Báo cáo khoa học: Effect of annealing time of an ice crystal on the activity of type III antifreeze protein pdf

8 471 0
  • Loading ...
    Loading ...
    Loading ...

Tài liệu liên quan

Thông tin tài liệu

Ngày đăng: 07/03/2014, 05:20

Effect of annealing time of an ice crystal on the activityof type III antifreeze proteinManabu Takamichi1,2, Yoshiyuki Nishimiya1, Ai Miura1and Sakae Tsuda1,21 Functional Protein Research Group, Research Institute of Genome-based Biofactory, National Institute of Advanced Industrial Science andTechnology (AIST), Toyohira, Sapporo, Japan2 Division of Biological Sciences, Graduate School of Science, Hokkaido University, Kita, Sapporo, JapanAntifreeze proteins (AFPs) are a structurally diverseclass of macromolecules that interact with water mole-cules located in the surface of an ice crystal at temper-atures below the melting point of the solution [1].Interaction of a substantial number of AFPs with theice surface modifies the shape of the ice crystal, result-ing in unique morphologies, such as a hexagonalbipyramid or hexagonal trapezohedron [2]. AFPs inhi-bit ice growth by adsorbing onto the ice surface(adsorption–inhibition model) [3,4] as the temperatureis lowered, and such inhibition becomes insufficientbelow a certain temperature that is favourable for theinitiation of crystal growth. For AFP solutions, this‘ice-growth initiation temperature’ (Tini) is differentfrom the melting point (Tm) of the ice crystal, and theTiniis referred to as the nonequilibrium freezing point(nonequilibrium Tf). The difference between Tmandthe nonequilibrium Tfis defined as thermal hysteresis(TH), and is generally used as a measure of the growthinhibition ability of an AFP [5]. Therefore, determina-tion of the Tiniand Tmof a seed ice crystal is an essen-tial procedure to evaluate the TH activity of AFP.However, the mechanism of ice binding that alters theTH value remains unclear [6].The TH value of AFPs has been evaluated usinga nanolitre osmometer and the ‘capillary technique’[5,7]. In the former, submicrolitre volumes of an AFPsolution are introduced into an oil droplet, the temper-ature of which is controlled by a Peltier device. TheTH value is determined by observing the growth of anKeywordsadsorption–inhibition; annealing time;notched-fin eelpout; thermal hysteresis;type III antifreeze proteinCorrespondenceS. Tsuda, Functional Protein ResearchGroup, Research Institute of Genome-basedBiofactory, National Institute of AdvancedIndustrial Science and Technology (AIST),2-17-2-1 Tsukisamu-Higashi, Sapporo062-8517, JapanFax: +81 11 857 8983Tel: +81 11 857 8912E-mail: s.tsuda@aist.go.jpWebsite: http://unit.aist.go.jp/rigb/fprg/index_e.html(Received 26 July 2007, revised 29 Septem-ber 2007, accepted 24 October 2007)doi:10.1111/j.1742-4658.2007.06164.xAntifreeze proteins (AFPs) possess a unique ability to bind to a seed icecrystal to inhibit its growth. The strength of this binding has been evalu-ated by thermal hysteresis (TH). In this study, we examined the dependenceof TH on experimental parameters, including cooling rate, annealing time,annealing temperature and the size of the seed ice crystal for an isoform oftype III AFP from notched-fin eelpout (nfeAFP8). TH of nfeAFP8 dramat-ically decreased when using a fast cooling rate (0.20 °CÆmin)1). It alsodecreased with increasing seed crystal size under a slow cooling rate(0.01 °CÆmin)1), but such dependence was not detected under the fast cool-ing rate. TH was enhanced 1.4- and 2.5-fold when ice crystals wereannealed for 3 h at 0.05 and 0.25 °C below Tm, respectively. After anneal-ing for 2 h at 0.25 °C below Tm, TH activity showed marked dependenceon the size of ice crystals. These results suggest that annealing of an icecrystal for 2–3 h significantly increased the TH value of type III AFP.Based on a proposed adsorption–inhibition model, we assume that type IIIAFP undergoes additional ice binding to the convex ice front over a 2–3 htime scale, which results in the TH dependence on the annealing time.AbbreviationsAFGP, antifreeze glycoprotein; AFP, antifreeze protein; nfeAFP8, an isoform of type III AFP from