop, H. Nature 1998, 394, 348−350. (11) Sugiyama, T.; Adachi, T.; Masuhara, H. Chem. Lett. 2007, 36, 1480−1481. (12) Yuyama, K.; Rungsimanon, T.; Sugiyama, T.; Masuhara, H. Cryst. Growth Des. 2012, 12, 2427−2434. (13) Yuyama, K.; Wu, C. S.; Sugiyama, T.; Masuhara, H. Photochem. Photobiol. Sci. 2014, 13, 254−260. (14) Sugiyama, T.; Adachi, T.; Masuhara, H. Chem. Lett. 2009, 38, 482−483. (15) Chang, Y. C.; Myerson, A. S. AIChE J. 1986, 32, 1567−1569. (16) Erdemir, D.; Lee, A. Y.; Myerson, A. S. Acc. Chem. Res. 2009, 42, 621−629. (17) Yuyama, K.; Sugiyama, T.; Masuhara, H. J. Phys. Chem. Lett. 2010, 1, 1321−1325. (18) Tu, J. R.; Miura, A.; Yuyama, K.; Masuhara, H.; Sugiyama, T. Cryst. Growth Des. 2014, 14, 15−22. (19) Ito, S.; Sugiyama, T.; Toitani, N.; Katayama, G.; Miyasaka, H. J. Phys. Chem. B 2007, 111, 2365−2371. (20) Glasoe, P. K.; Long, F. A. J. Phys. Chem. 1960, 64, 188−190. (21) Durbin, S. D.; Feher, G. J. Cryst. Growth 1986, 76, 583−592. G

DOI: 10.1021/cg501860k Cryst. Growth Des. XXXX, XXX, XXX−XXX

Article

Downloaded by UNIV OF NEBRASKA-LINCOLN on September 14, 2015 | http://pubs.acs.org Publication Date (Web): September 8, 2015 | doi: 10.1021/cg501860k

Crystal Growth & Design (22) Schneider, C. A.; Rasband, W. S.; Eliceiri, K. W. Nat. Methods 2012, 9, 671−675. (23) Monaco, L. A.; Rosenberger, F. J. Cryst. Growth 1993, 129, 465− 484. (24) Miller, D. C.; Kempe, M. D.; Kennedy, C. E.; Kurtz, S. R. Opt. Eng. 2011, 50, 013003. (25) Pusey, M. L. J. Cryst. Growth 1991, 110, 60−65. (26) Durbin, S. D.; Feher, G. J. Mol. Biol. 1990, 212, 763−774. (27) Nadarajah, A.; Pusey, M. L. Acta Crystallogr., Sect. D: Biol. Crystallogr. 1996, 52, 983−996. (28) Al-Haq, M. I.; Lebrasseur, E.; Tsuchiya, H.; Torii, T. Crystallogr. Rev. 2007, 13, 29. (29) Nanev, C. N.; Penkova, A. J. Cryst. Growth 2001, 232, 285−293. (30) Sazaki, G.; Yoshida, E.; Komatsu, H.; Nakada, T.; Miyashita, S.; Watanabe, K. J. Cryst. Growth 1997, 173, 231−234. (31) Sazaki, G.; Moreno, A.; Nakajima, K. J. Cryst. Growth 2004, 262, 499−502. (32) Cacioppo, E.; Pusey, M. L. J. Cryst. Growth 1991, 114, 286−292. (33) Gliko, O.; Pan, W.; Katsonis, P.; Neumaier, N.; Galkin, O.; Weinkauf, S.; Vekilov, P. G. J. Phys. Chem. B 2007, 111, 3106−3114. (34) Sleutel, M.; Van Driessche, A. E. Proc. Natl. Acad. Sci. USA 2014, 111, E546−553.

H

DOI: 10.1021/cg501860k Cryst. Growth Des. XXXX, XXX, XXX−XXX