Protein Expression and Purification 16, 261–275 (1999) Article ID prep.1999.1050, available online at http://www.idealibrary.com on
Effect of Accessible Immobilized NAD 1 Concentration on the Bioaffinity Chromatographic Behavior of NAD 1Dependent Dehydrogenases Using the Kinetic Locking-on Strategy Patricia Mulcahy, 1 Martina O’Flaherty, Mary McMahon, and Laura Oakey Department of Applied Biology and Chemistry, Institute of Technology Carlow, Kilkenny Road, Carlow, Ireland
Received November 17, 1998, and in revised form February 8, 1999
In preparation for studies aimed at establishing the relationship between immobilized NAD 1 concentration and the concentration of soluble locking-on ligand required to promote biospecific adsorption of NAD 1 -dependent dehydrogenases to immobilized NAD 1 derivatives (the “locking-on” strategy), two approaches were evaluated for varying substitution levels: (i) suitable dilution of the affinity matrix with unsubstituted Sepharose 4B and (ii) direct coupling of the required ligand concentration to the inert matrix. The latter approach was found to be the preferable strategy for evaluation of the locking-on tactic because it produced a more homogeneous distribution of immobilized NAD 1 concentration. Affinity chromatographic studies using S 6-linked NAD 1 derivatives synthesized to various substitution levels showed that the total accessible immobilized NAD 1 concentration has a direct effect on the locking-on behavior of pyridine nucleotide-dependent dehydrogenases. The one-chromatographic-step bioaffinity purification of L-lactate dehydrogenase (L-LDH, EC 1.1.1.27) from bovine heart illustrates the potential of the locking-on strategy for protein purification applications. © 1999 Academic Press Key Words: bioaffinity chromatography; bovine heart L-lactate dehydrogenase; yeast alcohol dehydrogenase; immobilized NAD 1 derivatives.
Rigorous analysis of immobilized cofactor derivatives, particularly in the 1970s, revealed some of the complexities associated with their application to the purification of nucleotide-dependent dehydrogenases (1–19). This, together with the rather disappointing purifications achieved with selected systems, contributed to the decline in interest in general ligand bioaf1
To whom correspondence and reprint requests should be addressed. Fax: 0503 70500. E-mail:
[email protected]. 1046-5928/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
finity chromatography in more recent years. This is reflected in the many reviews and textbooks on affinity chromatography which focus primarily on the less specific “pseudo-affinity” group specific adsorbents such as immobilized lectins and dyes. However, the kinetic “locking-on” and auxiliary tactics (20 –22) have the potential to overcome many of the difficulties associated with general ligand bioaffinity chromatography of NAD 1 -dependent dehydrogenases. The locking-on strategy uses soluble analogues of the enzymes specific substrate to produce a reinforcement of biospecific adsorption to immobilized NAD 1 derivatives which is sufficient to effect adsorptive selection of one enzyme from a group of enzymes (20 –22). Application of this strategy to the purification of NAD 1-dependent dehydrogenases from crude extracts has proven that use of the locking-on tactic yields bioaffinity systems capable of producing one-chromatographic-step purifications to homogeneity with yields approaching 100% (20 –23). Further development of the locking-on strategy has revealed a variety of parameters that require attention before successful application to enzyme purification. Most of these relate to the nature of the immobilized NAD 1 derivative, particularly the accessible immobilized NAD 1 concentration. While some early studies have been carried out on the effect of varying the immobilized cofactor substitution level of affinity matrices (6 – 8,24,25), these are of limited use to the present study because (i) these studies focused on immobilized AMP derivatives and/or (ii) nonspecific methods were employed to desorb bound enzymes and/or (iii) total immobilized cofactor concentration was determined rather than accessible immobilized cofactor concentration. Furthermore, substitution levels of the immobilized ligand were varied by dilution of the affinity matrix with underivatized Sepharose. Indeed, it has been suggested that matrices where the 261
262
MULCAHY ET AL.
ligand concentration has been lowered by dilution of the gel with underivatized Sepharose have the advantage over matrices that have been chemically synthesized to lower substitution levels (26). Both approaches were investigated in the present study and some experimental evidence was obtained which suggests that varying the immobilized ligand concentration via direct coupling methods is the preferable strategy. This approach achieved a more homogeneous distribution of immobilized NAD 1 throughout the matrix and this had a beneficial effect on the chromatographic behavior of certain “test” dehydrogenases chromatographed using the kinetic locking-on strategy. However, the less laborious preparation of immobilized NAD 1 derivatives with varying substitution levels via dilution with underivatized Sepharose-4B was carried out first in order to establish if there was a relationship between immobilized cofactor concentration and the concentration of locking-on ligand required to promote biospecific adsorption to the affinity matrix.
Enzyme and Protein Assays L-lactate dehydrogenase (L-LDH, EC 1.1.1.27). activity was measured spectrophotometrically measuring the decrease in absorbance at 340 nm. Assays were carried out in 50 mM potassium phosphate buffer (pH 7.4) containing 0.5 mM pyruvate and 0.2 mM NADH as substrates. The assay temperature was 30°C. Alcohol dehydrogenase activity from yeast (YADH, EC 1.1.1.1) was generally monitored spectrophotometrically at 30°C by measuring the increase in absorbance at 340 nm. Assays were usually carried out in 30 mM sodium pyrophosphate buffer (pH 8.8) containing 0.8 mM NAD 1 and 0.12 M ethanol. Alternatively, alcohol dehydrogenase activity was assayed in the opposite direction (measuring the decrease in absorbance at 340 nm) using a cocktail of 0.2 mM NADH and 3 mM acetaldehyde in 50 mM potassium phosphate buffer (pH 7.4). Protein concentration was routinely determined using the method of Lowry (27). Bovine serum albumin was used as the protein standard.
MATERIALS AND METHODS
From Sigma Chemical Company (Poole, Dorset, England)
Synthesis and Enzymatic Analysis of Immobilized Cofactor Derivatives
Cyanogen bromide-activated Sepharose 4B, Sepharose 4B, b-nicotinamide adenine dinucleotide (reduced form), b-nicotinamide adenine dinucleotide, 1,3 diamino-2-hydroxypropane, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide, oxalic acid (dipotassium salt), N-hydroxysuccinimide, bromoacetic acid, dicyclohexyl carbodiimide, p-nitrobenzoyl azide, diethyl formamide, 6-mercaptopurine riboside-59-phosphate, NMN, 59AMP, urea, Fiske and SubbaRow reducing agent, DL-dithiothreitol, alcohol dehydrogenase from yeast (EC 1.1.1.1); L-lactate dehydrogenase from bovine heart (Type 1X ; LDH1 (H 4) isoenzyme).
