lsoelectric focusing of human catalase

Eleclrophoresis 1990,II, 635-638

Antonio Alonso' Guillemo Viedo* Manuel Sancho' Jose Fernandez-Piqueras2 'Seccion de Biologia, Instituto Nacional de Toxicologia, Madrid Wnidad de Genetics, Departmento de Biologia, Facultad de Ciencias, Universidad Autonoma de Madrid

635

Characterization of human catalase by isoelectric focusing in presence of urea Human catalase from erythrocytes and liver were analyzed by polyacrylamide gel isoelectric focusing in presence and absence of urea using two different pH gradients, namely pH 6-8 and pH 6.7-7.7. In presence of urea, human catalase focused in the pH range 6.75-7.0, slightly anodal to that of hemoglobin A. In narrow pH gradients, human erythrocyte catalase was microheterogeneous. Neuraminidase from different sources and peptide-N-glycosidase F were applied to investigate the presence of sialic acid and/or carbohydrate chains in humancatalase. A shift in the focusing pattern of both erythrocyte and liver catalase towards the anode was observed after treatment with one of the commercially available neuraminidase preparations. This unusual result could be related to a contaminating protease since no effect was observed when the catalases were treated in presence of a serine protease inhibitor. In contrast, bovine liver and Maccaca erythrocyte catalase did not display any detectable change in their focusing patterns after treatment with any ofthe neuraminidase preparations.

1 Introduction Catalase (H,O,: H,Oz oxidoreductase, EC 1.11.1.6) is a hydrogen peroxide detoxifier which occurs in aerobic organisms. Human erythrocyte catalase (HEC) is a hemeprotein of oligomeric structure consisting of four subunits [ 11 and it has recently been reported that each tetrameric molecule contains four tightly bound molecules of N A D P H [2]. HEC was first electrophoretically analyzed by using starch gel [31 and polyacrylamide gel electrophoresis [41in order to compare the electrophoretic properties of catalase in normal individuals with those of individuals homozygous for acatalasemia, a hereditary defect characterized by the apparent lack of catalase [ 5 ] . More recently, HEC from normal and acatalasemic J apanese individuals has been analyzed by agarose gel isoelectric focusing (IEF) followed by enzyme activity detection or by immunoblotting [6,71. It has been shown in arecent article that an anomalous splicing of the mRNA of the catalase could cause acatalasemia in Japanese patients 181. In this article, we describe the separation of HEC fromnormal individuals by polyacrylamide gel I E F in presence of urea. The effects of different neuraminidase treatments on the I E F pattern of human catalase are also reported.

2 Materials and methods 2.1. Sample preparation HEC has been analyzed in blood samples from 150 healthy Spanish unrelated individuals. Samples were collected by venipuncture into 10 mL tubes containing EDTA as anticoagulant. Red cells were washed three times with saline and stored at - 40 "C until use. Samples were thawed and diluted with distilled water (1 :10 - 1:200) immediately before IEF. Best results were obtained for dilutions of hemolysates ranging from 1:lOO to 1:200. Less diluted samples showed excessive staining with impaired resolution while higher dilutions did not contain enough activity for clear detection. Correspondence: Dr. Antonio Alonso, Seccion de Biologia, lnstituto Nacional de Toxicologia, CiLuis Cabrera 9, E-2800 Madrid, Spain Abbreviations: BLC, bovine liver catalase; HbA, hemoglobin A ; HEC, human erythrocyte catalase; HLC, human liver catalase; IEF, isoelectric focusing; MEC, Maccaca erythrocyte catalase; PMSF, phenylmethylsulfonyl fluoride; PNGase F, peptide-A-glycosidase F (0VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990

Human livers were &tained from autopsies performed within 24 h after death. Human liver catalase (HLC) was anal!,ied employing crude liver extracts, prepared as follows: frozen liver pieces were cut with a scalpel to obtain approximately 1 mm thin slices, washed twice in phosphate buffered saline and three times in distilled water to eliminate blood. The washed slices were homogenized in distilled water (100 c(.L/20 mg of tissue) using a glass/glass homogenizer. The homogenates were centrifuged at 15 000 rpm for 10 min and the supernatants subjected to IEF. All steps during preparation were performed at 4 "C. Purified bovine liver catalase (BLC) was purchased from Boehringer Mannheim and Maccaca erythrocyte catalase (MEC) was analyzed in blood from M . sylvana and M .fascyculans.

