Purification and characterization of native and proteolytic forms of rabbit liver phosphorylase kinase

ht. J. Biocfieem.Vol. 22, No. 5, pp. 443-451, 1990 Printedin Great Britain. Al1 rights reserved

0020-71 IX/90 $3.00 + 0.00 Copyright Q 1990 Pergamon Press plc

PURIFICATION AND CHARACTERIZATION OF NATIVE AND PROTEOLYTIC FORMS OF RABBIT LIVER PHOSPHORYLASE KINASE JORGE BELETA,PILAR BENEDICTO,PEREAYMERICHand F. JAVIER GELLA Departamento de Bioquimica y Biologia Molecular, Unidad Docente de1 Hospital de la Sta. Creu i S. Pau, Universidad Autonoma de Barcelona, Avenida S. Antonio M Claret 167, 08025 Barcelona, Espaiia [Tef. 25619-00] (Receitred 2 August

1989)

Abstract-l. Two forms of phosphorylase kinase having mol. wt of 1,260,OOO(form I) and 205,000 (form II) have been identified by gel filtration chromatography of rabbit liver crude extracts. 2. Form I was the majority when the homogenization buffer was supplemented with a mixture of proteinase inhibitors. This form has been purified through a protocol including ultracentrifugation, gel filtration and affinity chromatography on Sepharoseheparin. 3. Form II was purified by a combination of chromatographic procedures including ion exchange, gel filtration and affinity chromatography on Sepharose-Blue Dextran and Sepharose-histone. 4. Upon electrophoresis in the presence of sodium dodecyl sulfate two subunits of 69,000 and 44,000 were identified for this low molecular weight enzyme. Thus, a tetrameric structure comprising two subunits of each kind can be proposed. 5. Treatment of form I with either trypsin or chymotrypsin gave an active fragment having a molecular weight similar to that of form II. On the contrary, other dissociating treatments with salts, thiols and detergents failed in producing forms of lower molecular weight. 6. The similarities between proteolyzed forms I and II were stressed by their behavior in front of antibodies raised against the muscle isoenzyme of phosphorylase kinase. 7. The study of the effect of magnesium and fluoride ions on the activity of both forms showed an inhibitory effect of magnesium when its concentration exceeded that of ATP. 8. The inhibition could nevertheless be reverted by including 50 mM NaF in the reaction mixture. 9. Form I and form II could be distinguished by their pH dependence in the presence of an excess of magnesium ions over ATP, whereas the affinity for both substrates was not significantly different.

INTRODUCTION Phosphorylase kinase is a regulatory enzyme that catalyzes the transfer of the y-terminal phosphate group of ATP to several protein substrates. Glycogen phosphoryla~, the main substrate, becomes activated by the action of phosphorylase kinase, thus triggering the degradation of tissular glycogen into glucose-lphosphate. Muscle phosphorylase kinase has been extensively studied. The enzyme is formed by four equal groups of subunits, each group being a tetramer of different subunits named c(, /I. y and S (Cohen et al., 1978). Subunits u.and p have been shown to have regulatory properties, the phosphorylation of them by different protein kinases leading to a more active form of the enzyme (Cohen, 1973; Wang et al., 1976). Subunit y on its side carries the catalytic centre of the enzyme (Skuster et a!., 1980; Reimann et al., 1984). Finally, subunit 6 is by all means identical to the calcium binding protein termed calmodulin (Cohen et al., 1978; Grand ef al., 1980), conferring its calcium ion sensitivity to the enzyme (Cohen, 1980). Muscle phosphorylase kinase is one of the largest soluble proteins known, with a mol. wt which ranges between 1,330,OOOand 1,280,OOOdepending on the authors (Cohen, 1973; Hayakawa el al., 1973). The knowledge about the structure of the non-muscular isoenzymes of phosphorylase kinase is by far 443

less complete (Taira et nl., 1982). Among them, the liver form is specially interesting due to the central role that liver glycogen plays in the homeostasis of blood glucose. Phosphorylase kinase has been isolated from liver in several forms, having mol. wts of 35,000 and 200,000 (Ch~sman et al., 1980), 110,000 and 1,300,OOO(Sakai et af., 1979) l,OOO,~O (Doorneweerd et ai., 1982) and 1,300,OOO(Vandenheede et al., 1979; Chrisman et al., 1982). Several authors have pointed out that the use of proteinase inhibitors favors the isolation of the high molecular weight forms, thus suggesting that the small ones were active proteolytic fragments of the native enzyme produced through the action of an endogenous proteinase present in the liver extracts. In the present paper we describe the identification, purification and characterization of two forms of phosphorylase kinase, having mol. wts of 1,260,OOO and 205,000, from rabbit liver. The results confirm the larger molecule as the one being initialfy present in the liver, and that the small form is produced during the handling of the extract in a process that can be blocked by the action of serine protease inhibitors. Evidence is also presented permitting to rule out the involvement of non-covalent forces, as well as disulfide bridges, in the mass loss process. The proteolytic degradation of the native enzyme has been reproduced under controlled conditions using purified

preparations

of phosphorylase

kinase.

Both

JORGE BELETA et al.

