TRYPSIN ADSORP‘E'EON ONTO REGEEfiEATEfi CELLULQfi DEALY $13 MEMEMNE “wash {123* {fig Dag?” of M. S. MCBLGAN STATE UNEVERLSETY Naotada. Kobamcte 2968 THESIS 5'" "L. LIBRARY Michigan 3mm Univcmiry ——oq ABSTRACT TRYPSIN ABSORPTION ONTO REGENERATED CELLULOSE DIALYSIS MEMBRANE by Naotada Kobamoto Trypsin adsorption onto cellulose material provides an excellent system for studying protein adsorption at a solid/ liquid interface. Further, these studies may provide some insight into the nature of protein adsorption onto the cellu- lose portion of the cell wall. Many kinds of cellulose materials (filter paper, cotton, and collodion membrane) have been used for earlier protein adsorption studies. Eventhough many isolation and purification processes used regenerated cellulose dialysis membranes, there has been no previous research on adsorption of proteins onto this type membrane. Trypsin accumulation onto dialysis membranes is the net result of the dynamic processes of adsorption and desorption. Both processes are affected by a number of factors. For example, the higher the concentration of the trypsin solu- tion in which the membranes are immersed the greater is the accumulation. The pH of a solution also changes markedly the amount of accumulation: e.g., there are maxima in the rate of accumulation of pHs 6.5 and 9.5 (both of these pH values Naotada Kobamoto are lower than the stated isoelectric point of 10.8 for trypsin). Desorption occurs at a lesser rate throughout the pH region of 6.5-10.5 and follows an inverse function to that for accumulation. Increasing the phOSphate buffer con- centration also lowers the rate of trypsin accumulation and correspondingly increases desorption. The thermal coefficient for accumulation depends on the temperature at which the membranes are kept prior to immersion in the protein solution. This fact suggests that the size of the pores--whose presence is demonstrated by the passage of lower molecular weight compounds through the dialysis membranes and by electron microsc0pic studies—-may be one of the major factors determining the rate of accumu- lation. If regenerated cellulose dialysis membrane is kept at 40C prior to adsorption, a thermal coefficient (which presumably is associated with thermal expansion of the pores) is calculated to be 5 kcal/mole. This is larger than the thermal coefficient of 5 kcal/mole measured for what is identified as chemisorption. Since the thermal coefficient for desorption is 9—15 kcal/mole, the heat of adsorption is calculated to be 6-10 kcal/mole. The stability of adsorbed trypsin was measured on air- dried (250C) membranes and those kept in distilled water for various lengths of time. For membranes maintained in the Naotada Kobamoto dry state 40-50% of the initial enzymic activity is lost within an hour, but the remainder is stable for days. By contrast the membranes stored in distilled water lost activity slowly but continuously for 6 days. TRYPSIN ABSORPTION ONTO REGENERATED CELLULOSE DIALYSIS MEMBRANE BY Naotada Kobamoto A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Biophysics 1968 ACKNOWLEDGMENTS The author wishes to express his sincere appreciation to Professor Leroy G. Augenstein for his guidance and en- couragement through the performance of this research and the preparation of this thesis. Appreciation is extended to the United States Civil Administration of the Ryukyu Islands for providing financial assistance through much of the period of this study. The latter part of this study was done under the support and auspices of Grant CA-06634-O6 from the National Institutes of Health. ii TABLE OF CONTENTS ACKNOWLEDGMENTS. . . . . . . . . . . . . . LIST OF FIGURES. . . . . . . . . . . . . . INTRODUCTION . . . . . . . . . . . . . . . MATERIALS AND METHODS. . . . . . . . . . . RESULTS AND DISCUSSION . . . . . . . . . . Selective Accumulation and Desorption Other Factors Affecting Accumulation. Importance of Porous Structure. . . . CONCLUSIONS. . . . . . . . . . . . . . . . REFERENCES 0 C O O O O O O O O I O O O O . iii of Trypsin. Page ii iv 10 .10 24 55 51 55 Figure l. 10. 