2.2 Enzyme treatments Twenty pL of neuraminidase (1 U/mL in a buffer recommended by the supplier) were added to 20 pL of hemolysate diluted 1: 100, 20 pL of liver supernatants or 20 1 L of purified BLC, followed by incubation at 4 "C, 25 "C or 37 "C for 6-24 h. Three different sources of neuraminidase were assayed: neuraminidase from Clostridium perfringens (purchased from Boehringer-Mannheim and Sigma) and neuraminidase from Vibrio cholerae (from Boehringer-Mannheim). The serine protease inhibitor phenylmethylsulfonyl fluoride (PMSF) was included at a final concentration of 5 mM in some experiments with neuraminidase. Peptide-N-glycosidase F (PNGase F from Flavobacteriurn meningosepticum, Boehringer-Mannheim) was used in a volume of 20 pL (200 U/mL in a buffer containing 20 mM potassium phosphate, 50 mM EDTA and 0.05 % sodium azide, pH 7.2) to 20 pL ofhemolysate, diluted 1:100, or to 20 pL of liver supernatants, followed by incubation at 25 "C for 16 h.

2.3 Preparation of polyacrylamide gels and IEF Miniaturized polyacrylamide gels (5 % T, 3 % C) containing carrier ampholytes of different pH ranges (Ampholine pH 6-8, Pharmalyte p H 6.7-7.7, Pharmalyte pH 6-8 and a mixture of Pharmalytes pH 6-8 and p H 4-6.5) were cast and polymerized in presence or absence of urea as previously described [9] with the modification that the acrylamide stock solution was made using a 0.5 % aqueous starch solution instead of distilled water. 0173-0835/90/0808-0635 %3.50+.25/0

636

A. Alonso el nl

Polyacrylamide gels with a 0-9 M urea gradient were generated by linearly mixing a 5 YOT, 3 % C polymerization solution, containing 0.75 % starch, 9 M urea and 2 % carrier ampholytes, pH 6-8, with a 5 % T, 3 96 C polymerization solution without urea, containing 0.5 % starch and 2 9 ' 6 carrier ampholytes, pH 6-8, using a two-chamber microgradient mixer (LKB 21 17 Multiphor I1 Gradient Marker). O n I E F the pH gradient was created perpendiculary to the urea gradient. I E F was performed using the Pharmacia equipment (flat bed apparatus FBE 3000, constant power supply ECPS 3000/150 and volt-hour integrator VH- 1). The platinized titanium electrodes rested directly on the gel surface with an interelectrode distance of 5 5 mm. Prefocusing was perl-ormed at limiting setings of 1 W, 350 V and 15 mA for 100 Vh. Samples were ap plied on the prefocused gels, using pieces of Whatman No. 3 filter paper (4 x 3 mm) at different positions and IEF was con tinued for 20 min with the same settings as during prefocusing. After removal of the paper tabs the gels were focused at 1200V,3Wand 15mAmaximumsettingsuntil1300-1600Vh were reached. The p l of the human catalase was estimated in the narrow p H gradient (6.7-7.7), using hemoglobin A as in ternal p l marker [ 101.

2.3 Visualization of catalase activity Catalase activity was detected by the starch iodine reaction [ 111. Focused gels were immersed for 10- 15s in a solution of 0.5 % H 2 0 2and0.15 % sodium thiosulfate and transferred to a 1.5 % KI solution, adjusted to pH 3.0 with acetic acid. Catalase activity was visualized in the gels, within minutes at room temperature, as unstained bands on a blue background. The stained gels were immediately photographed.

