Isolation and Characterization of the Nervous System-specific ...

TEE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 250, No. 5, Issue of March 10, PP. 1884-1891, 1975

Printed in U.S.A.

Isolation and Characterization of the Nervous System-specific

Protein 14-3-Z from Rat Brain

PURIFICATION, SUBUNIT COMPOSITION, AND COMI’ARISON TO THE BEEP BRAIN PROTEIN

(Received for publication, July 30, 1974)

PAUL J. R~ARANGOS, CLAIRE ZOMZELY-NEURATH, DANIEL C. M. LUK, AND CURTIS YORK

From the Rode Institute of Molecular Biology, Nutley, New Jersey 07110

SUMMARY

A procedure is described for the isolation of the nervous system-specific protein designated 14-3-2 from rat brain. The methods utilized are salt precipitation, DEAE-cellulose ion exchange chromatography, Sephadex G-150 gel filtration, and column isoelectric focusing. The native 14-3-2 protein has an isoelectric point of 4.7 in the absence of denaturing agents and 5.0 in the presence of 2.0 M urea. The protein, as isolated, appears homogeneous since it migrates as a single band on Tris-glycine (pH 8.9), sodium dodecyl sulfate (pH 7.2), and 8 M urea (pH 4.0) polyacrylamide gels. Sedi- mentation velocity and equilibrium data indicate a homogeneous component of molecular weight 78,000. Sedimen- tation of 14-3-2 in 6 M guanidine HCI containing 0.02% glutathione yielded a molecular weight of 39,000, indicating the dimeric nature of the protein as isolated. The rat brain protein seems to be composed of one subunit type, since polyacrylamide gel electrophoresis in 8 M urea yields a single protein component. Sodium dodecyl sulfate gel electrophoresis of rat brain 14-3-2 produced one sharp band with a relative mobility corresponding to a molecular weight of 48,000. Specific anti-14-3-2 serum has been prepared from both New Zealand white rabbits and goats.

Rat 14-3-2 is very similar in amino acid composition to the beef brain protein and to antigen CX. The antigenic properties of rat and beef 14-3-2 are also similar, since beef 14-3-2 antiserum reacts well with rat 14-3-2 and vice versa. Elec- trophoretic mobilities of denatured rat and beef 14-3-2 (0.1% sodium dodecyl sulfate and 8 M urea) are identical. Despite these similarities the two proteins are completely resolved on Tris-glycine gels. The sedimentation behavior of the beef and rat proteins are also different, indicating a difference in the association state and conformation of the two preparations.

The existence of tissue-specific proteins is a predictable con- sequence of cell differentiation, since it can be expected that cell types performing specific physiological functions would have to possess unique proteins. Recently a number of investigators

have reported on the existence and isolation of several nervous system-specific proteins (l-4). Several reviews discussing the localization and properties of these proteins have appeared recently (5, 6). The functions of the nervous system proteins isolated to date remain largely unknown, with the exception of the enzymes required for the biosynthesis and degradation of neurotransmitters and structural proteins such as myelin-specific proteins.

Two especially intriguing proteins that have been isolated from brain are the S-100 and 14-3-2 proteins (l-4). Both are soluble acidic proteins. Several lines of evidence indicate that the S-100 protein is probably localized in glial cells and the 14-3-2 in neurons (7). Both proteins have been implicated as playing a key role in nervous tissue function, since their levels increase dramatically coincident with the functional maturation of the nervous system (8). The S-100 protein from beef brain has been well characterized physically and chemically. The protein is highly acidic, has a molecular weight of 21,000 and is composed of three nonidentical subunits (9, 10). It is affected specifically by calcium, which causes it to undergo a dramatic conformational change giving rise to several st,able forms of the protein (11). Functionally the S-100 protein has been implicated in the leariling process by virtue of the fact that it increases in quantity during learning and the injection of anti-S-100 serum interferes with the learning process (12). The specific function of S-100 is not as yet known.

The 14-3-2 protein has been isolated from beef brain and a molecular weight of approsimately 50,000 reported (13, 14). This protein is also highly acidic, but few data on its physical and chemical properties arc available. Ucnnctt and Edelman have isolated a protein frorn rat brain which they called antigen a (1). This protciii is probably similar, if not identical with 14-3-2, since it reacts with beef 14-3-2 antiserum and has a similar rlectrophoretic mobility (15). The purification procedure described for antigen cy yields a mixture of what appears to be aggregation states of the protein, making it difficult to charactcr- ize the product physically and chemically.