444

trypsin and chymotrypsin produced an active form of the enzyme which had a molecular weight similar to that of the smaller form previously found in Liver crude extracts. MATERIALS

AND METHODS

Preparation of ante-pha~phoryia~e kinase

Antibodies against rabbit muscle phosphorylase kinase were raised in hens following the method described by Cohen and Cohen (1973). The obtained serum was brought twice to 50% saturation with ammonium sulfate, the precipitated protein being resuspended, dialyzed and stored frozen.

Materials

Determination of protein

Sepharose gels and Sepharose-heparin were from Pharmacia Biotechnology (Uppsala, Sweden). Ultrogel AcA34 was from LKB (Bromma, Sweden) DEAE-cellulose (DE 52) was obtained from Whatman (U.K.) Radioactive products were purchased from The Radiochemical Centre (Amersham, Buckinghamshire). Electrophoresis reagents were from Bio-Rad (Richmond, Calif.). Proteinases, proteinase inhibitor and molecular weight markers were from Sigma Chemical Co. (St Louis, MO.). r-13’P]ATP was prepared according to Giynn and Chapel1 (1964). Rabbit muscle phosphorylase b was purified as described by Fischer and Krebs (1958). Muscle phosphorylase kinase was isolated from rabbit muscle as described by Cohen (1973). Sepharose-histone was prepared according to Axen et al. (1967) and Cuatrecasas (1970). Sepharose-Blue Dextran was synthesized according to Ryan and Vestling (1974).

Protein was quantified using the method described by Lowry et al. (1951) using bovine serum albumin as standard.

Assay

qf phasphorylase kinase

Liver phosphorylase kinase was assayed using two different methods. Method A was a modifi~tion of the previously described by Vandenheede et al. (1979). The reaction mixture contained in a final volume of 50 ~1: 1mmol/l ATP, 1 mmol/l magnesium acetate, 30 gmol/I rabbit muscle glycogen phosphorylase 6, 50 mmoi/l NaF, 1mmol/I dithiothreitoi and 20 mmolji triethanolamine buffer adjusted to pH 7.5. This mixture, including an appropriate amount of the sample to be assayed, was incubated at 30°C. At the desired time intervals, 20~1 samples were withdrawn and the reaction stopped by mixing them with 20 vol of an ice-cold buffer containing lOmmol/l EDTA, 100 mmol/l NaF and 50 mmol/l 2 (N-morpholino)ethanesulfonic acid buffer (MES) adjusted to pH 6.1. Glycogen phosphorylase a produced by the action of the assayed enzyme was quantified from this mixture according to the method described by Gilboe er al. (1972). One unit of phosphorylase kinase was defined as the amount of enzyme that produces one international unit of phosphoryla~ a per minute under the above specified conditions. Method B was based on the incorporation of “P from 7 -[32P]ATPto phosphorylase b as described by Reimann et al. (1971). The reaction mixture was identical to that used in method A, except that in this case was supplemented with I x lo6 cpm of radioactive compound. Muscle phosphorylase kinase was assayed according to Cohen (1973). Prepararicm

Electrophoresis on polyacrylamide gels

Polyacrylamide gel electrophoresis (PAGE) both in the presence and in the absence of sodium dodecyl sulfate (SDS) was performed according to Shapiro et al. (I 967) and Davis (1964) respectively. For the electrophoresis of samples that had to be assayed after the separation soluble polyacrylamide gels were prepared using N,N’-bisacrylyl cystamine as cross linking agent. The polymerization was performed essentially as described by Hansen ef al. (1980), but using 15 PM riboflavin/u.v. light as catalyst. RESULTS Gel filtration

of liver extracts

An aliquot (5 ml) of the crude extract obtained as described in methods was applied to a Sepharose CL4B column (1.6 x 60 cm) equilibrated with 20 mmol/l t~ethanolamine buffer pH 7.5 containing sucrose, 250 mmol/l 1 mmol/l dithiothreitoi, 50 pmol/I EDTA, 2 gjl sodium azide (buffer B). Phosphorylase kinase activity (method B) was resolved in two distinct peaks (Fig. 1) named form I and form II. By calibrating the column with proteins of known molecular weight, values of 1,150,OOOand 215,000 were obtained for forms I and II respectively. The relative amount of both forms was function of the time elapsed between the obtention of the crude extract and its application to the column.

“I

I

I

of liver extracls

White New Zealand rabbits were injected into the ear marginal vein with 1 ml of a 25% solution of tiopenthal. The anesthetized animal was bled by sectioning the jugular veins and the liver was extracted. After removing the gall bladder the liver was minced and homogenized with 4 vol of 50 mmol/l phosphate buffer pH 7.0 containing 15 mmol/l 2-mercaptoethanol (buffer A). The homogenate was centrifuged at 10,OOOgfor 45 min and the supernatant fluid filtered through two layers of cheese-cloth and glass-wool, All operations were carried out at 4°C. A mixture of proteinase inhibitors, including N-tosyl-I-phenyl-alanine chloromethyl ketone (TPCK) 0.2 mmol/l, N-a-p-Tosyl-Llysine chloromethyl ketone (TLCK) 0.2 mmol/l, phenyl methyl sulfonyl chloride (PMSF) 0.2 mmol/l, benzamidine 0.5 mmol/l, leupeptin 4 pg/ml and antipain 4 pg/ml, was added to the buffers when indicated in the text.