3 Results On IEF of HEC in a urea gradient, the enzyme activity was detected at the application site in the range from 0-4 M urea (Fig. 1 A). This activity was also observed when the applica-

ElectrophorePis 1990. I / . 635-638

tion site was varied(resu1ts not shown). Above 5 ~ u r e aHEC , migrated towards the cathode. reaching its final focusing posi tion at p l 6.75-7.00, close to hemoglobin A (Fig. 1 A). At higher urea concentrations (8-9 M), the HEC activity was strongly inhibited (Fig. IA). For this reason, 6 M urea was chosen for further experiments (Figs. 1 B, I C, 2 and 3). In a narrow pH 6.7-7.7 range, in presence of 6 M urea, at least 6 different bands with catalase activity could be detected (Fig. 1 C and Fig. 2 B). When human erythrocyte hemolysates were treated with different neuraminidase preparations the results were different. Hemoly sates treated with different batches of neuraminidase from C. perfringens, purchased from Boehringer Mannheim, always displayed a modified IEF pattern of H E C when compared with untreated samples. The most cathodal H E C zone disappeared with simultaneous displacement of the most anodal zone towards the anode, in comparison with control samples (Fig. 2). Control samples, in cubated both in either water or heat-inactivated neuraminidase preparations, did not show a change of IEF pattern (Fig. 2). N o change in the IEF pattern was observed in samples incubated with neuraminidase from C. perfringens, purchased from Sigma, or with preparations obtained from V. cholerae (Boehringer-Mannheim; Fig. 3A, lanes 8 and 9). Also, neither human erythrocyte hemolysates incubated with PNGase F (Fig. 3A, lane 10) nor human erythrocyte hemolysates incubated with neuraminidase from C. perfringens (Boehringer) in the presence of PMSF showed any detectable change in the HEC band pattern. In presence of 6 M urea, HLC displayed a pattern similar to that of HEC (Fig. 3A). Also for HLC a shifttowards the anode was observed after treatment with neuraminidase from C. perfringens(Boehringer;Fig. 3A, lane5)butnotwith theother neuraminidase preparations (Fig. 3A, lanes 2 and 3); PMSF inhibited the effect (Fig. 3A, lane 6). After PNGase treatments, no modification in the HLC pattern was observed (Fig. 3A, lane 4). In contrast, BLC and MEC did not display any detectable change in the I E F pattern after treatment with any of the neuraminidase preparations assayed.

Figure 1. IEF of HEC in miniaturized polyacrylamide gels. (A) Pattern of HEX in a polyacrylamide gel with a urea gradient (Ampholine pH 6-8). ( B ) Pattern of HEC in presence of 6 M urea. Ampholine p H 6-8. (C) Pattern of HEC in presence of 6 M urea. Pharmalyte p H 6.7-7.7. Samples: erythrocyte hemolysates diluted 1 :200 with distilled water. The original size o f the gels with an interelectrode distance of 5 5 rnm is shown.

;

Isoelectric focusing of human catalasc

Electrophoresis 1990, I I , 635- 638

637

Figure 2. Changes in the IEF patterns of HEC after incubation with neuraminidase at 40 "C. (A) IEF in 6 M urea, Ampholine pH 6-8. (B) IEF in 6 M urea, Pharmalyte pH 6.7-7.7. Samples: (I). (3) and (5) control hemolysates. incubated in distilled water for 24 h. 12 h and 6 h, respectively; ( 2 ) , (4) and (6) hemolysates incubated with neuraminidase for 24 h, 12 h and 6 h, respectively: (7) hemolysate incubated with heat-inactivated newaminidax for 24 h. The final dilution of all samples was 1 :200. The original size of the gels with an interelectrode distance of 55 mm is shown.