89-Azo-linked immobilized NAD 1, S 6-linked immobilized AMP and S 6-linked immobilized NAD 1 derivatives were synthesized using 1,3 diaminopropanol as the spacer arm compound as described in Ref. (22). The synthesis of S 6-linked immobilized AMP and NAD 1 affinity matrices with varying substitution levels was achieved by varying the concentration of cyanogen bromide used to activate the Sepharose-4B (22). After coupling of the spacer arm compound to Sepharose-4B, the amount of spacer arm coupled was determined using the TNBS test (see below). Accessibility of the immobilized NAD 1 was determined as outlined in Ref. (22) using alcohol and aldehyde dehydrogenase. Certain of the S 6-linked immobilized NAD 1 derivatives were analyzed for the amount of accessible immobilized NAD 1 by also monitoring the enzymatic reduction of immobilized NAD 1 to NADH by lactate dehydrogenase. The progress of the reaction was monitored spectrophotometrically following the increase in absorption at 340 nm. The moist gels were first washed with 0.1 M potassium phosphate buffer (pH 7.4). 300 mg of the S 6-linked derivative was suspended in 3 ml (final volume) of the same buffer and incubated in a quartz cuvette at 30°C in the presence of 10 mM L-lactate and 40 units of L-lactate dehydrogenase from bovine heart (Type 1X; LDH1 (H 4 ) isoenzyme). The reference sample contained the same components, but replacing the 300 mg of S 6 -linked immobilized NAD 1 with 300 mg of S 6 -linked immobilized AMP derivative.
From Riedel de Haen (Seeize, Germany) Potassium dihydrogen phosphate, di-potassium hydrogen phosphate, sodium nitrite, dioxane, pyridine, dimethylformamide, sodium borate, and hydrogen peroxide. From Aldrich Chemical Company (Gillingham, Dorset, England) Sodium dithionite (sodium hydrosulfite), isobutyramide, cyanogen bromide, N-hydroxysuccinimide, and succinic anhydride. From Merk Schuchart (Darmstadt, FR, Germany) N-hydroxysuccinimide and NN9-dicyclohexylcarbodiimide.
ACCESSIBLE IMMOBILIZED @NAD 1 # AND THE KINETIC LOCKING-ON STRATEGY
Chemical Analysis of Affinity Matrices TNBS test. The qualitative TNBS color test was used to follow the course of substitution of Sepharose4B. A small quantity of derivatized gel (0.2– 0.5 g wet weight of matrix in distilled water) was added to 1 ml saturated sodium borate followed by three drops of a 3% aqueous solution of sodium 2,4,6-trinitrobenzenesulphonate (TNBS). The color reaction of the gel beads was allowed to develop for 2 h at room temperature. A rough estimation of the degree of substitution of the gels could be judged from the relative color intensity of the TNBS-reacted gel after washing with distilled water. The following color products are formed with various derivatives: unsubstituted Sepharose-4B, pale yellow; carboxylic acid and bromoacetyl derivatives, yellow; derivatives containing primary aliphatic amines, orange; derivatives containing primary aromatic amines, red-orange. A quantitative TNBS test was used in certain instances (e.g., synthesis of S 6linked immobilized AMP and NAD 1 derivatives of varying substitution levels) as a quantitative test to complement the results obtained using the qualitative TNBS procedure outlined above. The procedure followed was that described by Fallia and Santi (28). Phosphate analysis. Determination of the substitution levels of immobilized NAD 1 and AMP derivatives was based upon phosphorus analysis following chemical digestion of the affinity matrices. The method of Mosbach (29) was routinely used for the determination of substitution levels of immobilized nucleotide derivatives. Affinity Chromatography Analytical affinity chromatographic procedures. Chromatography was carried out at room temperature in miniature columns (1.7-cm internal diameter; 1-ml bed volume) with a hydrostatic head pressure adjusted to give a flow rate of approximately 1 column vol/3 min. The “breakthrough” elution volumes were determined by applying 4 mg glucose dissolved in 0.5 ml of irrigating buffer, and monitoring its presence in the effluent using dinitrosalicylic acid reagent. All columns were equilibrated with the appropriate irrigating buffer prior to application of the enzyme sample. Extraction, Concentration, and Affinity Purification of L-LDH from Bovine Heart Fresh bovine heart tissue was obtained from a local abattoir, transported on ice, washed thoroughly, and stored frozen at 220°C until required. Bovine heart extracts were prepared by homogenization with 4 vol of extracting buffer (50 mM potassium phosphate buffer, pH 7.4, containing 1 mM EDTA) per gram of cells in a Waring Blender. The temperature at this point and during all subsequent manipulations was maintained
263
at 0 to 4°C. The homogenate was centrifuged at 25,000g for 30 min. The clarified extract was then subjected to an ammonium sulfate fractionation step (20 to 80% saturated ammonium sulfate cut). Centrifugation was carried out at 25,000g for 30 min at each stage. The pellet was dissolved in 50 mM potassium phosphate buffer, pH 7.4, containing 0.5 M KCl. An S 6-linked immobilized NAD 1 derivative was synthesized using 1,3-diaminopropanol as the spacer arm compound and using Sepharose-4B activated with CNBr (50 mg CNBr/g wet wt., 0.3 mmol total NAD 1/g wet wt. and 0.2 mmol accessible NAD 1/g wet wt; see Ref. 22). Purification was achieved using the locking-on tactic in conjunction with the “stripping ligand” approach (22). A 5-ml packed bed column of the S 6-linked immobilized NAD 1 derivative was equilibrated with 50 mM potassium phosphate buffer, pH 7.4, containing 0.5 M KCl and 5 mM oxalate (starting irrigant). Oxalate was added to the crude bovine heart extract (the 20 to 80% saturated ammonium sulfate cut) to a final concentration of 5 mM. After application of this sample to the S 6-linked immobilized NAD 1 derivative, the column was washed with four column volumes of starting irrigant, followed by three column volumes of starting irrigant containing the “stripping ligand” (10 mM 59AMP), followed by a further two column volumes of starting irrigant. Subsequent omission of oxalate from the starting irrigant resulted in autoelution of the LLDH from the matrix. Fractions were analyzed for L-LDH activity and protein. Active fractions were pooled, dialyzed into distilled water, and freeze dried in preparation for SDS–polyacrylamide gel electrophoresis. SDS–Polyacrylamide Gel Electrophoresis SDS–PAGE was carried out according to Laemmli (30) using a separating gel of 12.5% acrylamide. The following protein markers were used for molecular weight calibration of the SDS-gels (SDS LMW proteins from Pharmacia): a-lactalbumin (14 kDa), trypsin inhibitor (20.1 kDa), carbonic anhydrase (30 kDa), ovalbumin (43 kDa), BSA (67 kDa), and phosphorylase B (94 kDa). Gels were fixed in 12% w/v TCA for 30 min and then stained for 2 h in amido black (0.2% w/v in methanol-acetic acid-water, 4:1:5). Stained gels were cleared of residual dye by leaching in the same solvent mixture, minus the dye. RESULTS AND DISCUSSION
Chromatography of Bovine Heart L-LDH on an 89AzoLinked Immobilized NAD 1 Derivative Using the Locking-on Tactic Previous studies on the development of biospecific affinity systems for the purification of NAD 1-dependent dehydrogenases have focused on various immobi-
264
MULCAHY ET AL.