11. LIST OF FIGURES Page The effect of trypsin concentration in solution on the rate of trypsin accumulation onto dialysis membranes . . . . . . . . . . . . . . . 12 The accumulation of trypsin onto dialysis mem- brane as a function of time and buffer content . 14 The changes in the relative protein content and enzymic activity remaining in the supernatant after accumulation for various lengths of time . 17 The effect of different "washing" solutions on the desorption of active trypsin from the mem- branes . . . . . . . . . . . . . . . . . . . . . 19 The stability of tryptic activity adsorbed on thg membranes kept dry or in distilled water at 24 C . . . . . . . . . . . . . . . . . . . . . . 25 The dependence of accumulation on the ionic strength of the pH 7 phOSphate buffer. . . . . . 26 The effect of the ionic strength of the phos- phate buffer on the rate of desorption measured by an assay of activity. . . . . . . . . . . . . 29 The effect of pH on the rate of accumulation and desorption . . . . . . . . . . . . . . . . . 31 The effect of temperature on trypsin accumula- tion as a function of the temperature at which the membranes are kept prior to the adsorption . 55 The effect of temperature on the accumulation or desorption of trypsin in pH 5.5 solutions having a 10-2 M phOSphate buffer . . . . . . . . 37 The temperature dependence of trypsin desorption from dialysis membranes when the membranes were kept at different temperatures prior to the desorption . . . . . . . . . . . . . . . . . . . 40 iv LIST OF FIGURES - Continued Figure Page 12. The temperature dependence of desorption (measured by BABE assay) after 2 minute adsorp- tgon fromOS M trypsin solutions maintained at 4 C or 50 C . . . . . . . . . . . . . . . . . . 43 15. The dependence of the rate of desorption of tryptic activity on the adsorption temperature. 47 INTRODUCTION There has been relatively little research on the nature of protein adsorption at liquid/solid interfaces. Zittle (1955) reviewed adsorption of enzymes and other proteins onto chromatographic materials and Schwimmer and Pardee (1955) have discussed the purification of enzymes using adsorption methods. Recently, James and Augenstein (1966) reviewed enzyme adsorption at various interfaces including liquid/ solid interfaces. Unfortunately our understanding of enzyme structure and function at a solid/liquid interface makes it difficult to comment in detail about enzymes in living cells, although we are sure, from electron microscopic studies, that a cell is primarily a collection of interfaces and membranes. Bradfute and McLaren (1964) have studied the adsorption of lysozyme and RNase onto barley root tips and how this affects enzyme activity. At a liquid/solid interface we must consider at least two cases: the substrates are localized at the interface and the enzymes are free in solution, or the enzymes are localized at the interface while the substrates migrate freely. The former case was studied by McLaren and Estermann (1957) and Estermann g£_al. (1959) for a system composed of heat-denatured lysozyme adsorbed at a kaolinite/water interface plus the enzyme chymotrypsin in the solution. The latter case has been investigated by Mortland and Gieseking (1955) with phosphotases adsorbed onto various clays sus- pended in water containing organic phOSphorous compounds and by McLaren and Estermann (1956) with chymotrypsin adsorbed at a kaolinite/water interface. The present research reflects an attempt to study the effects on biological activity of adsorbing an enzyme at a liquid/solid interface. Specifically, we have directed our attention to the following questions: How does trypsin ad- sorption and desorption depend on time, trypsin concentration, pH, ionic concentration, and temperature? Is there any selective adsorption and desorption of