6

4

Figure3. I E F pattern of HEC, HLC, BLC and MEC after different incubations. (A) Pharmalyte pH 6-8. Samples: (1) human liver supernatant: (2) human liver supernatant incubated with neuraminidase (C. perfringens, Sigma); ( 3 ) human liver supernatant incubated with neuraminidase (V. cholerae, Boehringer); (4) human liver supernatant incubated with PNGase F; ( 5 ) human liver supernatant incubated with neuraminidase (C. pevfringens, Boehringer); (6) human liver supernatant incubated with neuraminidase ( C . perfringens, Boehringer) in presence of PMSF: (7) human hemolysate; (8) human hemolysate incubated with neuraminidase (C.perfringens,Sigma); (9) human hemolysate incubated with neuraminidase ( V . cholerae, Boehringer); (10) human hemolysate incubated with PNGase F; ( 1 1 ) human hemolysate incubated with neuraminidase (C.perfringens, Boehringer); (1 2 ) human hemolysateincubatedwithneuraminidase(C.perfringens, Boehringer)inpresenceofPMSF. (B)PharmalytepH 6--8 and pH 4-6.5 in a 2 : l ratio.(l)BLC; (2) BLC incubated with neuraminidase (C. perfringens, Boehringer); (3) hemolysate from M . sylvana: (4) hemolysate from M . sq'lvana incubated with neuraminidase (C. perfringens, Boehringer); ( 5 ) hemolysate from M.,fassyculans; (6) hemolysate from M.fascycu1ans incubated with neuraminidase (C. perfringetzs. Boehringer). Incubations were performed overnight at 25 "C in all cases.

4 Discussion The fact that HEC did not migrate from the application point, in the absence of urea, may indicate that the enzyme aggregated and precipitated on contact with the gel, or that the enzyme is adsorbed to the gel matrix. As this phenomenon has been suppressed in the presence of defined concentrations of urea, the interaction must be noncovalent. In agreement with Aebi et al. 1121, who pointed out that the catalase tetramer dissociates into its dimer in urea, the pattern obtained in the presence of 6 M urea could correspond to catalase dimers. The HEC pattern after IEF in narrow pH gradients in the presence of urea displayed microheterogeneity (Fig. 1 C and 2B). It has been previously pointed out that at least part of this microheterogeneity is related to different redox states of the catalase molecule I131.

We have also investigated the possible presence of sialic acid and carbohydrate chains in HEC. Specific transformations in the I E F pattern of H E C and HLC have been observed after neuraminidase treatments with enzyme preparations from C. pevfringens (Boehringer). However, samples treated with neuraminidase preparations from C. perfringens, but purchased from Sigma, or preparations obtained from Vibrio cholerae (Boehringer) did not produce any detectable transformations in the I E F pattern. Therefore, the modifications of the I E F pattern of HEC must be attributed to the action of some contaminant enzyme, present in one of the commercial preparations. This contaminant enzyme could be inactivated by heating (Fig. 2, lane 7) and seems to be a protease, on the basis of inhibition by PMSF (Fig. 3A, lanes 6 and 12).

638

J . H. Waterborg

Electrophoresis 1990. 1 I . 6.18-64 1

The neuraminidase preparation from C. perfringens (Boehringer) did not change the IEF patterns ofpurified BLC and erythrocyte hemolysates M . sylimia and h.3.fascyculans (Fig. 3B). Thus, if protease activity is involved in the action of this neuraminidase preparation on the human catalase, it seems that this protease has some degree of specifity. If carbohydrate chains, attached to asparagine, were contained in HEC, their removal, e. g. by endoglycosidases, would result in anodal forms with increased negative charge. However, treatment with PNGase F did not modify the HEC and HLC patterns (Fig. 3A. lanes 4 and lo), indicating that these catalases do not contain carbohydrate chains.

W e wish to thgnk the anonymous referees of this paper whose criticisms have considerably improved its content. Thanks are also due to the lnstituto Anatdmico Forense de Madrid f o r providing human livers and to the Animalario del Hospital Ram& y Cajal de Madrid, J . Palacin and C. Ferndndez,f o r providing Maccaca blood samples. The photographic work of Mrs. Augela Zamora is highly appreciated. Received January 29, 1990

Jakob H. Waterborg Divison of Cell Biology and Biophysics, School of Basic Life Sciences, University of MissouriKansas City, MO