FIG. 1. Structures of the affinity chromatographic matrices used in the present study.
lized NAD 1 derivatives (e.g. Fig. 1). However, while these immobilized general ligands sometimes adsorb their complementary multisubstrate enzymes, they generally lack specificity towards any one individual enzyme. When “groups” of enzyme are adsorbed by such immobilized ligands, successful purification of a single enzymatic activity depends upon the degree of specificity that can be achieved during subsequent elution. These specific elution techniques generally involve competitive elution with the oxidized or reduced cofactor, either alone or in combination with the enzyme-specific substrate or substrate analogue. In this mode of use, immobilized cofactor derivatives have the following disadvantages: (a) While immobilized NAD 1 facilitate the adsorption of certain complementary dehydrogenases from crude cellular extracts, binding of a range of other enzymes to the insolubilized cofactor also occurs. This decreases the effectiveness of the affinity adsorbent by directly interacting with the immobilized nucleotide and the resultant enzyme preparations are not homogeneous. (b) It is not always easy to predict the best type of
elution method to achieve optimum resolution of the dehydrogenases of interest. (c) Some dehydrogenases exhibit very weak or no affinity for the immobilized cofactor and elute in the protein break-through peak. O’Carra et al. (20) and Mulcahy et al. (21) reported biospecific affinity chromatographic systems for the rapid purification of bovine liver glutamate dehydrogenase and molluscan octopine dehydrogenase to homogeneity from crude cellular extracts. These systems utilized 89-azo-linked immobilized NAD 1 derivatives (Fig. 1) in conjunction with soluble specific substrate analogues, or the actual specific substrates, to promote selective adsorption (the locking-on strategy). The potential advantages of the locking-on strategy for the purification of NAD 1-dependent dehydrogenases include: (a) Purification of one or more dehydrogenases to homogeneity in a single bioaffinity step (using one or more locking-on ligands). (b) Enzyme yields approaching 100%. (c) Purification to homogeneity within 2 h. (d) Application of the same biospecific affinity system
ACCESSIBLE IMMOBILIZED @NAD 1 # AND THE KINETIC LOCKING-ON STRATEGY
with minimal refinements to the same enzyme activity from a wide range of sources. However, attempts to further develop and apply the locking-on strategy to the biospecific affinity chromatographic purification of other NAD 1 and NADP 1-dependent dehydrogenases revealed some anomalous chromatographic behavior and certain unexplained phenomenon. In particular, some of the results reported by O’Carra et al. (20) proved difficult to reproduce exactly under apparently identical experimental conditions. Since one possible explanation for this is differences in immobilized NAD 1 concentrations, an examination of the relationship between immobilized NAD 1 concentration and the behavior of enzymes chromatographed using the locking-on tactic, seemed appropriate. The “test” enzyme chosen for these studies was bovine heart L-LDH. Lactate dehydrogenase is an enzyme with a compulsory order of substrate addition with NAD 1 as the leading ligand. Oxalate is a competitive analogue of the second substrate, lactate. O’Carra and coworkers (20) showed that addition of high concentrations of oxalate to the irrigating buffer caused strong reinforcement of the binding of the enzyme to an 89-azo-linked immobilized NAD 1 derivative. That this reinforcement is highly specific was shown by the fact that no similar reinforcement was observed when oxalate was replaced by other dicarboxylate ions or when the immobilized NAD 1 was replaced by a closely analogous NADP 1 derivative (20). Concentrations of oxalate initially used to achieve this locking-on effect were of the order of 100 mM oxalate. However, more recent studies showed that a progressive decrease of the oxalate concentration from 100 mM produced no apparent weakening of the locking-effect until submillimolar concentrations were reached (0.5 mM oxalate produced strong enhancement of the adsorption, 0.2 mM promoted weaker binding, and 0.05 mM oxalate has a negligible effect; see Ref. 20). Similar results were obtained in the present study using an 89-azo-linked immobilized NAD 1 derivative synthesized as described in the experimental section using a commercial preparation of CNBr-activated Sepharose-4B (Fig. 2B). As can be seen in Fig. 2B, 2 mM oxalate produces strong enhancement of the adsorption. However, 0.5 mM oxalate promotes weaker binding and 0.05 mM oxalate has a negligible effect (not shown). The 89-azo-linked immobilized NAD 1 derivative used for these studies was diluted twofold with underivatized Sepharose-4B and the concentration of oxalate required to lock-on L-LDH was reinvestigated. These results are presented in Fig. 2D. That a higher concentration of locking-on ligand is required to promote strong adsorption of L-LDH on the lower substituted matrix is evident from Fig. 2D. Furthermore, that the weaker adsorption of L-LDH to the diluted matrix in the presence of 2 mM oxalate (Fig. 2D) is not
265
FIG. 2. The effect of immobilized NAD 1 concentration on the “locking-on” of bovine heart L-LDH on an 89-azo-linked immobilized NAD 1 derivative. The 89-azo-linked immobilized NAD 1 was synthesized using commercial CNBr-activated Sepharose 4B (Sigma) and used neat (A and B) or diluted twofold with underivatized Sepharose 4B (C and D). The irrigant was 50 mM potassium phosphate buffer (pH 7.4) containing 0.5 M KCl. Chromatography was performed at room temperature.
due to overloading of the column is indicated by the fact that 10 mM oxalate promotes strong enhancement of the adsorption on the diluted matrix (not shown). This effect can be explained by the fact that in a bisubstrate reaction the concentration of one substrate usually affects the K m of the other depending upon the
266
MULCAHY ET AL.
TABLE 1 1
The Effect of NAD Concentration on Bovine Heart LDH K m and K i Values for Lactate and Oxalate, Respectively [NAD 1] mM
K m for Lactate (mM)
K i for oxalate (mM)
Type of inhibition
0.5 2.0 5.0
1.9 1.4 0.5
31.0 12.0 6.7
Competitive Competitive Competitive
Note. All assays were carried out using 50 mM potassium phosphate buffer (pH 7.4) at 30°C.