Fraction

number

Fig. 1. Gel filtration on Sepharose CL4B of rabbit liver crude extracts obtained in the absence (m) and in the presence (0) of a mixture of protease inhibitors. The arrow indicates the column void volume. Flow rate was 3 ml/cm2/hr and the eluate was collected in 1.2 ml fractions.

Liver phosphorylase kinase purification

445

When the crude extract was obtained including in the homogenization buffer a mixture of proteinase inhibitors (see Materials and Methods) and a similar gel filtration of the crude extract was carried out, only one peak of phosphorylase kinase activity, corresponding to form I, was obtained (Fig 1). This behavior suggested that form I corresponded with the native enzyme and that form II originated from form I on storage. In order to ascertain this assumption attempts were made to obtain a stable preparation of this form.

13.5

I 80

60

I 100

80

Sepharose CL4B Sepharostheparin

prot.

43 14

2.5

-

2.0

-

1.5

: 2

-

1.0

s

-

0.5

E 5

(u G

Fraction

I 120

140

160

number

Fig. 2. Gel filtration on Sepharose CL4B of the resuspended ultracentrifugation pellet (form I). Optical density at 280 nm (0) and phosphorylase kinase activity (D) are shown. The arrow indicates the column void volume. Flow rate was 2 ml/cm*/hr and the eluate was collected in 10 ml fractions.

~pharose-heparin column packed and equilibrated with buffer B containing 2.7 mol/l glycerol. In addition, both the sample and the column buffer contained 50 @mol/l calcium chloride. The amount of gel used was 2ml of packed resin per mg of protein present in the sample. After the sample application, the column was washed with 3 vol of equilibration buffer containing 40 mmol/l NaCl, the enzyme activity being eluted afterwards with two column volumes of the same buffer containing 300mmol/l NaCl. All chromatographic procedures were carried out at 4°C. The active fraction was then dialyzed overnight against 15vol of buffer B and was concentrated by ultrafiltration using a Millipore CX filter. This final process was completed at 4°C in not more than 36 hr. The active concentrated enzyme fraction was stored frozen at -25°C in small aliquots. Table 1 shows a summary of the results obtained using the described procedure. The figures, which are a mean of three independent preparations, have been adjusted to an initial liver fresh weight of 100 g. The enzyme activity was measured with method A. Further attempts of increasing the enzyme purity were unsuccessful due to the high molecular weight of the protein, which severely reduced the available capacity of most of the ion exchange and affinity supports tested.

Table 1. Summary of the fractionation procedure for the high molecular weight form of rabbit Ever phosphorylase kinase (form I) (me) 11012 3850 1857 467

-

2

The liver extract from one rabbit was obtained as described in Materials and Methods using buffer B containing the above mentioned proteinase inhibitors. The extract was adjusted to pH 5.2 by dropwise addition of 1 mol/l acetic acid in the cold. The sample was immediately centrifuged at 10,OOOg for 30 min at 4°C. The pellet was dissolved in one half of the initial volume with buffer B (adjusted to pH 8.0 and containing the proteinase inhibitors) by means of a Teflon-glass homogenizer. The resuspended pellet was clarified by centrifugation at 50,000 g for 150 min at 4”C, filtered through glass-wool and centrifugated again at 250,OOOg for 180 min at 4°C. Under these conditions a significant amount of the high molecular weight enzyme sedimented. The obtained pellet, which was firmly attached to the tubes, was removed by means of a spatula and transferred to a beaker containing a small amount of buffer B supplemented with the proteinase inhibitors. The sample was dissolved over a period of 2-3 hr by gentle magnetic stirring in the cold. The resuspended pellet was then freed from undissolved material by a short spun at 10,000 g and diluted with the same buffer up to a volume corresponding to one-tenth of the initial crude extract volume. This sample was applied to a Sepharose CL4B column (5.36 x 82cm) equilibrated with buffer B containing the previously mentioned proteinase inhibitors. The enzyme activity was recovered in three peaks (Fig. 2). The first one, which corresponded to the column void volume, had low quantitative importance, and was probably formed by enzyme aggregates produced during the previous ultracentrifugation step. The second majority peak, identified as form I, was the one used for further purification. A third small activity peak, that appeared in some of the preparations, corresponded in its eluting pattern with form II. The pooled form I from the Sepharose CL4B column was supplemented with glycerol up to a concentration of 2.7 mol/l and was applied to a

Total

3.0

$

Purl~cati~~ of a high molecular weight form of phossph~ry~ase kinase {form I)

Fraction Crude extract Acid precipitate 1st ultracentrif. 2nd ultracentrif.

-

Total act. (t-l) 7158 6442 460 1 2011

1326 1207

Sp. act (u/ma) 0.7 1.7 2.5

Yield W)

Purif. (fold)

100 90 64

2.6 3.8

4.3

28

6.6

19 17

47.7 131

31 85

1

JORGE

BELETA

et

al. 35r

-

300 a L

-

250 E 5

-

200

E 5 2

-150

s G

100 +

=

1 I 50

‘.

0

40

60

60

100 Fraction

120

I 140

I 160

I 180

2000

number

Fraction

number

Fig. 3. DEAEcellulose chromatography (form II). Supernatant was applied to the column (1 ml per 1.5 ml of packed gel) and the chromatography run as described in the text at a flow rate of 6.5ml/cm2/hr. The collected fractions were assayed for conductivity (+), absorbance at 280 nm (0) and phosphorylase kinase activity (0) (method B).