different trypsin conformers? What is the effect of adsorption on enzymic function and how stable is trypsin adsorbed onto the membranes? Adsorption of a number of proteins onto cellulose materi- ‘als (filter paper, fibers, and collodion membranes) has been reported. These previous studies have been mainly concerned with the practical aSpects of minimizing the loss of proteins during isolation and/or purification or of utilizing for purification purposes the selective adsorption of a desired protein from a protein mixture onto cellulose materials. In 1918 Wood noticed that if a piece of filter paper is immersed for a few hours in a trypsin solution some of the enzyme is adsorbed. The same was found for pepsin by Effront (1922), for pepsin, rennin, and catalase by Tauber (1956), and for albumin by Robinson, Price, and Hogden (1957). As a result, the filtering techniques used during the isolation of various proteins have been redesigned to avoid this complication (see for example Harris, 1959). Regenerated cellulose membranes have only been available in recent years. As a result, most past studies of the adsorption and also structure of cellulose membranes have dealt with collodion membranes. However, several aspects of our study strongly suggest that there are similarities in protein adsorption onto collodion membranes and our regen- erated cellulose (viscose) dialysis membranes. Accordingly, below we review in detail protein adsorption and how this is influenced by the structure of these membranes. There are, however, differences in the techniques of forming the two types of membranes: collodion is prepared in a laboratory without applying pressure whereas the dialysis membranes are prepared in a factory under high pressure. As a result, although there are similarities in the physical structure of dialysis and collodion membranes, apparently the latter have larger pores on average. Loeb (1919-29) observed that collodion membranes, which had been treated with a 1% gelatin solution, showed different osmotic behavior from ordinary collodion membranes for a long time afterwards. Casein, egg albumin, blood albumin, or edestin produced the same effect while peptone prepared from egg albumin, or starch had no such effect. Hitchcock (1925) found that gelatin adsorption onto various membranes increased with their permeability to water: for membranes of any given permeability, the amount of the protein adhering reached a constant limiting value, independent of the protein concen- tration, in solutions having more than 15 gm/L of the protein. He and also Palmer (1952) found that the maximum amount of gelatin adsorption onto the membranes was observed at the isoelectric point of the protein: Dow (1955) observed the same pH effect using egg albumin. Hitchcock (1925) adsorbed gelatin by shaking collodion membranes in solutions having different concentrations at 570C. He also placed membranes in solutions of egg albumin without shaking to avoid mechanical denaturation which might have led to precipitation. More egg albumin adsorbed onto the membranes than did gelatin. Plots of the amount adsorbed onto the membranes vs protein concentration left in the solu- tion, were hyperbolic in shape, rising steeply from the origin and becoming horizontal as the membrane became saturated. When varying amounts of HCl or NaOH were used to adjust the pH of the gelatin or egg albumin solutions, the maximum adsorption occurred near but not precisely at the isoelectric points of the proteins; the maximum adsorption of egg al- bumin (IEP = 4.6) was observed at 5.0 by Hitchcock (1925) and in the range of 5.8 to 4.8 by Palmer (1952). The amount of gelatin adsorption (IEP = 4.7) also depends upon the acid 0'1 anion: the plots for CH3COOH and H3PO4 were quite close to that for HCl, while considerably more adsorbed with H2804 present. However, Hitchcock (1925) did not show whether these anions had similar effects on the maximum adsorbed at the