Upon attempting to prepare 14-3-2 from rat brain we found it necessary to develop a purification scheme that would consistently yield a homogeneous sample of 14-3-2. It was also necessary to obtain the product in reasonable yields, since the amount of tissue available from rats is limited. The following report describes a purification procedure for rat brain 14-3-2 that con-

1884

by guest on April 13, 2018

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

1885

sistently yields a homogeneous product as determined by elcc- applied to each gel. The cathode was always situated at the

trophorctic and sedimentation results. Physical and chemical bottom of the electrophoretic apparatus in these experiments.

properties of the isolated 14-3-2 are also reported. It is antici- Sodium dodecyl sulfate gels were performed according to the

pated that knowlctlgc of the structure of this protein will facili- procedure of Weber and Osborn (19). The protein samples were incubated in a solution containing lgj; sodium dodecyl sulfate,

tatc the elucidation of its function in the nervous system. 1% p-mercaptoethanol, 0.01 M sodium phosphate, pH 7.2, at 50” for 1 hour. The samples were then adjusted to a 10% sucrose and

MATERIALS ASD MISTHODS

Protei?~ Dctermi?latio7z-The concentration of pure 14-3-2 fractions was determined spectrophotometrically after determining that the E$p was equal to 0.5. Preliminary identification of 14-3-2 fractions was accomplished by reaction against beef anti- 14-3-2 serum on Ouchterlony plates. The antiserum and beef protein were a kind gift from Dr. 13. Moore (Washington Uni- versity School of Medicine, St. Louis, MO).

DEAE-ccll~closc Colrcmrc.y-The column routinely used was 2.6 X 70 cm. I>EAE-52 (Whatman oreswollen) was oreoared bv washing in 0.5 N HCl and 0.5 N NaOH according to the procedure recommended by the manufacturer. The final equilibration buffer was 10 rnM Tris-phosphate, pH 7.5. About 2 to 4 g of protein were usiially applied to each column in a volume of 50 ml or less. All columns were run at 4” with a flow rate of 30 ml per hour. The sample was applied and 3 column volumes of 0.15 M

NaCl in 10 mM Tris-phosphate, pH 7.5, was run through the column followed by a linear gradient of 0.15 to 0.35 M NaCl in 10 mM Tris-phosphate, pH 7.5, with 900 ml of each solution in the rc- spective gradient reservoirs. The column eluent was monitored at 280 and 260 nm with an LKB Uvicord monitor.

Sephadez Columns-Scphadex G-150, superfine (Pharmacia), was used in columns (1.5 X 100 and 1.5 X 20 cm). The buffer used was 10 mM Tris-phosphate, pH 7.5.

I’reparation of Antisera-Antisera were prepared from both New Zealand white rabbits and goats. In each case 1 ml of a 1 mg per ml solution of rat brain 14-3-2 in isotonic (NaCl) Tris phosphate (10 rnM, pH 7.5) was mixed with 1 ml of Freund’s complete adjuvant (Difcoj and emulsified. This emulsion was injected intradermally in quantities of 0.5 ml per injection site. Injections were given once every 3 weeks with the antibody appearing after the second or third inoculation. The serum was tested weekly on Ouchterlony double diffusion plates (Cordis Laboratories) using pure rat 14-3-2 as antigen. The rabbits were bled from the ear vein and 10 to 50 ml of blood were obtained at each bleeding. The goats were bled from the jugular vein. About 500 ml of blood were obtained from the goats during periods of peak antibody titre. After low speed centrifugation of the clotted blood, the serum was centrifuged at 100,000 X g for 1 hour at 4’. The lipid layer which had been concentrated at the top of the centrifuged serum was removed and the antibody-containing serum stored in small volumes at -20”.

Isoelectric Focusing-Electrofocusing was performed in a 440. ml Ampholine column (LKB) using the LKB gradient maker. The top electrode was the cathode (0.25 N NaOH) and the bottom electrode the anode (lcj, phosphoric acid). Both electrode solutions contained urea (2 M). The pH 4 to 6 carrier ampholytes were used routinely and both the light and dense solutions were adjusted to a concentration of 2 M urea (Mann ultrapure). p- Mercaptoethanol was added (0.257,) to both the light and heavy solution. One-half of the Amuholine solution (10 ml) was added \ I to the light and heavy solutions, yielding a final Ampholine concentration of 2cj,. The sample solution was also divided equally among the light and heavy solutions. The columns were run for 72 hours with a final potential of 600 volts at a temperature of 8”. Columns were drained at a rate of 1 to 2 ml per min through a LKB Uvicord monitor reading at 260 and 280 nm.

Polyacrylamidc Gel Electrophoresis-Disc gels were run in a Buchler water-cooled apparatus using tubes (0.5 X 13 cm). All gel reagents were electrophoresis grade. Tris-glycine gels were run at pH 9.0 according to the method of Ornstein and Davis (16. 17). The gel concentration was 7.5% and they were run at 3 ma per tube. Protein staining was accomplished in 0.5yo Amido black (7% acetic acid) for 30 min. The procedure of Wrav and Stubblefield (18) was employed in the preparation and runnmg of the 8 M urea gels. The gel pH was 4.0 and electrophoresis performed at 3 ma per tube for 4 hours. The sample (0.2 mg per ml) was incubated in an 8 M urea solution containing 1% &mercaptoethanol at 50” for 1 hour. Fifty microliters of this solution were

O.OOZoi, bromphenol blue concentration. To determine molecular weight, 5 or 10 /*g of each protein were treated electrophoretically on each gel. Molecular weights were calculated by running each marker and the unknown on separate gels as well as on the same gel.