Fig. 4. Gel filtration chromatography of form II on Ultrogel AcA 34. The dialyzate from 2 1. crude extract was applied to a 5 x 75 cm column and chromatographed at a flow rate of 3ml/cm2/hr. The collected fractions were assayed for absorbance at 280 nm (0) and phosphorylase kinase activity (method B) (e).

PMr~cat~o~ of rabbit ~~osp~ory~~e kinase (form Zr)

The active dialyzed fraction was applied to a Sepharose-Blue Dextran column equilibrated in the same buffer (1.5 mg protein per ml packed gel). The column was washed with 4 vol of buffer A, 3 vol of 50 mM ATP in buffer A adjusted to pH 7.4 and 12 vol of buffer A. The enzyme was finally eluted with 12 vol of buffer A containing 2 M NaCl and 10% ethyleneglycol. Flow rate was 200 ml/cm 2/hr throughout the elution procedure. The active fraction was immediately dialyzed for 4 hr against buffer A.

The Iiver crude extract prepared as described under

Materials and Methods was adjusted to pH 5.2 by dropwise addition of 1 N acetic acid and centrifuged at 10,OOOg for 30min at 4°C. The precipitate was dissolved in one half of the initial volume with 50 mM phosphate buffer, pH 8.0, 150 mM sucrose, 15 mM 2-mercaptoethanol and centrifuged at 100,OOOg for 120min at 4°C. The obtained supernatant was filtered through glass-wool and applied to a DEAE-cellulose column previously equilibrated in the same buffer. The column was washed using 2 vol of buffer and eluted with 4 vol of a NaCl gradient from 0 to 0.5 M in the same buffer. Phosphorylase kinase eluted as a singie peak (Fig. 3) that was pooled. The pooled DEAE-cellulose fraction was brought to 45% saturation with ammonium sulfate at 4°C and centrifuged at 15,OOOg for 20 min. The obtained pellet was dissolved in a 2% of the crude extract volume with buffer A containing 150 mM NaCl and was dialyzed overnight at 4°C against the same buffer. The dialyzate was clarified by centrifugation and applied to an Ultrogel AcA 34 column equilibrated in the same buffer. Phosphorylase kinase activity eluted in two peaks (Fig. 4), corresponding to the previously described forms I and II (see Fig. 1). Form II was majority when the present protocol was used. The pooled peak II was dialyzed overnight at 4°C against buffer A and applied to a Sepharose-histone column equilibrated in the same buffer. The column was washed with 4~01 of buffer A and eluted with 8 vol of a NaCl gradient from 0 to 0.8 M in buffer A. Phosphorylase kinase eluted as a single peak (Fig. 5) that was pooled and dialyzed for 4 hr at 4°C against buffer A.

0.14r-------

Fraction

-------1

O6

number

Fig. 5. Sepharose-histone chromatography. Pooled peak II from the gel filtration was applied to a column (1 mg protein per ml of packed gel) and chromatographed at a flow rate of 50 ml/cm2/hr. The collected fractions were assayed for absorbance at 280 nm (a), conductivity (+) and phosphorylase kinase activity (a) (method B).

Liver phosphorylase

kinase

purification

447

Table 2. Summary of the fractionation procedure for the low molecular weight form of liver phosphorylaae kinase (fnrm 111

Total prot Fraction

(ma)

Crude extract Acid precipitate 100,000g sup. DEAE-cellulose 45% SAS precip. 1st ACA 34 Sepharose-histone Sepharose-Blue Dextran 2nd ACA 34

11139 4036 1820 223 167 19 2.8 0.6 0.055

The dialyzate was concentrated by adsorption on a small column of DEAE-cellulose equilibrated in buffer B and eluted with the same buffer containing 0.4 M NaCl. The protein rich fraction was dialyzed against 6.7 M glycerol in buffer B. The concentrated fraction was finally resolved by gel filtration on Ultrogel AcA 34 equilibrated in buffer B containing 3.3 M glycerol. A column of 1.6 x 60 cm was used for a fraction coming from 1 kg fresh liver. Flow rate was 2.5 ml/cm*/hr. Phosphorylase kinase eluted in a liver. Flow rate was 2.5 ml/cm*/hr. Phosphorylase kinase eluted in a single peak with a K,, identical to that of form II in the first gel filtration step. The active peak was pooled and concentrated by dialysis against 6.7 M glycerol in buffer B. After concentration the enzyme was stable at -20°C for several weeks. Table 2 shows the extent of purification and yields at different stages of the purification protocol. The data are the mean of three different preparations and have been normalized to an initial liver fresh weight of 100 g in order to ease the comparison with Table 1. The enzyme activity was measured using method A. Purity and physical characterization of the isolated forms Form II showed two bands on PAGE. The first band contained ca 40% of the protein and was not active while the second one contained 60% of the total protein and exhibited phosphorylase kinase activity. This activity was measured in solubilizable gels by cutting out 2 mm slices and dissolving them in buffer B containing 1 M 2-mercaptoethanol. The dissolved slices were assayed for phosphorylase kinase activity using method B. The active band was cut, dissolved as before, and run together with molecular weight markers on polyacrylamide in the presence of SDS. Two polypeptide chains were identified (A and B) with mol. wts of 44,000 and 69,000. The molecular weights of the purified phosphorylase kinase (form II) was determined by calibrating the second Ultrogel AcA34 gel filtration column used during the purification with proteins of known molecular weight. That procedure gave a value of 205,000 for the enzyme. From the obtained data, a tetrameric structure can be proposed for the purified enzyme, which would contain two A subunits and two B subunits. Form I on its side exhibited several bands on SDS-PAGE. Nevertheless, the procedure yielded a stable preparation which retained its molecular