iso- electric point of gelatin. Concerning this point, Elford (1955) reported that the maximum adsorption of oxyhaemoglobin was at a pH to the acid side of the isoelectric point (4.5- 5.6) when HCl and NaOH were used for adjusting the pH of the protein solutions. In general, the maximum amount of gelatin which adsorbs increases both with the thickness and the permeability of the membrane. In particular, the maximum which adsorbs to a given membrane is a linear function of the pore size. The straight-line relationship, however, could not be extrapolated to zero average pore size since no gelatin adsorbed onto the membranes of very small pore size. Hitchcock (1925) believes that the discontinuity lies at the point where the pores become too small to admit any gelatin at all. By comparing the water permeability through the membrane before and after gelatin adsorption he estimated that the adsorbed gelatin could have decreased the cross section of the pores by 76% or the average pore radius by 0.5. To calculate pore sizes, he applied Poiselle's law to the rate at which water flowed through the collodion membranes and made three simplifying assumptions about the nature of the pores: (i) they are cylindrical capillaries; (ii) they are perpendicular to the surface and run straight through the membrane; and (iii) the capillaries are circular in cross section. These assumptions were also used by Elford and Ferry (1955) for calculating the average pore diameter. Subsequent electronmicrographic studies (Hansmann and Pietsch, 1949; Helmcke, 1954; Beutel- Spacher, 1954) have invalidated the assumptions underlying- the above computations; i.e., collodion membranes have a spongelike structure of porous cells of various sizes (0.15-1 u across) in the interior and an irregular netlike structure on the surface with the average pore opening to the exterior being 0.04-0.05 u in diameter. While this does not invali— date the observation that adsorption is related to membrane permeability, it is not feasible to determine a unique func— tional dependence on absolute pore size. Since only a minute fraction of the adsorption appears to occur on the outer surface area of the collodion membranes, Elford (1955) attempted to obtain the internal pore space by measuring the increase of the weight and volume of dried membranes upon wetting. Unfortunately, when dried collodion membranes are hydrated, some of the water becomes bound since both intramicellar and intermicellar swelling occurs (Sollner and Carr, 1945). Thus, neither the weight nor volume increase nor their difference could be taken as a measure of the true pore space inside the membrane. So far, no one has been able to calculate the surface area in the pore network. MATERIALS AND METHODS The cellulose dialysis membranes (approximately 28 u thick in the wet state) were obtained from Scientific Glass Apparatus Co., Inc., Bloomfield, N. J. Pieces 2.5 x 2.2 cm (11.0 cm2 of actual surface area) were washed several times with distilled water to remove glycerin and other organic materials which might be present. One piece of membrane was put into each polystyrene vial containing 5 ml of a particu- lar concentration of protein plus the ions used to control pH. To make up the 5 uM solutions (assuming M. W. = 25,800) used in the adsorption, 7.1 mg of twice—crystallized, salt- free trypsin from Worthington Biochemical Corporation were dissolved in 100 ml of double-distilled water. The pH of the double—distilled water was 5.5 and the addition of the trypsin lowered it to 4.7. These solutions stored in a poly— styrene bottle at 40C showed no appreciable decrease of tryptic activity during a week. Adsorption was carried out normally at 40C: the ele- vated temperatures needed for studying the temperature dependence of adsorption were obtained in a thermostatically- controlled water bath. To observe the effect of ionic con- centration, trypsin was adsorbed from solutions containing 10-1 to 10"4 