type of reaction mechanism. For many sequential mechanisms the higher the concentration of the second or “following” substrate, the smaller is the K m for the first or “leading” substrate (and vice versa). As lactate dehydrogenase has a compulsory order sequential mechanism of substrate addition, increasing the concentration of the leading substrate (immobilized NAD 1) produces a decrease in K m for the following substrate (lactate for LDH). Since the locking-on ligand used (oxalate) is a competitive inhibitor of the following substrate (lactate), this increase in affinity for the specific substrate also results in an increase in affinity for the substrate analogue (Table 1). It has been established with other NAD 1-dependent dehydrogenases (for example, YADH, see below) that when the immobilized NAD 1 concentration is kept constant, the K i value of the locking-on ligand determines the effective concentration of soluble locking-on ligand: the percentage enzyme locked-on gradually decreases as the K i value of the competitive inhibitor/locking-on ligand increases. Thus decreasing the concentration of immobilized NAD 1 (leading substrate) by dilution results in a need for a higher concentration of locking-on ligand to promote adsorption since the lower concentration of immobilized ligand translates into a higher K i value for the locking-on ligand. Characteristics of S 6-Linked Immobilized NAD 1 Derivatives Synthesized to Varying Substitution Levels Although the preliminary investigations discussed in the previous section suggested that the substitution level of immobilized NAD 1 derivatives influences the concentration of locking-on ligand required to promote strong reinforcement of binding of NAD 1-dependent dehydrogenases, it could be argued that the diluted 89-azo-linked immobilized NAD 1-underivatized Sepharose-4B mixture is not homogeneous with respect to NAD 1 distribution throughout the matrix. Since this is likely to affect the chromatographic characteristics of enzymes, a more analytical approach was adopted. Four preparations of S 6 -linked immobilized NAD 1 were synthesized to give varying substitution levels as
described in the experimental section. It was found that decreasing substitution levels could be achieved by decreasing the concentration of CNBr used to activate Sepharose-4B (see Table 2). 1,3-diaminopropanol was used as the spacer arm compound for all of these preparations in order to reduce the possibility of nonbiospecific interactions, which can occur when hydrophobic spacer arms are used. The approach used for synthesis of S 6-linked immobilized NAD 1 derivatives was the solid phase modular approach where the spacer arm was first coupled to Sepharose-4B, followed by the synthesis of S 6-linked immobilized AMP, followed by the condensation of this AMP with NMN to produce S 6 -linked immobilized NAD 1. It is clear from Table 2 that the commercial preparation of CNBr-activated Sepharose-4B resulted in the attachment of the highest concentration of AMP (4.3 mmol AMP/g wet wt), while the immobilized ligand concentrations achieved when the inert matrix was activated with CNBr in the laboratory were significantly lower (1.1 to 0.5 mmol AMP/g wet wt.). After condensation of the S 6 -linked immobilized AMP with NMN, the change in phosphate concentration suggested that the four preparations contained an immobilized NAD 1 concentration ranging from 1.9 mmol to 0.3 mmol/g wet wt. matrix (Table 2). This indicates that not all of the S 6-linked immobilized AMP was converted to S 6-linked immobilized NAD 1 (Table 2). The percentage conversion rate of AMP to NAD 1 was between 45 and 69%. It thus seems that 69% conversion is the maximum conversion rate achievable in the presence of saturating concentrations of NMN. Obviously the concentration of NMN was not at saturating level in the case of the highest substituted matrix as only a 45% conversion of AMP to NAD 1 was achieved. The presence of S 6-linked AMP did not appear to have an impact on the purification of NAD 1TABLE 2 Properties of the S 6-Linked Immobilized NAD 1 Derivatives Synthesized to Give Varying Substitution Levels (1,3Diaminopropanol Was Used As the Spacer Arm Compound and Decreasing Substitution Levels Were Achieved by Decreasing the Concentration of CNBr Used to Activate Sepharose-4B)
Preparation
mmol NAD 1 (per g. wet wt.)
mmol AMP (per g wet wt.)
Percentage conversion of AMP to NAD 1
A B C D
1.9 0.7 0.4 0.3
2.4 0.4 0.2 0.2
44.9 60.9 69.0 60.0
Note. (A) Commercial CNBr-activated Sepharose-4B from Sigma; (B) 250 mg CNBr/g wet wt. Sepharose-4B; (C) 100 mg CNBr/g wet wt. Sepharose-4B; (D) 50 mg CNBr/g wet wt. Sepharose-4B. Substitution levels were determined using the method described in (29).
267
ACCESSIBLE IMMOBILIZED @NAD 1 # AND THE KINETIC LOCKING-ON STRATEGY
TABLE 3 The Concentration of Accessible S -Linked Immobilized NAD 1 Determined Using Enzymatic Reduction with Yeast Alcohol Dehydrogenase 6
A B C D
mmol NAD 1 (per g. wet wt.)
mmol NADH a (per g wet wt.)
Correction factor b
[NADH] Corrected (mmol/g/wet wt.)
Percentage reduction
1.9 0.7 0.4 0.3
0.56 0.27 0.23 0.15
1.24 1.06 1.19 1.27
0.7 0.3 0.3 0.2
35.8 43.3 67.5 63.3
Note. Four S 6-linked immobilized NAD 1 derivatives were synthesized to give varying substitution levels (1,3-diaminopropanol was used as the spacer arm compound and decreasing substitution levels were achieved by decreasing the concentration of CNBr used to activate Sepharose-4B: (A) Commercial CNBr-activated Sepharose-4B from Sigma; (B) 250 mg CNBr/g wet wt. Sepharose-4B; (C) 100 mg CNBr/g wet wt. Sepharose-4B; (D) 50 mg CNBr/g wet wt. Sepharose-4B). a Calculated from the change in absorbance at 340 nm after enzymatic reduction using (a) the molar extinction coefficient of NADH and (b) using a standard curve of NADH prepared in the presence of Sepharose 4B (to correct for light scattering). Results calculated using these two methods varied by less than 0.6%. b Using the same concentrations determined for the immobilized NAD 1 derivatives, soluble NAD 1 was enzymatically reduced under corresponding experimental conditions. The percentage enzymatic reduction of soluble NAD 1 to NADH was used to calculate the correction factors.
dependent dehydrogenases using these affinity matrices in conjunction with the kinetic locking-on tactic (unpublished data). In order to establish the immobilized NAD 1 population accessible for interaction with enzymes, the S 6linked immobilized NAD 1 derivatives were enzymatically reduced with yeast alcohol dehydrogenase. The results obtained are presented in Table 3. It was found that depending upon the initial concentration of immobilized NAD 1, the percentage enzymatic reduction to NADH varied from 36 to 68%. The percentage reduction of NAD 1 to NADH was greater for the lower substituted gels, with the 0.4 and 0.3 mmol NAD 1/g wet wt. matrices gels yielding percentage reductions of 67.5 and 63.3%, respectively. However, the 35.8% reduction to NADH achieved for the highest substituted gel (1.9 mmol NAD 1/g wet wt.) is closest to the value obtained by Barry and O’Carra (2). These investigators reported that of the total immobilized NAD 1 (substitution level of 2 mmol NAD 1/ml packed gel) only 20% was available for enzyme interaction. The S 6-linked immobilized NAD 1 derivatives with NAD 1 concentrations of 0.7 and 0.4 mmol NAD 1/g wet wt. matrix were found to have similar accessible NAD 1 concentrations (approximately 0.3 mmol NAD 1/g wet wt.). It thus seemed reasonable to presume that the chromatographic behavior of NAD 1-dependent dehydrogenases would be similar with these two matrices. However, while this was the case for YADH (see below), bovine heart L-LDH distinguished between these two matrices and exhibited a greater affinity for the higher substituted matrix. Since these accessibilities were determined using YADH, there was a possibility that different values would have been obtained if a
different enzyme had been used for the reduction. Differing accessibilities with different enzymes could be explained in terms of steric hindrance effects introduced by the Sepharose-4B matrix itself. However, although bovine heart L-LDH has a lower molecular weight than YADH (140,000 and 150,000 daltons, respectively) it is difficult to believe that such a small difference in molecular weights would have a significant effect on accessibility. The possibility that L-LDH is binding to the residual AMP present on these S 6linked immobilized NAD 1 derivatives was ruled out. Perhaps a plausible explanation for these experimental observations lies in the earlier reports of Barry and O’Carra (2), who showed that a variety of different NAD 1-dependent dehydrogenases have differing affinities for any one immobilized cofactor derivative. Presumably this is due to subtle differences in the mode of cofactor binding which is exacerbated as a result of the immobilization linkage. Whatever the reason, these results suggest that enzymatic reduction studies on immobilized NAD 1 derivatives should ideally be performed with each individual dehydrogenase under investigation. Direct Chemical Coupling Procedures Versus Dilution with Underivatized Sepharose-4B for the Preparation of Immobilized NAD 1 Derivatives with Varying Substitution Levels The effect of substitution level on the chromatographic behavior of NAD 1-dependent dehydrogenases on S 6-linked immobilized NAD 1 derivatives was investigated using the affinity matrices discussed in the previous section. Furthermore, the effect of varying the
268
MULCAHY ET AL.