5 References I l l Bonaventura, J., Schroeder. W. A. and Suen Fang.. Arch. Biachenr. Biophvs. 1972. 150. 606-617. 121 Kirkman, H. N . and(3aetani.G. F.,Proc.Natl.Acad.Sci. USA 1984. 81.4343-4341. 131 Acbi. M . and Suter. 11.. A t h . H u m . Genel. 1971.2, 143-199. 141 Ogata, M . and Miaugaki. G., H u m Gener. 1979.48, 329-338. IS1 Takahara, S. and Miyanioto, H.. J . Olorhi. Soc. Jpti. 1948. 81. 163- 164. I61 Ogata, M. and Satoh, Y., Elecrrophoresis 1988. 9, 128-131. 171 Ogata, M., Suzuki, K. and Satoh, Y.. Elec/rophoresis 1989. 10. 194- 198. IS1 Wen. J. K.. Osumi, T.. Hashimoto, T. and Ogata. M.. J . Mol. B i d . 19YO.211.383-393. 191 Alonso, A,./. For. Sri. 1988. 33. 1267-1272. I101 Righctti, P. G., Isoelectric Focusing: Theory, Methodology arid Applicarions. Elsevier, Amstcrdam 1983. 1111 Vallejos, C. E.. in: Isoz.vmes in Plant Generics andBreeding, Elsevier, Amsterdam 1983. Part A. pp. 469. I121 Scherz. B., Kuchinskas, E. J.. Wyss. R. S. and Aebi. H., Euro. J . Biochem. 1976,69,603-6 13. 1131 Miirikofer-Zwez, S., Cantz. M., Kaufmann. H.. von Wartburg. J. P. and Aebi, H., Eztro. J . Biochem. 1969. 11.49-57.

Involvement of cysteine residues in the electrophoretic mobility of histone H3 in acid-urea-Tritongels Carbamylation of cysteines 96 and 1 10 in histone H3 increases the electrophoretic mobility of this histone in acetic acid-urea-Triton X-100 polyacrylamide gels but has no effect in gels lacking Triton. Residue 96 appears to be a major determinant in the affinity of histone H3 for the nonionic detergent Triton. Carbamylation and carboxymethylation of cysteine 96 caused a major loss of the gel retardation caused by Triton. Carbamylation of cysteine 110 did not affect Triton binding but prevented ionization of the thiol side-chain moiety in the acetic acid-urea-Triton X-100 gel.

1 Introduction Core histones can be identified and distinguished from other proteins by their interaction with the nonionic detergent Triton X-100 in acetic acid-urea (AU) poiyacrylamide gels 1-21. The basis for this interaction is not understood, but it is observed for core histones of all species tested and has not been reported for other proteins. This study of the modification of cyste cysteine residues in histone H3 species with one or two cysteine groups has identified residue 96 as a location of major importance f for Triton binding. A much smaller role may be played by cysteine 1 10.An artifact was detected in two alfalfa histone H3 proteins which were prepared from isolated Correspondence: J. H. Waterborg, Ph. D., Division of Cell Biology and Biophysics, School of Basic Life Sciences, Room 4 14 BSB. 5 100 Rockhill Road, University ofMissouri-KansasCity,KansasCXy,MO 64 1 10-2499. USA Abbreviations: AU, acetic acid-urea; AUT, acetic acid-urea Triton X- 100; HPLC, high performance liquid chromatography; TFA, trifluoroacetic acid

5 VCH Verlagsgesellschaft mbH. D-6940 Weinheim, 1990

nuclei I3 1. It was similar or identical to the early effects of carbamylation of cysteine 110 in histone H3. Analysis of this modification and its preventions has resulted in revised estimates for the steady state acetylation of alfalfa histone H3.1 and H3.2 variants.

2 Materials and methods Calf thymus histone variant H3.1 with two cysteines and the lowest gel mobility in acetic acid-urea-Triton X-100 (AUT) gels 14-5 I was purified to homogeneity from a commercial preparation of whole calf thymus histones (Worthington) by reverse phase chromatography on Zorbax Protein Plus (Dupont). Histones were solubilized in 7.2 M freshly deionized urea, I M dithiothreitol, 0.75 M ammonium hydroxide and 0.05 % phenolphtalein for 5 min at room temperature and acidified by the addition of 1/20 volume of glacial acetic acid. Up to 4 mg of histones were injected on a 4.6 mm x 25 cm column, equilibrated with 0.1 % trifluoroacetic acid (TFA) in 01 73-0835/90/0808-0638 %3.50+.25/0