Total act. UJj

Sp. act. (u/me)

4043 3546 2116 1298 1043 586 447 377 193

Yield (%j

0.36 0.88 1.16 5.82 6.23 30.5 159 662 3513

100 88 52 32 26 14.5 11.1 9.3 4.8

Purif. (fold)

I 2.4 3.2 16 17 84 438 1824 9678

weight and kinetic features for at least one week when stored at 4°C. The phosphorylase kinase present in crude extracts, on the contrary, was completely converted into form II after 24 hr at 4°C. When electrophoresed in polyacrylamide gels under non-denaturing conditions the activity of the purified liver enzyme was detected in the same position than the purified muscle isoenzyme. The molecular weight of the purified form I, determined by calibrating a Sepharose CL4B column with proteins of known molecular weight, was found to be 1,260,000, thus confirming the identity of this enzyme with the high molecular weight form previously detected in crude liver extracts. Proteolytic degradation of puriJied form I

An aliquot of purified form I was incubated at 30°C with trypsin at a proteinase to purified enzyme ratio of 1: 1000 by weight. After 20 min the reaction was stopped by addition of Soybean trypsin inhibitor at a final concentration of 4pg/ml. The mixture (2 ml) was applied to a Sepharose CMB column (1.6 x 60 cm) equilibrated with buffer B. The phosphorylase kinase activity was recovered in two peaks (Fig. 6). The first peak eluted with the void volume,

r

O30

4 Fraction

number

Fig. 6. Gel filtration on Sepharose CL6B of the purified form I of liver phosphorylase kinase before (0) and after (0) being treated with trypsin as described in the text. The arrow shows the column void volume. Flow rate was 2 ml/cm2/hr and the eluate was collected in 1.2 ml fractions.

JORGE BELETA et

448

al.

whereas the bulk of the activity was retained in a well

defined peak of 200,000 mol. wt. A control sample, that had not been treated with trypsin, eluted in a single peak excluded from the gel. Reduction in the molecular size was also obtained when the purified phosphorylase kinase was treated with chymotrypsin at a ratio of I:50 by weight as described above for trypsin. The proteolytic activity was in this case stopped by the addition of 0.5 mmol/l TPCK. On the contrary, the use of dissociating agents such as salts (0.4 mol/l NaCl), non-ionic detergents (0.4% Brij-35, Triton X-100) or reducing agents (0.5 mol/l 2-mercaptoethanol), in conditions similar to those described for the proteinases and in the chromatographic buffers, did not cause any modification in the molecular weight of the purified form I of the enzyme. Inhibition by antibodies raised against the muscle &enzyme

Fixed amounts of both liver phosphorylase kinase forms I and II, and of the proteolyti~ally active fragment of the former one were incubated at 30°C with increasing amounts of an antibody preparation obtained as described under Materials and Methods. After 30min the samples were centrifuged and the residual activity in the supernatant was assayed using method B. The same procedure was followed to test the erect of the antibody on a sample of purified skeletal muscle enzyme. A preparation of unspecific antibodies obtained from the serum of a nonimmunized animal showed no effect on the enzyme activities for the range of antibody concentrations used in the experiment. Figure 7 shows that the muscle enzyme could be inhibited down to ~5% of its initial activity by limited amounts of antibody. The high molecular weight form of the liver enzyme, on the contrary, required a substantially higher concentration of anti-

Anti

phosphorytase

kmose (&/mtl

Fig. 7. Inhibition by anti-skeletal muscle phosphorylase kinase of rabbit phosphorylase kinase. (0) liver form I, (a) liver form I after trypsin treatment and form II, (0) skeletal muscle enzyme.

0--L----0 5

15

20

Mognesium acetate

10

(mM1

25

Fig. 8. Effect of magnesium ions on the activity of purified form II. Enzyme activity was assayed according to method A in the presence (0) and in the absence (0) of 50mM NaF. body to half its activity. The same antibodies preparation showed even less affinity for the low molecular liver forms as reflected by the slight slope of the inhibition curve. Kinetic properties of the puriJied enzymes

At a fixed ATP concentration of 1 mM, magnesium ion was inhibitory for both enzyme forms when added at concentrations >2 mM to the reaction mixture. The presence of 50mM sodium fluoride greatly reduced this Mg2 + inhibitory effect. Fluoride ion itself inhibited the enzymes when its final concentration was higher than 50 mM. Figures 8 and 9 show

0

I

I

50

100

. I 150

200

250

(mM) Fig. 9. Effect of fluoride ions on the activity of purified form fI. Enzyme activity was assayed according to method A. M$+ concentration in the assay mixture was 1mM (0) and 10 mM (a). NoF