Notched-fin eelpout; TH, thermalhysteresis.FEBS Journal 274 (2007) 6469–6476 ª 2007 The Authors Journal compilation ª 2007 FEBS 6469ice crystal on the device through a microscope within acooling gradient. Previous studies have used a similarprocedure for the capillary technique with a larger icecrystal [5,6]. Duman (2001) demonstrated that consid-erably higher levels of TH are determined using thenanolitre osmometer compared with the capillary tech-nique for the same AFP sample [6]. The sample usedwas the AFP of the beetle Dendroides canadensis,which displayed TH of 1.4 °C using the capillary tech-nique, and 5.5 °C using the nanolitre osmometer [6].In addition, several physicochemical factors influenceTH activity [8], and there is a linear and negative cor-relation between TH activity and the logarithm of themass fraction of ice in a sample for cerambycid beetleRhagium inquisitor AFP [9,10]. A similar observationwas also reported for the common mealworm Tenebriomolitor, for which differential scanning calorimetrywas employed [11].For fish AFPs, TH activity of approximately 1 °Cisclose to the maximum value observed. Fish AFPs canbe classified into four types (I–IV) or as antifreeze gly-coproteins (AFGPs). Type I AFP, with relative mole-cular weights ranging from 3.3 to 5 kDa, containsalanine-rich amphipathic a-helices [12]. Type II AFP isa globular protein with an approximate relative molecu-lar weight of 14 kDa that exhibits high structuralhomology to a carbohydrate-recognition domain ofC-type lectin [13]. Type III AFP is a 6.5 kDa compactglobular protein characterized by a unique internal two-fold symmetry motif [14]. Type IV is assumed to form afour-helix bundle structure [15]. AFGPs are glycopro-teins whose relative molecular weight ranges from 3 to34 kDa, and comprises a repetitive tripeptide (Ala-Ala-Thr) whose Thr side chain is modified by a disaccharidemoiety [16]. The structural characteristics of these fishAFPs and their target ice crystal surface are differentfrom those of insect AFPs. A faster cooling ratedecreases the apparent TH value of fish AFPs, and lar-ger ice crystals tend to initiate crystal growth at highertemperatures [17]. The TH activity of a small species ofAFGP depends on the freezing rate (i.e. cooling bathtemperature), while a large AFGP showed no suchdependence [18]. Chapsky and Rubinsky (1997) exam-ined the kinetic ice binding of fish type I AFP using atechnique called temperature-gradient thermometry[19]. For fish type III AFP, no time-dependence of THactivity has been examined to date. These data raise thequestion about the key determinant that has a signifi-cant influence on the TH activity of an AFP species.The present study examined the TH value of arecombinant 65-residue type III AFP called nfeAFP8,which was recently discovered in the Notched-fineelpout [20]. nfeAFP8 exhibits 94% sequence identitywith HPLC12, a well-examined type III AFP isoformfound in the ocean pout Macrozoarces americanus [21].For TH measurements, we utilized a custom-madephotomicroscope system equipped with a temperaturecontroller and a sample holder, for which data consis-tency with a commercial freezing-point osmometer hadbeen verified. We then examined the dependence ofTH value on the cooling rate, the seed ice crystal size,crystal annealing time, and annealing temperature. Thedata obtained revealed that the TH value is signifi-cantly increased when the ice crystal is annealed for along time at a lower annealing temperature. Based onthese results, we discuss the ice-binding mechanism oftype III AFP that causes TH enhancement over a veryslow time scale.ResultsBefore determining the TH activity of nfeAFP8, theperformance of our photosystem was verified usingsolutions of NaCl and glucose. The sample solution inthe capillary tube (Fig. 1) was initially cooled with thetemperature controller until completely frozen. Thesample was then warmed until an ice crystal wasapparent. As expected, the ice crystal prepared in eachsample exhibited a ‘disk-like’ morphology, which didnot subsequently change [22]. We carefully manipu-lated the temperature controller to fix the