substitution level by (i) suitably diluting the matrix with unsubstituted Sepharose-4B or (ii) direct coupling of the required ligand concentration to the inert matrix were compared. The test enzyme used for these studies was bovine heart L-LDH. This lactate dehydrogenase was so strongly adsorbed to the S 6-linked immobilized NAD 1 derivative (1.9 mmol NAD 1/g wet wt.) through binary complex formation that the enzyme had to be competitively eluted with 59AMP or NADH (Figs. 3A and 4A, respectively). That at least part of this affinity was for the residual AMP present on the affinity matrix (see Table 2) is clear from the results presented in Fig. 3B. Furthermore, demonstration of the locking-on tactic using this matrix and L-LDH seemed precluded at first since it would be difficult to show the reinforcement in strength of binding brought about by a soluble locking-on ligand. However, use of AMP as a “stripping” ligand allowed a distinction to be made between those experimental conditions that promoted locking-on of the L-LDH activity and those that did not. From the results presented in Figs. 3C and 3D it is clear that while 0.05 mM oxalate did not promote locking-on of the LDH to the S 6-linked immobilized NAD 1 derivative (1.9 mmol NAD 1/g wet wt.), 0.5 mM oxalate was effective as locking-on ligand. When bovine heart L-LDH was chromatographed on a lower substituted S 6-linked immobilized NAD 1 derivative (chemically synthesized to a substitution level of 0.7 mmol NAD 1/g wet wt.) in the absence of locking-on ligand, some leakage of the enzyme occurred before elution from the column in the presence of NADH (Fig. 4B). The lower substituted matrices did not retard L-LDH activity to any significant extent (Figs. 4C and 4D). When similar studies were carried out on S 6-linked immobilized NAD 1 derivatives diluted with underivatized Sepharose 4B, a similar LDH chromatographic pattern was observed in that the affinity of the enzyme for the S 6-linked immobilized NAD 1 decreased with decreasing substitution level (Fig. 5). However, some discrepancies were observed. For example, the chromatographic behavior of LDH on the undiluted matrix (0.4 mmol NAD 1/g wet wt; see Fig. 4C) was quite different to that seen with the diluted matrix (0.5 mmol NAD 1/g wet wt; see Fig. 5C). Based on the proportion of immobilized NAD 1 accessible for enzyme interaction (0.3 and 0.2 mmol/g wet wt. for the undiluted and diluted matrix, respectively) the reverse chromatographic pattern would have been expected; i.e., that the LDH would have had a higher affinity for the matrix used in Fig. 4C. The possibility was considered that the stronger affinity of LDH for the diluted matrix may be in part due to the higher concentration of residual AMP present on the S 6-linked immobilized NAD 1 derivative (0.2 and 0.6 mmol AMP/g wet wt. for the undiluted and diluted matrix, respectively). However, while bovine heart LDH shows weak affinity for
FIG. 3. Chromatography of bovine heart L-LDH on an S 6-linked immobilized NAD 1 derivative (A, C, and D) and on an S 6-linked AMP derivative (B). The irrigant was 50 mM potassium phosphate buffer (pH 7.4) with additions as indicated by the horizontal lines. Chromatography was carried out at room temperature.
an S 6-linked immobilized AMP derivative with a substitution level of 2.2 mmol/g wet wt matrix (Fig. 3B), matrices with immobilized AMP concentrations of less
ACCESSIBLE IMMOBILIZED @NAD 1 # AND THE KINETIC LOCKING-ON STRATEGY
FIG. 4. Chromatography of bovine heart L-LDH on S 6-linked immobilized NAD 1 derivatives synthesized to give varying substitution levels. (A) 1.9 mmol NAD 1/g wet wt; (B) 0.7 mmol NAD 1/g wet wt; (C) 0.4 mmol NAD 1/g wet wt; (D) 0.3 mmol NAD 1/g wet wt. The irrigant was 50 mM potassium phosphate buffer (pH 7.4). Chromatography was carried out at room temperature.
than 1 mmol/g wet wt. did not adsorb this enzyme. The accessible immobilized NAD 1 ligand concentration reported in Table 3 was determined using YADH rather
269
FIG. 5. Chromatography of bovine heart L-LDH on S 6 -linked immobilized NAD 1 derivatives diluted with underivatized Sepharose 4B to give varying substitution levels. (A) undiluted S 6 -linked NAD 1 (1.9 mmol NAD 1 /g wet wt.); (B) twofold diluted S 6 -linked NAD 1 (1.0 mmol NAD 1 /g wet wt.); (C) fourfold diluted S 6 -linked NAD 1 (0.5 mmol NAD 1 /g wet wt.); (D) eightfold diluted S 6 -linked NAD 1 (0.3 mmol NAD 1 /g wet wt.). The irrigant was 50 mM potassium phosphate buffer (pH 7.4). Chromatography was carried out at room temperature.