Liver phosphorylase kinase purification

449

TabIe3. ApparentMichael% constantsfor the purified liver phosph~~la~ Substrate ATP Phosphorylase

’

5.0

5.4

’

5.9

’

6.2

’

6.6

’

7.0

’

7.4

’

7.6

’

6.2

’

9.6

1

9.0

PH

Fig. 10. Effect of pH on the activity of purified form I. Phosphorylase kinase activity was determined according to method A except in the buffer that was as described in the text. Mg’+ concentration in the assay mixture was 1mM (0) and IOmM (0).

the described effects for form II. Similar results were obtained with form I (not shown). The effect of pH on the purified liver phosphorylase kinase forms I and II was assayed with 30mM triethanolamine-30 mM MES buffer adjusted to different pH values. As can be seen in Figs 10 and 11, the profile obtained for both forms when ATP and magnesium ions were in an equimolar ratio in the reaction mixture was different to that obtained in the presence of excess magnesium (10mM). Maximum

. I

I

I

I

I

I

I

I

I

I

5.5

6.0

6.5

7.0

7.5

8.0

6.5

9.0

9.5

1

I

PH

Fig, 11. Effect of pH on the activity of purified form II. Phosphorylase kinase activity was determined according to method A except in the buffer that was as described in the text. Mg*+ concentration in the assay mixture was 1 mM (0) and 10mM (0).

b

kinase forms (PM) Form I

Form II

65.2

66 4.6

1.78

activity was obtained at pH 7.5 and equimolar ATP/MgZf for the proteolized and native forms of the enzyme, whereas in the presence of an excess of magnesium ions the activity was reduced and the pH optima shifted towards acidic pH for form II and towards alkaline pH for form I. Al1 the activity measurements were made in the presence of 50 mM NaF to compensate for the described inhibitory effect of free magnesium ions. The effect of substrate concentration on the activity of the purified enzymes was determined using method A. It was found that excess ATP over Mg2+ concentration strongly inhibited phosphorylase kinase activity. To overcome this circumstance the K,,, for ATP was determined in the presence of 10 mM magnesium ion and 50 mM NaF. By using this approach the K,,, values for ATP obtained with both phosphorylase kinase forms were very similar (see Table 3). When the concentration of phosphorylase b was varied keeping ATP at 1 mM, a K,,, value for the proteic substrate of 4.6 FM was obtained for form II, assuming a mol. wt of 194,800 for the phosphorylase dimer (Koide et al., 1978). The value obtained for the native enzyme (form I) when assayed under the same conditions was slightly lower, 1.78 PM (Table 3). DISCUSSION Several authors have isolated forms of phosphorylase kinase both from rabbit and rat liver differing in their molecular weight. Of those, Chrisman et al. (1980) reported a high molecular weight form (> 106) as the sole species present in rabbit liver crude extracts, but upon purification only low molecutar weight forms (co.2 x 106) were obtained. By using proteinase inhibitors both Chrisman et ai. (1982) and Vandenheede et al. (1979) were able to purify high molecular weight forms from rat liver, but no low molecular weight form similar to those described on rabbit extracts were found. Conversely, forms of 35,000 (Chrisman et al., 1980) 110,000 (Sakai et al., 1979) and 200,000 (Chrisman et al., 1980) have been described for rabbit liver. We have purified the low molecular weight form present in liver crude extracts obtained in the absence of proteinase inhibitors (form II) and found a mass of 200,000. The analysis on the molecular structure of the purified enzyme reveals that it is formed by two subunits of &i, 69,000 and two other subunits of 44,000. The smaller subunits are identicaf in size to the y subunit of the muscle isoenzyme, which carries the active site of the enzyme (Skuster et al., 1980; Reimann et al., 1984) while the larger subunits of M, 69,000 are probably proteolytic fragments of the native regulatory subunits. Assuming that the native liver phosphorylase kinase has a molecular substructure similar to that described for the muscle isoenzyme (Chrisman et al., 1982) the proteolytic process also involves separation of subunits since the native

450

JORCEBELETA et al.

enzyme (form I) would have four catalytic subunits, whereas only two remain on the proteolyzed form, as described on this paper. The high molecular weight form I has been purified following a procedure that maintains its molecular weight. The procedure is simple to perform and can be easily reproduced, while giving a purification factor that can only be substantially improved by using the more elaborate procedure described by Chrisman et al. (1982) for the glycogen associated enzyme for rat liver. By using the specific activity given by these authors for the homogeneous rat enzyme the amount of phosphorylase kinase present in the purified preparation described can be calculated as at least 10% by weight. Finally, this partially purified enzyme can be easily distin~ished from both the skeletal muscle isoenzyme and the low molecular weight forms from liver by its reactivity towards an antibody preparation raised against the muscle isoenzyme. Both purified phosphorylase kinase forms could be inhibited by raising magnesium ion concentration to values higher than those of ATP in the reaction mixture. This inhibition could be reverted by complexation of the free magnesium with excess fluoride ion. In the following paper (Beleta ef al., 1990) evidence is presented suggesting that purified form I is transformed to its phospho~lation-activated state during the assay of its activity, except if very short incubation times are used. Assuming thus that the data presented here reflects mainly the properties of the activated enzyme, the results obtained in this study are compatible with those described by Chrisman et al. (1984) for the phosphorylated high molecular weight phosphorylase kinase from rat liver. However, in contrast with these results, other authors have found an activation of the same enzyme by free magnesium ions (Hashimoto et al., 1984). The data also show that the low molecular weight form of the enzyme still retains the previously described magnesium ion sensitivity characteristic of the native active enzyme, even after losing the possibility of being regulated through a phosphorylationl dephospho~lation mechanism (Beleta et ai., 1990). A similar effect of magnesium ions on a low molecular weight form of phosphorylase kinase had been previously reported by our group for the enzyme isolated from lymphocytes (Gella et al., 1981). The activity profile as a function of pH of both kinases seems to be influenced too by the magnesium ion concentration in the reaction mixture. The profile obtained for the high molecular weight form in the presence of I mM Mg2+ is very similar to the one described by Chrisman et al. (1982) for the purified active enzyme. When an excess of this ion was included in the assay, the activity was reduced for any pH, being higher on the alkaline side of the curve. This behavior sharply contrasts with the results obtained when the same enzyme was treated with either trypsin or chymotrypsin before the assay. In this latter case and in the presence of excess magnesium ion the pH optimum was cu 6.0, as was the case when