M phOSphate buffer. To study the effect of pH, this parameter was varied from 2 to 11 using H3P04, NaH2P04, Na2HPO4, and Na3PO4: the ionic strength was kept constant at 10'2 M. Changes in the protein content in the solution before and after adsorption were determined by measuring the optical density of the supernatant at 280 nm in a Beckman DU Spectro- photometer. The tryptic activity was determined by adding 0.2 ml of trypsin solution to 4 ml of benzoyl—arginine-ethyl ester (BAEE) solution and following the rate of hydrolysis by measuring the increase in OD at 255 nm (Worthington Bio- chemical Catalog No. 8, 1957). The pH 8 BABE solution was made by dissolving 6.64 g of Na2HP04 and 0.4 g of NaH2P04 in double—distilled water, adding 200 mg of BAEE and enough water to make 1 liter. The BAEE-urea solutions contained 8 M urea in addition (Augenstein, 1959). In order to reduce the errors which might result if drops of solution adhered to the membranes after the adsorp- tion process, the membranes with trypsin adsorbed were washed by dipping them several times into distilled water, since, as will be shown later, trypsin does not desorb into distilled water. The amount of trypsin on the membranes was studied by measuring the optical density at 220 nm in a Cary 15 spectrophotometer after they had been dried at room tempera- ture over night. The tryptic activity desorbed from the membranes was measured as follows: the desorption was usually carried out by stirring 4 ml of BAEE solution containing the membrane for 2 minutes, then the membrane was removed and an additional 4 ml of BAEE solution was quickly added and the rate of hydrolysis was measured at 255 nm. RESULTS AND DISCUSSION Selective Accumulation* and Desorption of Trypsin Both the rate and the amount of protein which accumu— lates on the membrane depends upon its initial concentration in the solution: the higher the initial concentration the greater the accumulation. Using a constant amount of dialysis membrane (one piece in each vial), the accumulation from trypsin concentrations of 1, 5, 5, and 10 uM in distilled water was measured at 40C. The increase in the OD of mem- branes onto which trypsin accumulated was plotted in Fig. 1 as a function of the trypSin concentration in the solution. Other factors which affect the accumulation of trypsin onto the dialysis membranes are shown in Fig. 2. The maximum value of 0.128 for the 5 uM solution shown in Fig. 2 would correSpond to a uniform layer of protein 200 A thick on each side of the membrane--with 72 hour adsorption in the same trypsin solution the corresponding value increases to 1500 A. The plots of accumulation with time can be resolved as the sum of two exponential components. The time constants (t37) derived from this analysis indicate that more than 85% of *The term "accumulation" is used instead of the more common "adsorption," since the amount of protein which is measured on a membrane at any given time is the net resutling from both adsorption and desorption (see next section). 10 11 Figure 1: The effect of trypsin concentration in solution on the rate of trypsin accumulation onto dialysis mem- branes. The amount of protein adsorbed was determined by the increase of OD (220 nm) following adsorption for the three lengths of time shown. INCREASE IN OPTICAL DENSITY (220 mp.) 'o 0.8 0.6 0.4 0.2 12 l I 1 l l l l l l J 24 hr 2 4 6 8 l0 TRYPSIN CONCENTRATION (AM) FIGURE 1 15 .8: CNN um no CH wmmmnocfl map Eouw COCHEumumU mm3 mmcmunfimfi mnu co :amuoum mo ucsoam mze .ucmucoo Hmmmsn Cam NEH» mo coauocsm m mm OGMHQEOE mHmMHMHU ouso cammmuu mo coaumHDEDoom One "N musmflm 14 N HMDme 62:): m2; ZO_._.Huom mo coaumuommw Gnu so mCOHuDHOm :mcflnmm3= ucmummwflv mo uommmm OSB "w musmflm 19 w HMDme 62:): m2; ZO_._.dmOmwo m n... _ me o _ a _ _ III—”HI AM. I I «wanna mfizamoza m In 2. zoiamommo .. r 554 mzqmmzmz zo oz_z_<_2mm I O I I O 8:3 20:38 mmqm z. ommmommo IN RIG-II T 4 \ cubs: omI_.:._.m_o Z_ 02.Im<>> OZ _ _ _ _ /x ‘ No.0 . v0.0 $0.0 All/\IIOV OlldAHJ. BALLV‘IBH 20 droplets which adhere after removal from the solution, since additional washing in distilled water does not reduce further the activity which can be recovered. This constancy indi- cates that adsorbed trypsin does not desorb from the mem- brane into distilled water (pH 5.5). There is, however, rapid desorption in pH 8 phosphate buffer solution which is essentially complete in 1.5 minutes. Thus, in all of the present experiments we have placed membranes with trypsin adsorbed on them into 4 ml of BAEE solution for at least 2 minutes in order to measure the amount of active trypsin which can be recovered. To characterize the relative proportions of the various trypsin conformers accumulated onto the membrane, the amount of tryptic activity desorbed was compared with the decrease in total protein on the membranes following both short (2 minutes) and long (24 hours) adsorption times. Specifically, each membrane with protein adsorbed onto it was placed in 4 ml of BAEE solution and stirred for a specific length of time varying from 15 sec to 128 min before measuring the OD of the trypsin still remaining on the membrane. To determine the enzymic activity desorbed, another 4 ml of BAEE solution was added to the trypsin-BAEE mixture. For those films formed in 2 minutes of adsorption, the desorbed activity reached its maximum when only 65% of the total trypsin had been desorbed from the membrane. However, from the membranes kept in trypsin solutions for 24 hours, the activity 21 recoverable was related directly to the decrease of the film OD on the membrane (i.e., activity and total protein were desorbed at the same rate). In fact, some residual tryptic activity could still be recovered even from membranes which had been washed for 128 minutes with the result that 98% of the trypsin had already been desorbed. These differences could reflect either differential adsorption and/or desorption or else denaturation at the interface. Thus, it was necessary to investigate the stabil- ity of trypsin accumulated on the membrane. Accordingly, after 2 minutes of adsorption from a solution containing trypsin in distilled water, one group of membranes were air- dried at room temperature (24°C) and each membrane of another group was put in a vial containing 5 ml of distilled water and kept at the same temperature. Although the amount of protein present on the membranes did not decrease in either situation, the amount of trypsin which could be desorbed in an active form decreased rapidly with time. The data in Fig. 5 show that in the air-dried samples enzymic activity is lost initially faster than on the immersed membranes. However, after an initial rapid drOp to 60%, the activity recoverable from the membrane is fairly stable and even after one week 40% of the original activity can still be recovered. By contrast, the membranes kept in distilled water show a continuous decrease in the amount of activity recoverable from the membranes so that all activity is lost in one week. 22 Figure 5: The stability of tryptic activity adsorbed on the membranes kept dry or in distilled water (5 ml for each piece of membrane) at 240C. OD measurements on the supernatant show no appreciable desorption of the trypsin from the membranes into the distilled water. FRACTION 0F ACTIVITY REMAINING Q5 .0 005 25 1 I WET MEMBRANE I l I I I 5 l0 I5 20 PRESERVATION TIME (HR) FIGURE 5 25 24 Since, as noted above, active molecules accumulate faster than inactive molecules, we can conclude that the denatura- tion of trypsin following its adsorption is facilitated by the presence of water. Other Factors Affiecting Accumulation Since we are interested in the effect of adsorption on the enzymic activity of trypsin, in the rest of the experi- ments attention was focused on the change of tryptic activity with various experimental conditions. Further, to minimize the effect of diffusion of the trypsin molecules