270
MULCAHY ET AL.
matographed on an eightfold diluted S 6-linked immobilized NAD 1 derivative (substitution level of 0.3 mmol NAD 1/g wet wt.) in the absence of locking-on ligand showed only marginal affinity for this matrix (Fig. 6A). Inclusion of 0.02 mM oxalate in the irrigant locked the majority of the enzyme onto the same matrix, but the LDH activity only partially autoeluted when the locking-on ligand was discontinued (Fig. 6B). The remaining L-LDH was only eluted by the subsequent use of NADH as a competitive counter-ligand. Increasing the
FIG. 6. Chromatography of bovine heart L-LDH in the absence (A) and presence (B and C) of varying concentrations of oxalate (0.02 and 0.2 mM, respectively). Chromatography was carried out on an S 6linked immobilized NAD 1 derivative diluted eightfold with underivatized Sepharose-4B to give a substitution level of 0.3 mmol NAD 1/g wet wt. of gel. The irrigant was 50 mM potassium phosphate buffer (pH 7.4).
than L-LDH. However identical accessible NAD 1 concentrations were determined using bovine heart L-LDH and therefore the possibility was ruled out that different accessible immobilized NAD 1 values are obtained if the accessibility studies are carried out using bovine heart L-LDH for enzymatic reduction of the S 6-immobilized NAD 1 derivatives. The discrepancies could alternatively be attributed to the diluted affinity matrix having a more heterogeneous distribution of immobilized NAD 1 than the chemically synthesized gel. This is further suggested by the results presented in Figs. 6 and 7. L-LDH chro-
FIG. 7. Chromatography of bovine heart L-LDH on S 6-linked immobilized NAD 1 derivatives in the presence of 0.2 mM oxalate. (A) S 6-linked immobilized NAD 1 diluted eightfold with underivatized Sepharose-4B to give a substitution level of 0.3 mmol NAD 1/g wet wt. gel; (B) S 6-linked immobilized NAD 1 diluted twofold with underivatized Sepharose-4B to give a substitution level of 0.3 mmol NAD 1/g wet wt. gel; (C) S 6-linked immobilized NAD 1 synthesized to give a substitution level of 0.3 mmol NAD 1/g wet wt. The irrigant was 50 mM potassium phosphate buffer (pH 7.4). Chromatography was carried out at room temperature.
ACCESSIBLE IMMOBILIZED @NAD 1 # AND THE KINETIC LOCKING-ON STRATEGY
oxalate concentration to 0.2 mM ensured that all the LDH was locked-on. However, once again the LDH activity only partially autoeluted when the locking-on ligand was discontinued (Fig. 6C) and NADH had to be used to complete elution of the enzyme. That this continued adsorption is probably due to the heterogeneous distribution of immobilized NAD 1 throughout the diluted affinity matrix is further suggested by the results presented in Fig. 7. For these studies three different S 6-linked immobilized NAD 1 preparations were used: (i) an S 6-linked immobilized NAD 1 diluted eightfold with underivatized Sepharose-4B to give a substitution level of 0.3 mmol NAD 1/g wet wt. (Fig. 7A); (ii) an S 6-linked immobilized NAD 1 diluted twofold with underivatized Sepharose-4B to give a substitution level of 0.3 mmol NAD 1/g wet wt. (Fig. 7B); and (iii) an undiluted S 6-linked immobilized NAD 1 derivative chemically synthesized to give a substitution level of 0.3 mmol NAD 1/g wet wt. (Fig. 7C). Thus while all three affinity matrices had similar substitution levels, the eightfold diluted matrix would be expected to contain the most heterogeneous distribution of immobilized ligand. In the absence of locking-on ligand the majority of the applied L-LDH activity eluted from all three columns within 5 or 6 column vol (e.g., Figs. 4D and 5D). However, while 0.2 mM oxalate locked-on the L-LDH activity to all three affinity matrices, the pattern of elution differed significantly (Fig. 7). With the eightfold diluted matrix the majority of the enzyme activity did not autoelute after omission of the locking-on ligand from the matrix (Fig. 7A). Conversely when the twofold diluted matrix was used for chromatography the majority of the L-LDH autoeluted after discontinuation of oxalate (Fig. 7B), although the peak of activity was quite broad. With the undiluted matrix all of the activity autoeluted after omission of oxalate from the irrigant (Fig. 7C). This pattern of chromatographic behavior can perhaps be explained in terms of the distribution of immobilized NAD 1. Once the enzyme activity is locked-on to the immobilized ligand it becomes localized in regions of the matrix with high concentrations of NAD 1 and has difficulty dissociating from the immobilized cofactor even when the locking-on ligand is discontinued. It has been reported that lowering substitution levels by dilution with unmodified Sepharose has an advantage over matrices that have been chemically synthesized to the desired substitution level. The explanation given for this conclusion is reduced nonbiospecific adsorption effects with diluted affinity matrices (7). However, the results presented here strongly suggest that chemical synthesis of the required immobilized ligand concentration is the preferable strategy.
271
The Effect of Accessible Immobilized Ligand Concentration on the Concentration of Locking-on Ligand Required to Promote Adsorption of YADH to an S 6-Linked Immobilized NAD 1 Derivative O’Carra et al. (20) reported that hydroxylamine and pyrazole are competitive inhibitors of YADH activity (competitive with respect to ethanol) and that both are effective as locking-on ligands. Further investigations were carried out in the present study in order to establish if the dissociation constants of locking-on ligands influenced the concentration of locking-on ligand required for adsorption of YADH to S 6-linked immobilized NAD 1 derivatives. Pyrazole was found to be a much stronger competitive inhibitor of YADH than hydroxylamine (K i values of 0.05 and 0.83 mM, respectively) and in keeping with this, lower concentrations of pyrazole were found to facilitate the locking-on of YADH to an S 6-linked immobilized NAD 1 derivative. For example, using an S 6-linked immobilized NAD 1 derivative with a substitution level of 1.9 mmol total NAD 1/g wet wt. and 0.7 mmol accessible NAD 1/g wet wt., 5 mM pyrazole produced strong enhancement of YADH adsorption, while 5 mM hydroxylamine had a negligible effect. (All chromatography was carried out in 100 mM sodium pyrophosphate buffer pH 8.8.) These results prove that when the immobilized NAD 1 concentration is kept constant, the K i value of the locking-on ligand determines the effective concentration of soluble locking-on ligand required for adsorption. Affinity chromatographic studies were performed on the four S 6-linked immobilized NAD 1 derivatives synthesized to various substitution levels (Table 3) using YADH and a decreasing concentration gradient of hydroxylamine as locking-on ligand (Fig. 8). The results revealed that YADH eluted maximally at progressively lower concentrations of soluble locking-on ligand as the substitution level of the affinity matrix was increased (Fig. 8). YADH exhibited only marginal affinity for the lowest substituted gel (0.3 mmol NAD 1/g wet wt.) in the presence of 100 mM hydroxylamine (Fig. 8A). That this was not due to overloading of the matrix was confirmed by the observation that higher concentrations of hydroxylamine locked-on all the applied YADH activity to the same affinity adsorbent. Conversely, with the highest substituted matrix (1.9 mmol NAD 1/g wet wt.) the enzyme was strongly locked-on at 100 mM hydroxylamine and eluted maximally at 17 mM hydroxylamine (Fig. 8D). That the two other S 6-linked immobilized NAD 1 preparations (substitution levels of 0.4 and 0.7 mmol NAD 1/g wet wt.) would show a requirement for intermediate concentrations of hydroxylamine was confirmed when the YADH eluted maximally at 35 and 37 mM hydroxylamine, respectively (Figs. 8B and 8C). However, it was quite puzzling initially that the yeast enzyme did not appear to distinguish between the two
272
MULCAHY ET AL.