diameter ofthe seed crystal at approximately 20 lm, and examinedwhether there was a difference between Tiniand Tm.It appeared that the temperatures were affected bya change in the temperature controller of only± 0.01 °C, consistent with equality between TiniandTmfor the solutions of NaCl and glucose, affirmingthat they possess no specific ice-binding ability (i.e.Tini¼ equilibrium Tf). In Fig. 2A, the concentrationdependence of the equilibrium Tf(¼ Tm) for NaCl andglucose are shown by open circles and open squares,respectively. The corresponding data measured using afreezing-point osmometer are also plotted in Fig. 2A(closed circles and closed squares). The linear profilesobtained with our photosystem and the osmometerperfectly overlapped for the NaCl and glucose solu-tions. These results indicated that our system preciselyevaluates the Tiniand Tmvalues of a seed ice crystal ina solution.An example of TH determination for nfeAFP8 usingour photosystem is shown in Fig. 2B, for whichnfeAFP8 was dissolved to a final concentration of0.1 mm. The figure depicts a series of images of an icecrystal formed in solution at )0.46 °C; the crystals hada bipyramidal structure due to adsorption of nfeAFP8onto their surfaces. Adsorption of the AFP stopped iceTime-dependent enhancement of thermal hysteresis M. Takamichi et al.6470 FEBS Journal 274 (2007) 6469–6476 ª 2007 The Authors Journal compilation ª 2007 FEBScrystal growth during the 0.01 °CÆmin)1temperaturedecrease (images a and b), but failed to stop growth at)0.81 °C (image c) and allowed subsequent crystalgrowth (images d–h). The nonequilibrium TfofnfeAFP8 was hence determined to be )0.81 °C. As aseparate experiment determined the Tmto be )0.30 °Cfor an ice crystal in the same sample, 0.51 °C is theTH value of 0.1 mm nfeAFP8 under these experimen-tal conditions.Figure 3A shows the concentration dependence ofthe TH activity of nfeAFP8 measured at cooling ratesof 0.01 °CÆmin)1(closed circles) and 0.20 °CÆmin)1(open circles). The TH activity obtained with a slowcooling rate (0.01 °CÆmin)1) was significantly (1.7-fold)higher than that obtained with the fast cooling rate(0.20 °CÆmin)1). Figure 3B is a plot of the TH depen-dence on the cooling rate of 0.1 mm nfeAFP8. The THvalue decayed exponentially with increasing coolingrate, and reached a plateau at a rate of approximately0.10 °CÆmin)1.When the relationship between TH and crystal sizewas assessed, a marked difference was noted betweenthe cooling rates of 0.01 and 0.20 °CÆmin)1(Fig. 3C).The TH activity for a slow rate of cooling(0.01 °CÆmin)1) markedly decreased with increasing icecrystal size, while that measured using a fast coolingrate (0.20 °CÆmin)1) exhibited only a slight decrease. Ahigh TH value (approximately 0.7 °C) was obtainedonly when a slow cooling rate (0.01 °C min)1) wasused for a small ice crystal.Figure 4A is a plot of TH activity measured for0.1 mm nfeAFP8 after the annealing of an ice bipyra-mid for 0–3 h. The annealing was performed at a Tmof )0.05 °Cor)0.25 °C. Each experiment used thesame cooling rate of 0.20 °CÆmin)1. The TH valuePhotomicroscope CCD Camera Object glas s Computer Temp. controlle r Display A B Crystal & Temp. monitor) LN 2 dewa r Stage Capillary holder Viewing hole Capillary (sample) Capillary Sample solution Oil Air Capillary holde r Freezing plate Fig. 1. Experimental set-up (photosystem)for viewing the growth and melting of icecrystal to determine thermal hysteresisvalue of an AFP solution. (A) The photosys-tem composed of a photomicroscope (LeicaDMLB100), a temperature controller (Lin-kam THMS 600), a Colorvideo 3CCD camera(Sony), and an image-processing computer.(B) The capillary cell containing a 0.75 lLsample solution, and setting of the sampleinto the holder.4.03.5AB0. 0 .25 1.000.750.50NaClGlucosephoto-system & osmometerΔTf(oC)Concentration (M)20μmhgfedcbaFig. 