through the solvent to the membrane surface, accumulation and desorption .were carried out with stirring so as to continuously renew the membrane/water interface. The data in Fig. 2 show that the concentration of salt or buffer ions affects greatly the amount of trypsin which accumulates: the lower the ionic strength the greater the accumulation-~e.g., with no buffer (i.e., only double dis- tilled water) as much as 15% of the trypsin in 5 ml of a 5 uM solution will accumulate onto 11 cm2 of membrane surface in.1 hour: even with10‘2 M phosphate buffer present 5% will accumulate in this time and approximately 10% when a maximum is achieved after 4 days. The results of another set of experiments on the effect of salt concentration on accumula- tion of active trypsin at pH 7 are shown in Fig. 6. (Since it was difficult to accurately resolve the initial portions 25 .2 HIGH um coaumuomom oc mHHmoHuomum mH mumnu cam E OIOH m>onm mommmuusfl Sumcmuum UHGOH c033 mawflmmu mum> mommmuoww cofiumuompm mo mpmu was .coaumuomow mo mnumcma mDOHum> umumm mmcmunfimfi map m0 mcaanB muscHE N Eva? coausaom mm¢m CH COHO>OUOH mufl>fluum Uflumhuu mm Consumma mm3 coflumuomvm mo UGDOEM mss .ummmsfl muwnmmosm h mm map m0 numcmuum UHCOH map so coaumasfisoom mo mocmwcmmwv one .0 musmflm 26 m mmDOHm mwk_u_ any mwumDmmEQIn—mOIQ m0 Ihozmmhm U_ZO_ ~-o_ nuo_ «-o. :_dq4fi «A.~T__ A NIO_ H I V l 3 I O 4.. I V O I n I W I n I _I T To. W. m N ] 2" I...) I I I w .m. 1 1 “L .p. r _ _ _ E. _ _ _ _ _ 27 of the plots of log fraction of activity still to be absorbed vs time--they represent only about 7-10% of the total activity accumulated--the accumulation rates in Fig. 6 were obtained only from the slowest component of the semilog plots. The rate is taken as the reciprocal of the 1/e time for these semilog plots.) The results indicate that accumu- lation decreases rapidly when ionic strength is larger than 10'3 M and at 10‘1 M practically no active protein adheres to the membrane. .Salt concentration also affects greatly the rate of desorption. Fig. 7 shows clearly that the rate of desorption increases rapidly for ionic strength greater than 10‘3 M. Not only the concentration of ions but also pH affects markedly the amount of active material which accumulates after a given time. The data in Fig. 8 illustrate the effect of pH on the rate of accumulation of active trypsin while holding the ionic strength of phosphate buffer constant at 10‘2 M. Two peaks appear in the rate of accumulation on the membrane at pH's of 6.5 and 9.5: for pH less than 4.5 or greater than 10.5 protein accumulates on the membranes so slowly that accurate measurements could not be made. The pH dependence of desorption, shown in Fig. 8 has two minima corresponding to the maxima at 6.5 and 9.5 for accumulation of active molecules. At both pH extremes and in the region of pH 7.5-8.5 desorption occurs very rapidly. 28 14".‘l IIJIII . .coflumH IDEDoom mo um£u ou coauocsm omuo>ca am we o>u50 coflumuomoo one .2 oIOH ou mommonoca mCOHusHOm oSu mo Sumcouum canoe soak mommouocfl Maoflmmu soaumuomoa .AQOHuDHom cammwnu 21 m Eonm coaumuomom mo moussflfi N mCHHDU UoEHOm ouoB mEHHmV mufl>fluom mo momma cm mp Cousmmofi coaumHOmoC mo oumu onu so nommsn oumnmmonm on“ NO numcouum UHCOH onu mo uoowmo one up ousmflm 29 IO" L - IIIIIIII r IIITTTTT T _ —I N I :o I -I I0 I -9 _I -I Q’ L I l lllLlIIll IJIIIIILI 2 - Nw I'O I I I o o o (“H“)(‘QI I ] NOIIdHOSEICI :IO 31w ) MOLES IONIC STRENGTH OF PHOSPHATE BUFFER ( LITER FIGURE 7 50 .COHuCHOm Cflmmwuu 21 m m Eoum COHpmuomUm ouCCHE N >9 CoEHOM mEHHm Co Couswmoe mm3 Coaumuomoa .Coflusaom mmdm m CH mCHHHHum onsCHE N an oCmnnEoE onu Eoum hauCoCUomnsm Couo>ooou on UHDOU nUHnB mufl>fluum Uflumhuu onp mm Cousmmoa moB Coaumasfisoom mo quoEo onB .muommsn oumnmmonm S Nloa mCHmD CoCHmpCHmE mmB mm one .COHumuomoU UCm COHumHCEDUUm mo oumn onu Co mm mo uoommo onB "m ousmam 51 ACCUMULATION \ DESORPTION ‘0. o 1 ID 0 0.4r- I I ['0 N O O I'UIw) L9: [ I ] NOIIdHOSBO HO