ACCESSIBLE IMMOBILIZED @NAD 1 # AND THE KINETIC LOCKING-ON STRATEGY
S 6-linked immobilized NAD 1 preparations which had quite different substitution levels of 0.4 and 0.7 mmol NAD 1/g wet wt. (Fig. 8B and 8C, respectively), but clearly distinguished between two S 6-linked immobilized NAD 1 preparations with more similar substitution levels of 0.3 and 0.4 mmol NAD 1/g wet wt. (Figs. 8A and 8B, respectively). This chromatographic elution pattern, however, could be rationalized when interpreted in terms of the available or accessible immobilized NAD 1 concentration. Enzymatic reduction with YADH revealed accessible immobilized NAD 1 concentrations of 0.7, 0.3, 0.3, and 0.2 mmol NAD 1/g wet wt. for the matrices with total immobilized NAD 1 concentrations of 1.9, 0.7, 0.4, and 0.3 mmol NAD 1/g wet wt., respectively. Thus although two of the S 6-linked immobilized NAD 1 preparations had quite different total immobilized NAD 1 concentrations of 0.4 and 0.7 mmol NAD 1/g wet wt., these preparations had identical accessible immobilized NAD 1 concentrations (0.3 mmol NAD 1/g wet.). It is also quite clear from the results that at low substitution levels, relatively small differences in accessible immobilized NAD 1 concentration can have quite a dramatic affect on the locking-on of an NAD 1-dependent dehydrogenase. For example, YADH exhibited only marginal affinity for the lowest substituted matrix (0.2 mmol accessible NAD 1/g wet wt.) in the presence of 100 mM hydroxylamine as locking-on ligand (Fig. 8A). Conversely the yeast enzyme was completely locked-on to an immobilized NAD 1 derivative with a slightly higher substitution level (0.3 mmol accessible NAD 1/g wet wt.) at the same hydroxylamine concentration and eluted maximally at 35 mM hydroxylamine (Fig. 8B). These studies thus confirm the importance of determining both the total immobilized ligand concentration and the accessible immobilized ligand concentration when studying the bioaffinity chromatographic behavior of NAD(P) 1-dependent dehydrogenases on immobilized derivatives of the dinucleotides using the locking-on tactic. As discussed above for L-LDH, an attempt was made to explain this in terms of a reduced concentration of leading substrate resulting in increased K i values for the locking-on ligand. Kinetic studies with YADH showed that as the concentration of NAD 1 increased, the K m value for the specific substrate (ethanol) and the K i value for the competitive inhibitor of the specific substrate (hydroxylamine) decreased. Progressively increasing the soluble NAD 1 concentration from 0.08 mM, to 0.3 and 0.7 mM, resulted in a significant de-
273
crease in the K m values of YADH for ethanol (10.5, 3.5, and 1.9 mM, respectively) and in the K i values for hydroxylamine (0.83, 0.21, and 0.18 mM, respectively). It has already been shown in the previous section that the concentration of soluble locking-on ligand required to lock-on a particular NAD 1-dependent dehydrogenase to an immobilized NAD 1 derivative is dependent on the dissociation constant of the competitive inhibitory substrate analogue; the lower the K i value, the lower the effective concentration of soluble locking-on ligand. Further studies are currently underway aimed at investigating and further developing the potential of the locking-on tactic for protein purification applications and for mechanistic investigations of a wide range of NAD(P) 1-dependent dehydrogenases. Purification of Bovine Heart L-LDH in a Single Bioaffinity Chromatographic Step The single chromatographic-step purification of L-LDH from bovine heart tissue was carried out on an S 6 -linked immobilized NAD 1 derivative. Previous investigations had shown that the L-LDH could be locked-on biospecifically to this affinity adsorbent using oxalate. The bovine heart extract was first concentrated using ammonium sulfate fractionation. The crude cellular extract (after a 20 to 80% saturated ammonium sulfate fractionation step) was directly applied to the S 6 -linked immobilized NAD 1 derivative in the absence or presence of 5 mM oxalate. In the former case, the L-LDH activity eluted with the protein breakthrough peak and there was no significant change in the enzyme specific activity (0.72 mmol/min/mg of protein). When applied to the immobilized NAD 1 derivative in the presence of 5 mM oxalate (locking-on ligand), the L-LDH activity was adsorbed to the matrix and was separated from the main protein breakthrough peak (Fig. 9A). A short pulse of 59-AMP was added to the irrigant a couple of column volumes prior to discontinuation of the locking-on ligand so that contamination of the final L-LDH preparation (by other proteins bound weakly to the affinity adsorbent) would be avoided. The L-LDH cleanly eluted upon omission of oxalate from the irrigant (Fig. 9A). This procedure resulted in a homogeneous preparation of L-LDH as demonstrated using SDS–PAGE (Fig. 9B). The percentage recovery of L-LDH after bioaffinity chromatography, the final enzyme specific activity, and the purifica-
FIG. 8. The effect of accessible immobilized NAD 1 concentration on the concentration of hydroxylamine required to lock-on YADH to S 6-linked immobilized NAD 1 derivatives (hydrophilic spacer arm). The four S 6-linked immobilized NAD 1 preparations of varying substitution levels were prepared by varying the concentration of CNBr used to activate Sepharose 4B as described under Materials and Methods. The total immobilized NAD 1 concentration was determined using the method of Mosbach (29) and the concentration of accessible S 6-linked immobilized NAD 1 was determined using enzymatic reduction with yeast alcohol dehydrogenase. Chromatography was carried out in 100 mM sodium pyrophosphate buffer (pH 8.8) at 4°C.
274
MULCAHY ET AL.
FIG. 9. Bioaffinity purification of L-LDH from bovine heart in a single chromatographic-step using the locking-on tactic: (A) Elution profile of bovine heart L-LDH on an S 6-linked immobilized NAD 1 derivative (see experimental section for details) and (B) Protein-stained SDS–polyacrylamide gel electropherogram of bovine heart L-LDH (center lane) purified as in A. The outer lanes in (B) contain the protein molecular weight standards.
tion factor were 96.6%, 478 mmol/min/mg, and 673, respectively. This sample purification clearly demonstrates the potential of biospecific affinity systems based on the locking-on strategy for protein purification applications. Our research efforts are currently focused on the design, synthesis, and development of new immobilized cofactor derivatives. These developments will ensure that the kinetic locking-on strategy will be widely applicable to the one-step purification of a variety of other cofactor-dependent enzymes.
ACKNOWLEDGEMENTS This work was funded by the Forbairt Applied Research Programme and the Graduate Training Programme under the Operational Programme for Industrial Development.