2. (A) Concentration dependence of the equilibrium freezingpoint of NaCl and glucose solutions. Open symbols represent thedata determined using the photosystem, and closed symbols arethose obtained using a commercial osmometer (Vogel, modelOM802). The error bars represent standard deviations of the dataobtained from at least three experiments. (B) Snapshots showingthe change in an ice bipyramid created in a 0.1 mM solution oftype III AFP from Notched-fin eelpout (denoted nfeAFP8, coolingrate 0.01 °CÆmin)1). (a) T ¼ –0.46 °C; (b) T ¼ –0.61 °C; (c)–(h)T ¼ –0.81 °C. The nonequilibrium freezing point of )0.81 °C wasevaluated from this experiment.M. Takamichi et al. Time-dependent enhancement of thermal hysteresisFEBS Journal 274 (2007) 6469–6476 ª 2007 The Authors Journal compilation ª 2007 FEBS 6471tended to increase in proportion to the crystal anneal-ing time. When the ice bipyramid was annealed for 3 hat a Tmof )0.05 °C, TH activity increased 1.4-foldcompared with time zero. This TH increase reached2.5-fold when the 3 h annealing was performed at aTmof )0.25 °C.Dependence of TH on the size of the seed ice crystalwas also examined for 0.1 mm nfeAFP8 at a coolingrate of 0.20 °CÆmin)1. The closed circles in Fig. 4B rep-resent TH values obtained after 2 h annealing of theice bipyramid at a Tmof )0.25 °C, and the open cir-cles, which are reproduced from Fig. 3C, depict com-parable data obtained without ice crystal annealing.TH decreased exponentially with increasing crystal sizeonly after 2 h annealing at a Tmof )0.25 °C, and noTH dependence on crystal size was evident when THwas measured soon after creation of the ice bipyramid.DiscussionIn the present study, we first examined the TH depen-dence of nfeAFP8 on the rate of cooling (Fig. 3), uti-lizing our in-house-built photomicroscope system.Data consistency between the system and a freezing-point osmometer (Fig. 2) validated the performance ofthe system for evaluating TH value. Using this system,we demonstrated that a slower rate of cooling ampli-fies TH, while a faster rate of cooling progressivelydiminishes TH until it reaches a plateau. A similar,but not identical, observation has been made for asmall species of AFGP using a freezing-point osmome-ter; the AFGP exhibited significant antifreeze activityupon slow cooling, but lacked such activity if cooledquickly [18]. In addition, a time-dependent change inTfof a type I AFP solution has been demonstrated00. 0.1 0.2 0.3 0.400. (mM)0.01oC/min0.20oC/min0 0.10 0.200.05 0.15 Cooling rate (oC/min)Thermal hysteresis (oC)Thermal hysteresis (oC)(slow)(fast)slow fast00. oC/min0.20oC/minIce crystal size (μm)03515 252010 30Thermal hysteresis (oC)(slow)(fast)CABFig. 3. (A) Concentration dependence of TH for a 20 lm long ice crystal measured using 0.01 °CÆmin)1(closed circles) and 0.20 °CÆmin)1(open circles) rates of cooling. (B) Dependence of TH on the rate of cooling (°CÆmin)1). (C) Dependence of TH on the size of the ice crystal.Each point represents the mean of three experiments and the error bar represents the standard deviation. All experiments were performedimmediately after preparation of an ice crystal in a 0.1 mM solution of nfeAFP8.Annealing time (h) 1.0 2.0 3.0Annealing Temp. = Tm - 0.25 °C1.42.51.0Ice crystal size (μm) 25201030 35Annealing Time = 2 hAnnealing Time = 0 hThermal hysteresis (°C) Thermal hysteresis (°C)ABAnnealing Temp.= Tm - 0.05 °CFig. 4. Influence of annealing time and ice crystal size on TH. (A)Thermal hysteresis of nfeAFP8 measured after annealing of an icebipyramid for a certain period of time (0–3 h). The measurementwas repeated for three times at the annealing temperatures ofTm)0.25 °C (closed circles) and Tm)0.05 °C (open circles) using acooling rate of 0.20 °CÆmin)1. (B) Dependence of thermal hysteresisof nfeAFP8 on the size of the seed ice crystal. Closed circles arethe data obtained after 2 h of annealing, and open circles are thosewithout annealing. All other experimental parameters are the sameas Fig. 3.Time-dependent enhancement of thermal hysteresis M. Takamichi et al.6472 FEBS Journal 274 (2007) 6469–6476 ª 2007 The Authors Journal compilation ª 2007 FEBSusing temperature-gradient thermometry [19]. No suchtime-dependent change in TH was reported for insectAFPs, while a linear and negative correlation wasdetected between TH and the logarithm of the massfraction of ice crystal versus the total water mass forAFP from cerambycid