purified form II was used. Finally, when the same low molecular weight form was assayed in the presence of 1 mM Mg’+ a curve similar to the one obtained with form I was obtained. It is worth noting that the

catalytic subunit of the muscle isoenzyme of phosphorylase kinase has been reported to have a pH optimum of 8.2 (Skuster et al., 1980), very different to the ones described in this paper for the liver low molecular weight forms. The effect of pH on the activity of the various forms of phosphorylase kinase under different experimental conditions described in this paper can be of assistance when comparing the kinetic and regulatory properties of other described preparations of the enzyme in order to determine its degree of proteolytic degradation. Both purified forms had the same affinity for ATP. The values obtained were also very similar to those found by several other authors for both high and low molecular weight forms of the liver enzyme (Chrisman et al., 1980; Sakai et al., 1979; Doorneweerd et al., 1982; Vandenheede et al., 1979). and for the catalytic subunit of the muscle isoenzyme (Skuster et al., 1980). These data seem to indicate that the ATP recognition site is not affected neither by the presence of regulatory subunits nor by the activation state of the enzyme. The K,,, for phosphorylase b was 2.5fold higher for the low molecular weight forms than for the native enzyme, reflecting perhaps the involvement of regulatory sites in the recognition of the proteic substrate. Values in the same range (from 1.5 to 9.0 PM) have been published by other authors (Chrisman et al., 1980; Sakai et ai., 1979; Gella et uf., 1981; Vandenheede et al., 1977) for several non-muscular forms of the enzyme. In contrast, the muscle isoenzyme has been shown to have a much lower affinity for the same substrate (Krebs et al., 1964). Overall, the molecular kinetic properties determined in this study for the low molecular weight phosphorylase kinases (form II and proteolyzed form I) show striking similarities with those of low molecular weight forms previously isolated from lymphocytes (Gella er al., 1981), and liver Chrisman et al., 1980; Sakai et al., 1979) and, partially, with the catalytic subunit of the muscle isoenzyme (Skuster et al., 1980). Our results support and extend the belief pointed out by Sakai et al. (1979) that the low molecular weight form found in rabbit liver is an active proteolytic fragment of the native enzyme. This form includes a catalytic subunit probably similar or identical to the catalytic subunit of the muscle isoenzyme. The treatment of the purified high molecular weight form with a range of dissociating agents did not produce any modification of the molecular weight of the enzyme. On the contrary, a low molecular weight form has been obtained for the first time under controlled conditions by limited proteolysis of the purified enzyme either with trypsin or chymotrypsin. REFERENCES

Axen R., Porath J. and Ernback S. (1967) Chemical coupling of peptides and proteins to polysaccharides by means of cyanogen halides. Nature 214, 1302-1304. Beleta J., Benedict0 P. and Gella F. J. (1990) Regulatory properties of rabbit liver phosphorylase kinase. Int. J. Biochem. 22, 453-460.

Chrisman T. D., Vandenheede J. R., Khandelwal R. L., Gella F. J., IJpton J. D. and Krebs E. G. (1980) Purification and regulatory properties of liver phosphorylase kinase. Adv Enzym. Reg. 18, 145-159.

Liver phosphorylase kinase purification Chrisman T. D., Jordan J. E. and Exton J. H. (1982) Purification of rat liver phosphorylase kinase. J. biol. Chem. 257, 10,798-10,804.

Chrisman T. D., Sobo G. E. and Exton E. H. (1984) The magnesium requirements of nonactivated and activated rat liver phosphorylase kinase. FEBS Lert. 167, 295 300. Cohen P. (1973) The subunit structure of rabbit skeletalmuscle phosphorylase kinase, and the molecular basis of its activation reactions. Eur. J. Biochem. 34, l--14. Cohen P. (1980) The role of calcium ions, calmodulin and troponin in the regulation of phosphorylase kinase from rabbit skeletal muscle. Eur. J. Biochem. 111, 563-574. Cohen P. T. W. and Cohen P. (1973) Skeletal muscle phosphorylase kinase deficiency: detection of a protein lacking any activity in ICR/IAn mice. FEBS Let?. 29, 113-116. Cohen P., Burchell A., Foulkes J. G. and Cohen P. T. W. (1978) Identification of the Ca-dependent modulator protein as the fourth subunit of rabbit skeletal muscle phosphorylase kinase. FEBS Left. 92, 287-293. Cuatrecasas P. (1970) Protein purification by affinity chromatography. J. b&l. Chem. i45, 3059-3065. _ Davis B. J. (1964) . , Disc electrouhoresis II. Method and application to human serum proteins. Ann. N. Y. Acad. Sci. 121, 404427.