REFERENCES 1. O’Carra, P., and Barry, S. (1972) Affinity chromatography of lactate dehydrogenase—Model studies demonstrating the potential of the technique in the mechanistic investigation, as well as in the purification, of multisubstrate enzymes. FEBS Lett. 21(3), 281–285. 2. Barry, S., and O’Carra, P. (1973) Affinity chromatography of
ACCESSIBLE IMMOBILIZED @NAD 1 # AND THE KINETIC LOCKING-ON STRATEGY
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
nicotinamide-adenine dinucleotide-linked dehydrogenases on immobilized derivatives of the dinucleotide. Biochem. J. 135, 595– 607. Barry, S., and O’Carra, P. (1973) A simple general method for the preparation of ‘6-immobilized’ analogues of AMP, NAD and of other adenine-containing compounds for affinity chromatography. FEBS Lett. 37(2), 134 –139. O’Carra, P., Barry, S., and Griffin, T. (1973) Spacer arms in affinity chromatography: The need for a more rigorous approach. Biochem. Soc. Trans. 1, 289 – 290. Lowe, C. R., and Dean, P. D. G. (1973) Affinity chromatography of lactate dehydrogenase on immobilized nucleotides. Biochem. J. 133, 515–520. Lowe, C. R., Harvey, M. J., Craven, D. B., and Dean, P. D. G. (1973) Some parameters relevant to affinity chromatography on immobilized nucleotides. Biochem. J. 133, 499 –506. Harvey, M. J., Craven, D. B., Lowe, C. R., and Dean, P. D. G. (1974) N 6-immobilized 59-AMP and NAD 1: Preparations and applications. In “Methods in Enzymology” (Jakoby, W. B., and Wilcheck, M., Eds.), Vol. XXXIV, Part B, pp. 242–253, Academic Press, New York. Harvey, M. J., Lowe, C. R., Craven, D. B., and Dean, P. D. G. (1974) Affinity chromatography on immobilized adenosine 59monophosphate. 2. Some parameters relating to the selection and concentration of the immobilized ligand. Eur. J. Biochem. 41, 335–340. Harvey, M. J., Lowe, C. R., and Dean, P. D. G. (1974) Affinity chromatography on immobilized adenosine 59-monophosphate. 5. Some applications of the influence of temperature on the binding of dehydrogenases and kinases. Eur. J. Biochem. 41, 353–357. O’Carra, P., Barry, S., and Griffin, T. (1974) Spacer arms in affinity chromatography: use of hydrophilic arms to control or eliminate non-biospecific adsorption effects. FEBS Lett. 43(2), 169 –175. O’Carra, P., Barry, S., and Griffin, T. (1974) Interfering and complicating adsorption effects in bioaffinity chromatography. In “Methods in Enzymology” (Jakoby, W. B., and Wilcheck, M., Eds.), Vol. XXXIV, Part B, pp. 108 –126, Academic Press, New York. O’Carra, P., Barry, S., and Corcoran, E. (1974) Affinity chromatographic differentiation of lactate dehydrogenase isoenzymes on the basis of differential abortive complex formation. FEBS Lett. 43(2), 163–169. O’Carra, P. (1974) Affinity chromatography. Bioaffinity chromatography: Theoretical aspects and complicating factors. Biochem. Soc. Trans. 2, 1289 –1293. O’Carra, P. (1974) Affinity chromatography of enzymes. In “FEBS Symp: Industrial Aspects of Biochemistry” (Spencer, B., Ed.), pp. 107–134, North Holland, Amsterdam. Lee, C. Y., Lappi, D. A., Wermuth, B., Everse, J., and Kaplan, N. O. (1974) 8-(6-aminohexyl)-amino-adenine-nucleotide derivative for affinity chromatography. Arch. Biochem. Biophys. 163, 561–569.
275
16. Barry, S., Griffin, T., and O’Carra, P. (1974) Synthesis of affinity gels. A reassessment of the advantages and disadvantages of the modular solid phase approach. Biochem. Soc. Trans. 2, 1319 – 1322. 17. Lowe, C. R., Harvey, M. J., and Dean, P. D. G. (1974) Affinity chromatography on immobilized adenosine 59-monophosphate. 3. The binding of glycerokinase and lactate dehydrogenase in relation to column geometry and dynamics. Eur. J. Biochem. 41, 341–345. 18. Lowe, C. R., Harvey, M. J., and Dean, P. D. G. (1974) Affinity chromatography on immobilized adenosine-59-monophosphate. Some kinetic parameters involved in the binding of group specific enzymes. Eur. J. Biochem. 42, 1– 6. 19. O’Carra, P. (1978) Theory and practice of affinity chromatography. In “Chromatography of Synthetic and Biological Macromolecules” (Epton, R., Ed.), Vol. 2, Chap. 11, pp. 131–158, Ellis Horwood Ltd/Chem. Soc., London. 20. O’Carra, P., Griffin, T., O’Flaherty, M., Kelly, N., and Mulcahy, P. (1996) Further studies on the bioaffinity chromatography of NAD 1-dependent dehydrogenases using the locking-on tactic. Biochim. Biophys. Acta 1297, 235–243. 21. Mulcahy, P., Griffin, T., and O’Carra, P. (1997) Biospecific affinity chromatography of octopine dehydrogenase from molluscs. Protein Exp. Purif. 9(1), 109 –114. 22. O’Flaherty, M., McMahon, M., and Mulcahy, P. (1999) A kinetic locking-on strategy for bioaffinity purification: Further studies with alcohol dehydrogenase. Protein Exp. Purif. 15, 127–145. 23. O’Flaherty, M., McMahon, M., Forde, J., and Mulcahy, P. (1999) The kinetic locking-on strategy for bioaffinity purification: Further studies with bovine liver glutamate dehydrogenase. Accompanying paper. 24. Lowe, C. R., and Dean, P. D. G. (1974) In “Affinity Chromatography,” Wiley. 25. Dean, P. D. G., and Harvey, M. J. (1974) The importance of ligand concentration, pH and temperature in affinity chromatographic systems. Biochem. Soc. Trans. 2, 1306 –1308. 26. Parikh, I., March, S. T., and Cuatrecasas, P. (1974) Topics in the methodology of substitution reactions with agarose. In “Methods in Enzymology” (Jakoby, W. B., and Wilcheck, M., Eds.), Vol. XXXIV, pp. 77–102, Academic Press, New York. 27. Lowry, O. H., Rosenbrough, N. J., Farr, A. L., and Randall, R. J. (1951) Protein measurement with the folin phenol reagent. J. Biol. Chem. 193, 265–275. 28. Fallia, D., and Santi, D. V. (1973) A simple method for quantitating ligands covalently bound to agarose beads. Anal. Biochem. 52, 363–368. 29. Mosbach, K. (1974) AMP and NAD 1 as general ligands. In “Methods in Enzymology” (Jakoby, W. B., and Wilcheck, M., Eds.), Vol. XXXIV, pp. 229 –242, Academic Press, New York. 30. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of bacteriophage T4. Nature 227, 680 – 685.