beetle R. inquisitor [9,10].We examined the dependence of TH on the crystalsize under fast (0.20 °CÆmin)1) and slow(0.01 °CÆmin)1) cooling rates to clarify the correlationbetween the cooling rate and seed ice crystal size.These results show that TH decreases with increasingcrystal size at the slow cooling rate, and does notdepend on the crystal size at the fast cooling rate.These results imply that the cooling rate dominates theTH dependence on the seed crystal size. As actual tem-perature control of the freezing plate in our system isstepwise, a different cooling rate will produce a differ-ent time lapse of cooling on a sample. In other words,there is a longer annealing period before starting theTH measurement at a slower cooling rate. We there-fore examined whether an annealing period of 2–3 hinfluenced the TH value of nfeAFP8. An annealingexperiment was performed within the hysteresis gap(0.25 and 0.05 °C below Tm), and TH was measured ata fixed cooling rate (0.20 °CÆmin)1). These parametersrevealed that the annealing time amplifies TH, theamplification level is enhanced by lowering of theannealing temperature, and TH shows dependence oncrystal size only after 2–3 h of annealing. As the time-dependent amplification of TH explains the detectionof a higher TH value at the slower cooling rate, it maybe concluded that the time of annealing is an essentialinfluence on the TH value of nfeAFP8.For type III AFP, the 3D structure and the ice-bind-ing mechanism have been extensively examined. High-resolution X-ray and NMR structures have revealedthat a remarkably flat and amphipathic surface isconstructed on type III AFP, enabling complementarybinding to the flat ice prism plane [14,21]. When thetype III AFP solution contains a seed ice crystal, theAFP molecules come closer to the ice through the dif-fusion process, for which an association constant of107)108s)1Æm)1can be assumed from the Smulochow-ski model [23]. The AFP molecules then undergoprompt binding onto the flat ice plane, which has beenthought to progress irreversibly according to theadsorption–inhibition mechanism at the ice–waterinterface [3,4]. Subsequently, convex ice fronts areformed between the ice-bound AFPs on the flat plane,as depicted in Fig. 5A. At this stage, the ice crystaladopts a bipyramidal shape. During subsequentTH measurements, the convex ice front overgrows(Fig. 5B, dashed line), allowing uncontrolled growth atthe nonequilibrium Tf. This leads to the detection of anonzero TH value at time zero (0 h of annealing).The height and curvature of the convex ice front aredefined by spacing between adsorbed AFP molecules(i.e. the Kelvin effect [24]), such that the TH valuereflects the fraction of ice-bound AFPs on the flat iceplane. Indeed, TH increases in proportion with thetotal concentration of nfeAFP8, as can be seen inFig. 3A. The fraction of AFPs on the flat ice plane isnot, however, changed during the present time-depen-dent experiment on TH (Fig. 4) as we fixed the con-centration of AFP at 0.1 mm. A plausible explanationfor the time-dependence of TH is that nfeAFP8 under-goes ‘secondary’ binding onto the convex ice frontover a very slow time scale (2–3 h; Fig. 5C). Indeed, arecent study on ice etching revealed that type III AFPcan bind to several ice planes in addition to the (10–10) prism plane [14]. TH is the level of supercoolingrequired to nucleate ice growth from the ‘weak’ convexice fronts, the growth of which is not strongly inhib-ited by AFP. If AFP undergoes secondary bindingCAFPlow THTH measurementAannealing time (2–3 h)TH measurementBDsolutioniceiceicehigh THiceFig. 