Doorneweerd D. D., Tan A. W. H. and Nuttall F. Q. (1982) Liver phosphorylase kinase: characterization of two interconvertible forms and partial purification of phosphorylase kinase a. Mol. Cell. Biochem. 47, 45-53. Fischer E. H. and Krebs E. G. (1958) The isolation and crystallization of rabbit skeletal muscle phosphorylase b. J. biol. Chem. 231, 65-71. Gella F. J., Beleta J. and Concustell E. (1981) Lymphocyte phosphorylase kinase. Inr. J. Biochem. 13, 1241-1246. Gilboe D. P., Larson K. L. and Nuttall F. Q, (1972) Radioactive method for the assay of glycogen phosphorylases. Analyt. Biochem. 47, 20-27. Glynn I. M. and Chapel1 J. B. (1964) Simple method for the preparation of “P-1abelled adenosine triphosphate of high specific activity Biochem. J. 90, 147-149. Grand R. J. A., Schenolikar S. and Cohen P. (1981) The aminoacid sequence of the 6 subunit (calmodulin) or rabbit skeletal muscle phosphorylase kinase. Eur. J. Biochem.

113, 359-367.

Hansen J. N., Pheiffer B. H. and Boehnert J. A. (1980) Chemical and electrophoretic properties of solubilizable disulfide gels. Analyt. Biochem. 105, 192-201. Hashimoto E., Mizuta K. and Yamamura H. (1984) Effect of free Mg2+ on liver phosphorylase kinase. J. Biochem. %, 299-303.

451

Hayakawa T., Perkins J. P., Walsh, D. A. and Krebs E. G. (1973) Physicochemical properties of rabbit skeletal muscle phosphorylase kinase. Biochemistry 12, 567-573. Koide A.. Titani K.. Ericsson L. H.. Kumar S.. Neurath H. and Walsh K. A.‘(1978) Sequence of the amino-terminal 349 residues of rabbit muscle glycogen phosphorylase including the sites of covalent and allosteric control. Biochemistry

17, 5657-5672.

Krebs E. G.,.Love D. S., Bratvold G. E., Trayser K. A., Mever W. L. and Fischer E. H. (1964) Purification and properties of rabbit skeletal muscle ’ phosphorylase b kinase. Biochemistry 3, 1022-1033. Lowry 0. H., Rosenbough N. J., Farr A, L. and Randall R. L. (1951) Protein measurement with the Folin phenol reagent. J. biol. Chem. 193, 265-275. Reimann E. M., Walsh D. A. and Krebs E. G. (1971) Purification and properties of rabbit skeletal muscle adenosine-3,5’ monophosphate dependent protein kinases. J. biol. Chem. 246, 19861995. Reimann E. M., Titani K., Ericsson L. H., Wade R. D., Fischer E. H. and Walsh D. A. (1984) Homoloev of the subunit of phosphorylase kinase\ with CAMP-dypendent protein kinase. Biochemrstry 23, 41854192. Ryan L. and Vestling C. S. (1974) Rapid purification of lactate dehydrogenase from rat liver and hepatoma: a new approach. Archs Biochem. Biophys. 100, 279-284. Sakai K., Tsumara S., Okimura Y., Yamamura H. and Nishizuka Y. (1979) Liver glycogen phosphorylase kinase. J. biol. Chem..254, 66311-6637. _ _ _ Shaniro A. L.. Vinuela E. and Maize1 J. V. (1967) Molecular \ weight estimation of polypeptide chains by electrophoresis in SDS-polyacrylamide gels. Biochem. biophys. Res. I

Commun.

28, 815-820.

Skuster J. R., Chan J. K. F. and Graves D. J. (1980) Isolation and properties of the catalytically active subunit of phosphorylase b kinase. J. biol. Chem. 255,2203-2210. Taira T., Kii R., Sakai K., Tabuchi H.. Takimoto S., Nakamura S., Takahashi J., Hashimoto E., Yamamura H. and Nishizuka Y. (1982) Comparison of glycogen phosphorylase kinases of various rat tissues. J. Biochem. 91, 883-888.

Vandenheede J. R., Keppens S. and De Wulf H. (1977) Inactivation and reactivation of liver ohowhorvlase b kinase. Biochim. biophys. Acta 481, 463270: _ Vandenheede J. R.. De Wulf H. and Merlewede W. (1979) Liver phosphorylase b kinase. Cyclic-AMP mediated activation and properties of the partially purified rat-liver enzyme. Eur. J. Biochem. 101, 51-58. Wang J. H., Stoll J. T., Huang T. S. and Krebs E. G. (1976) A study on the autoactivation of rabbit muscle phosphorylase kinase. J. biol. Chem. 251, 45214527.