NOIIV‘IFIWOOOV so 31w ' I2 l0 FIGURE 8 52 The inverse nature of the dependence of accumulation and desorption on pH (Fig. 8) or salt concentration (Figs. 6 and 7) indicates that the accumulation of trypsin molecules on the membrane is determined by the true adsorption offset by appreciable desorption: i.e., Accumulation = Adsorption - Desorption. The accumulation is strongly dependent on pH: it is smallest where desorption is greatest at highly acidic or alkaline pH's and in the region of pH 8 or at high salt concentrations. Unlike previous reports on other proteins adsorbed onto collodion membrane, accumulation of trypsin is not maximum at the isoelectric point (10.8). .Rather it must be determined in a complex way by the pK values of groups in both the pro- tein and the dialysis membrane. With the collodion membranes used earlier the maximum adsorption was observed at the iso- electric point of the proteins used, and it was concluded that this reflected aggregation of the protein molecules in the pore Spaces. But, unfortunately they used only proteins whose isoelectric points were in the neutral pH region. »According to Sollner.§§._l. (1941). the collodion membrane is acidic due to the presence of COOH groups and sulfate groups. In dialysis membranes, the most prevalent free ionic groups are OH (Valko, 1957) and the effect of pH and the concentration of salt may be to change the environment of these hydroxyl groups. Alternatively, the added ions may compete with trypsin molecules for binding to the adsorption sites of the membrane. 55 Importance of Porous Structure In order to observe the effect of temperature on the amount of protein accumulation, the samples were immersed in trypsin solutions maintained at various temperatures from 40 to 570C. The data in Fig. 9 show that the higher the temperatures the greater the rate of accumulation. This means that adsorption must be enhanced more by an increase in T than is desorption. The temperature dependence is affected greatly by the state of the membranes. In Fig. 9, it can be seen that the membranes which are maintained at a given temperature before and during adsorption have a smaller temperature dependence for the accumulation of active molecules than those kept at 40C prior to adsorption: the SIOpes of the two straight lines correspond to thermal coefficients of 2 kcal/mole and 8 kcal/mole respectively. The above experiments on the temperature dependence of accumulation were carried out with trypsin dissolved in double—distilled water (pH 5.5) where desorption is minimal. Fig. 10 also shows the temperature dependence of accumulation from a pH 5.5 solution having 10‘2 M phosphate buffer (the membranes were maintained at the various temperatures prior to and during the adsorption). Although the rates of accumulation are much slower with this salt concentration, still the thermal coefficient is 5 kcal/mole. 54 Figure 9: The effect of temperature on trypsin accumu- lation (measured by BAEE assay) as a function of the temperature at which the membranes are kept prior to the adsorption. I I37 (min) RATE OF ACCUMULATION I: 55 3 T I T T 2 P r \x \x I P .. 0.9 I- - 0.8 P - 0.7 I- .. 0.6- .. 0.5 P "I 0.4L x MEMBRANES KEPT AT 4°C PRIOR -I TO ABSORPTION THERMAL COEFFICIENT AT 8 KCOI/ 0.5 r- mole" O MEMBRANES KEPT AT EACH TEMP. O 2 _ PRIOR TO AND DURING AOSORPTION q ' THERMAL COEFFICIENT AT 3 Kcol/ mole O.I J I I J 3.2 3.3 3.4 3.5 3.6 3.7 l000 T (K°) FIGURE 9 56 Figure 10: The effect of temperature on the accumulation or desorption of trypsin in pH 5.5 solutions having a 10-2 M phOSphate buffer: films for desorption were prepared by 2-minute adsorption from 5 uM trypsin in double-distilled water (amounts of trypsin were measured by BAEE assay). 57 c-q A q — —I q A q — a Hagan—n. _ N nlu _I A I L '1 U M U C C I A i N T m I T: /P R O S E 1. O I _____r b L PV—p—.L_ p 2 $376.34. 3. 2. I.9.8.7.6.54. 3 2 I 00000 O O 0 000000 0 O 0 O «EEK». n Q zofiamoflo mo ZOF<432300< mo m._.