5. Schematic drawing of ice growthinhibition of type III AFP based on the‘adsorption–inhibition’ model [3,4]. (A) Aconvex ice front is created after primarybinding of type III AFP onto the flat iceplane. (B) Overgrowth of the convex icefront, giving a low TH value. (C) Possiblesecondary ice binding of type III AFP ontothe convex ice front during the annealingperiod (2–3 h). (D) AFPs bound onto theconvex ice front narrow the growth area(dashed line), leading to an enhancement ofTH activity.M. Takamichi et al. Time-dependent enhancement of thermal hysteresisFEBS Journal 274 (2007) 6469–6476 ª 2007 The Authors Journal compilation ª 2007 FEBS 6473over a very slow time scale after the primary ice bind-ing, it stabilizes the convex ice front (Fig. 5D), leadingto the time-dependent change in TH activity. This con-cept further explains the increased level of time-depen-dent amplification of TH at the lower annealingtemperature (Fig. 4A). The convex ice front willbecome more accentuated by lowering the annealingtemperature, which will raise the opportunity for bind-ing of AFPs to the growing convex ice front, therebyenhancing TH activity. We recently monitored the flu-orescence of seed ice crystals at 30 min intervals in asolution of a fusion protein comprising nfeAFP8 andgreen fluorescence protein (nfeAFP8–GFP) (see supple-mentary Fig. S1). We observed a slight increase offluorescence intensity in a region proximal to theseed ice surface, accompanying a slight growth of icecrystals in proportion with the annealing time (0–3 h).It is noteworthy that such an intensity change wasobserved only at the lower annealing temperature(Tm)0.25 °C). These preliminary data are also consis-tent with the idea of progressive binding of AFP overa time scale of 2–3 h.A larger number of convex ice fronts are presumablylocated on a larger-sized ice crystal. If secondarybinding of AFP has the ability to stabilize convex icesurfaces, such stabilization becomes imperfect with anincreasing number of weak points on the ice crystal.This supposition is consistent with the data obtainedafter 2 h of annealing (Fig. 4B, closed circles), forwhich a reduction in TH was apparent with increasingsize of the ice crystal. The lack of significant sizedependence of TH activity (Fig. 4B, open circles)might be ascribed to nonstabilization of the weakpoint of any size of the crystals, because of the lack ofsecondary binding of AFP at time zero (0 h of anneal-ing). It should be noted that no apparent time-depen-dent TH increase was detected for different types ofAFP (type I AFP from great sculpin and type II AFPfrom Japanese smelt; data not shown). Therefore, it isnot clear whether such a slow accumulation process isa general mechanism for any type of AFP. Several paststudies have shown that the bipyramidal ice crystalcan be maintained for a long time [25], but no THdata have been reported for such an ice bipyramid. Webelieve that a simple TH measurement before and after2 h will enable us to learn more about the secondarybinding of AFP over a very slow time scale, which willbe detected as enhancement of the observed TH value.To summarize, we have successfully prepared arecombinant type III AFP isoform and carefully mea-sured its TH activity using an in-house-built photomi-croscope system. Measurement of TH dependence onvarious experimental parameters revealed, for the firsttime, that annealing time significantly causes anenhancement of TH activity of fish type III AFP. Theadsorption–inhibition model may explain this time-dependent change in TH if type III AFP undergoes sec-ondary binding to the convex ice front following bind-ing to the primary ice plane.Experimental proceduresSample preparation of AFPIIIA sample of the type III AFP isoform nfeAFP8 was pre-pared for TH measurements as described previously [20]with the following modifications. A soluble fraction con-taining a recombinant nfeAFP8 after sonication of Escheri-chia coli BL21 (DE3) was dialyzed against 50 mm citricacid buffer (pH 2.9). After dialysis, cation-exchange chro-matography was performed using an Econo-Pac High Scartridge (Bio-Rad, Hercules, CA, USA) with a linear NaClgradient (0–0.5 m)in50mm sodium citrate buffer (pH 2.9).The purified nfeAFP8 was concentrated (approximately7mgÆmL)1), and then diluted prior to TH determination.For all the TH experiments, the sample was dissolved in0.1 m ammonium bicarbonate (pH 7.9).TH measurement systemThe experimental set-up using a commercial photomicro-scope system (denoted as the ‘photosystem’) is illustrated inFig. 1. The two main instruments in this system were a LeicaDMLB100 photomicroscope (Leica Microsystems, Wetzlar,Germany) and a Linkam THMS 600 temperature controller(Linkam Scientific Instruments Ltd, Tadworth, Surrey, UK)equipped with a liquid nitrogen Dewar flask. The latter con-trolled the temperature of the freezing plate with an accuracyof 0.01 °C by combining the use of an electric heater andliquid nitrogen. The left portion of Fig. 1(B) illustrates theposition of the sample solution in a quarter of a Hirschmanncapillary tube (Hirschmann Labogera¨te, Heilbronn, Ger-many) (length 30 mm, diameter 0.92 mm); each end of thetube was sealed using mineral oil (approximately 1 lL each)to prevent vaporization of the sample solution (approxi-mately 0.75 lL). The sample-containing capillary was thenloaded into a custom-made copper capillary holder (diameter17 mm, thickness 2.5 mm) (Fig. 1B, right), which was placedinto the freezing plate on the cooling stage (Fig. 1A). Afterpositioning of the capillary tube, 15 lL of ethylene glycolwas poured through the viewing hole (Fig. 1B, right) toachieve thermal conductivity between the holder and capil-lary tube. The ice crystal image was captured using a Color-video 3CCD camera (Sony, Tokyo, Japan) (Fig. 1A), andthe temperature status was simultaneously viewed on a dis-play and saved as a video file on a personal computer(Fig. 1A). We determined Tfand Tmby monitoring a slowerTime-dependent enhancement of thermal hysteresis M. Takamichi et al.6474 FEBS Journal 274 (2007) 6469–6476 ª 2007 The Authors Journal compilation ª 2007 FEBSplayback of the video. The size and length of the ice bipyra-mid was evaluated from the crystal image captured by thevideo. For accurate determination of the crystal length, weinitially captured the image of a 30 lm fiber inserted into thesample solution, and found that the captured length neededto be multiplied by 1.2 to determine the actual length of thecrystal. Measurement of the Tfand Tmvalues was repeatedat least three times, and the values were averaged. Moredetails of the TH measurement procedure are described inResults.A model OM802 commercial freezing-point osmometer(Vogel, Giessen, Germany) was used to check the perfor-mance of the photosystem. We initially placed a 50 lL sam-ple into the cooling bath ()7 °C) of the osmometer. Whenthe sample temperature reached )7 °C, the instrumentautomatically inserted a frosty probe into the sample solu-tion to initiate the growth of ice crystals. Accordingly,latent heat was emitted owing to the phase transition fromliquid to solid state, which raised the temperature of thesample. After completion of the rise in temperature, an ice–water equilibrium state was achieved at a certain negativetemperature, which was the Tfvalue in this instrument.Before performing measurements, the osmometer was nor-malized with a standard fluid of 300 mOsmol.AcknowledgementsWe are grateful to Brian Sykes at the University ofAlberta (Edmonton, Canada) for fruitful discussionson the ice-binding mechanism of nfeAFP8.References1 Jia Z & Davies PL (2002) Antifreeze proteins: anunusual receptor–ligand interaction. Trends Biochem Sci27, 101–106.2 Davies PL & Hew CL (1990) Biochemistry of fishantifreeze proteins. FASEB J 4, 2460–2468.3 Raymond JA & DeVries AL (1977) Adsorption inhibi-tion as a mechanism of freezing resistance in polarfishes. Proc Natl Acad Sci USA 74, 2589–2593.4 Knight CA & DeVries AL (1994) Effects of apolymeric, nonequilibrium ‘antifreeze’ upon ice growthfrom water. J Crystal Growth 143, 301–310.5 DeVries AL (1986) Antifreeze glycopeptides andpeptides: interactions with ice and water. MethodsEnzymol 127, 293–303.6 Duman JG (2001) Antifreeze and ice nucleator proteinsin terrestrial arthropods. Annu Rev Physiol 63, 327–357.7 Chakrabartty A, Yang DS & Hew CL (1989) Structure–function relationship in a winter flounder antifreezepolypeptide. II. Alteration of the component growthrates of ice by synthetic antifreeze polypeptides. J BiolChem 264, 11313–11316.8 Li N, Andorfer CA & Duman JG (1998) Enhancementof insect antifreeze protein activity by solutes of lowmolecular mass. J Exp Biol 201, 2243–2251.9 Zachariassen KE & Husby JA (1982) Antifreeze effectof thermal hysteresis agents protects highly supercooledinsects. Nature 298, 865–867.10 Zachariassen KE, DeVries AL, Hunt B & Kristiansen E(2002) Effect of ice fraction and dilution factor on theantifreeze activity in the hemolymph of the cerambycidbeetle Rhagium inquisitor. Cryobiology 44, 132–141.11 Hansen TN & Baust JG (1988) Differential scanningcalorimetric analysis of antifreeze protein activity in thecommon mealworm, Tenebrio molitor. Biochim BiophysActa 957, 217–221.12 Sicheri F & Yang DS (1995) Ice-binding structure andmechanism of an antifreeze protein from winterflounder. Nature 375, 427–431.13 Gronwald W, Loewen MC, Lix B, Daugulis AJ,Sonnichsen FD, Davies PL & Sykes BD (1998) Thesolution structure of type II antifreeze protein reveals anew member of the lectin family. Biochemistry 37, 4712–4721.14 Antson AA, Smith DJ, Roper DI, Lewis S, Caves LS,Verma CS, Buckley SL, Lillford PJ & Hubbard RE(2001) Understanding the mechanism of ice binding bytype III antifreeze proteins. J Mol Biol 305, 875–889.15 Deng G, Andrews DW & Laursen RA (1997) Aminoacid sequence of a new type of antifreeze protein, fromthe longhorn sculpin Myoxocephalus octodecimspinosis.FEBS Lett 402, 17–20.16 Tachibana Y, Fletcher GL, Fujitani N, Tsuda S, MondeK & Nishimura S (2004) Antifreeze glycoproteins:elucidation of the structural motifs that are essential forantifreeze activity. Angew Chem 43, 856–862.17 Fletcher GL, Hew CL & Davies PL (2001) Antifreezeproteins of teleost fishes. Annu Rev Physiol 63, 359–390.18 Raymond JA & DeVries AL (1972) Freezing behaviorof fish blood glycoproteins with antifreeze properties.Cryobiology 9, 541–547.19 Chapsky L & Rubinsky B (1997) Kinetics of antifreezeprotein-induced ice growth inhibition. FEBS Lett 412,241–244.20 Nishimiya Y, Sato R, Takamichi M, Miura A & TsudaS (2005) Co-operative effect of the isoforms of type IIIantifreeze protein expressed in Notched-fin eelpout,Zoarces elongates Kner. FEBS J 272, 482–492.21 So¨nnichsen FD, DeLuca CI, Davies PL & Sykes BD(1996) Refined solution structure of type III antifreezeprotein: hydrophobic groups may be involved in theenergetics of the protein–ice interaction. Structure 4,1325–1337.22 DeLuca CI, Comley R & Davies PL (1998) Antifreezeproteins bind independently to ice. Biophys J 74, 1502–1508.M. Takamichi et al. Time-dependent enhancement of thermal hysteresisFEBS Journal 274 (2007) 6469–6476 ª 2007 The Authors Journal compilation ª 2007 FEBS 647523 Smoluchowski MV (1917) Versuch einer mathematis-chen Theorie der Koagulationskinetik kolloiderLo¨sungen. Z Phys Chem 92, 129–168.24 Wilson PW (1993) Explaining thermal hysteresis by theKelvin effect. Cryo-Lett 14, 31–36.25 Haymet ADJ, Ward LG, Harding MM & Knight CA(1998) Valine substituted winter flounder ‘antifreeze’:preservation of ice growth hysteresis. FEBS Lett 430,301–306.Supplementary materialThe following supplementary material is availableonline:Fig. S1. Time-dependent changes in photomicroscopeimage of an ice bipyramid in the solution of nfeAFP–GFP.This material is available as part of the online articlefrom http://www.blackwell-synergy.comPlease note: Blackwell Publishing is not responsiblefor the content or functionality of any supplementarymaterials supplied by the authors. Any queries (otherthan missing material) should be directed to the corre-sponding author for the article.Time-dependent enhancement of thermal hysteresis M. Takamichi et al.6476 FEBS Journal 274 (2007) 6469–6476 ª 2007 The Authors Journal compilation ª 2007 FEBS . Effect of annealing time of an ice crystal on the activity of type III antifreeze protein Manabu Takamichi1,2, Yoshiyuki Nishimiya1, Ai Miura1and. results in the TH dependence on the annealing time. AbbreviationsAFGP, antifreeze glycoprotein; AFP, antifreeze protein; nfeAFP8, an isoform of type III AFP
- Xem thêm -

Xem thêm: Báo cáo khoa học: Effect of annealing time of an ice crystal on the activity of type III antifreeze protein pdf, Báo cáo khoa học: Effect of annealing time of an ice crystal on the activity of type III antifreeze protein pdf, Báo cáo khoa học: Effect of annealing time of an ice crystal on the activity of type III antifreeze protein pdf