Sulfhydryl Group Determination-The determination of total and exposed sulfhydryl groups was performed using 5,5’-dithiobis(2-nitrobenzoic acid) obtained from Sigma. The procedure used was that of Ellman (20, 21). Total sulfhydryl content was determined in sodium dodecyl sulfate at pH 8.0 and exposed groups were assayed in 0.1 M sodium phosphate, pH 8.0, and 0.1 M Tris- phosphate buffer, pH 7.6.

Analytical UItracentrifugation-All experiments were performed in a Beckman model E ultracentrifuge equipped with ultraviolet optics, a photoelectric scanner, and RTIC unit. All scans were made at 280 nm (from monochrometer). A partial specific volume of 0.73 cc per g was used in calculating the 14-3-2 molecular weight. Protein concentration was 0.3 mg per ml in a solution containing 0.01 or 0.10 M Tris-phosphate, pH 7.5, and 0.5 hr NaCl. Both sedimentation velocity and low speed sedimentation equilibrium methods were employed. After all sedimentation equilibrium experiments, the rotor was accelerated to maximum speed overnight, then decelerated to the equilibrium speed, and scans were made to establish the base-line.

Amixo Acid A&y&s-Samples (50 pg) were hydrolyzed with 1 ml of 5.7 N HCl in an evacuated and sealed tube at 110” for 22 hours. The hydrolysate was analyzed according to the method of Spackman et al. (22) with a Jeol model 6AH automatic amino acid analyzer. For the determination of tryptophan, sampleswere hydrolyzed with 0.2 ml of 4 M methane sulfonic acid containing 0.27, tryptamine (23).

RESULTS

Rat Brain 14-M? Purificalion

Extraction-Rats (Charles River) were injected with 1 ml of

Nembutal (10 mg per ml) and killed by exsanguination via the abdominal aorta. The whole brain was removed and a 25% homogenate prepared in 10 mM Tris-phosphate, pH 7.5, using a Teflonglass homogenizer. One hundred brains were usually prepared each day (150 g of tissue). After centrifugation of the homogenate at 12,000 rpm for 10 min at. 4” in a Sorvall RC 2B centrifuge, the resulting post-mitochondrial supernatant was centrifuged at 100,000 x g for 1 hour at 4”. The high speed supernatant obtained was used for the isolation of the 14-3-2 protein and is referred to as the crude soluble fraction.

Salt Precipitation-The crude soluble fraction was brought to 400/, (NH&SO, (Mann, ultrapure) saturation by slow addition of the solid. This solution was stirred for 45 min after the (NHJ)*S04 dissolved. The resulting suspension was centrifuged at 15,000 rpm for 20 min at 4”, and the supernatant decanted. The 40% (NHJzS04 supernatant fraction was brought to 60% (NH,)804 and stirred for 45 min. This solution was centrifuged at 15,000 rpm for 20 min. The supernatant was discarded and the 60% (NH&SO4 pellets were resuspended in 10 mM Tris-phosphate, pH 7.5. This solution was then dialyzed against 100 volumes of buffer with one change. The resulting solution was centrifu ed at 15,000 rpm for 20 min and the supernatant decanted. 3 his solution is referred to as the P-60 fraction.

DEAE-cellulose Chromatography-The P-60 fractions from 400 rats (about 600 g of tissue) were pooled and concentrated to 50 ml or less with an Amicon ultrafiltration apparatus using a PM-10 membrane. This solution was centrifuged at 15,000 rpm for 15 min. The supernatant usually contained 3 to 4 g of pro-


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

1886

14-3-2

0 IO 20 30 40 50 60 70 SO 90 100 110 120 130

FRACTION NUMBER

FIQ. 1. DEAE-cellulose chromatography of rat brain P-60 fraction. Approximately 3 g of the P-60 fraction in 50 ml of Tris-phosphate buffer were applied on a column (2.6 X 70 cm). The flow rate was adjusted to 30 ml per hour. Fraction volume was 8 ml. The 14-3-2-containing fraction was identified using antiserum after each peak was pooled and dialyzed.

20 40 60 80 100 120 140

FRACTION NUMBER

FIG. 3. Isoelectric focusing of the 14-3-2-containing Sephadex fraction. Isoelectric focusing was performed in the pH 4 to 6 range at a 2% Ampholine concentration in the 440-ml column. An LKB linear gradient maker was used to prepare the sucrose gradient (5 to 47%). The sample contained 80 A280 units of protein in 15 ml of Tris-phosphate buffer (5 mM). The column was drained at a rate of 50 ml per hour into 2-ml fractions.

14-3-Z

3 2.0

s (Y

2 1.0 .!,;;;,I~ light solution 5% sucrose. Slow addition of the sample to the Ampholine solutions (room temperature) avoided any precipitation. During the first 12 hours of the run a precipitation ring appeared at the top of the gradient. Initially this precipitate was withdrawn (with current off) with a capillary tube connected to a syringe, after 18 hours of running. Addi- tional cathode solution was then layered over the upper electrode to compensate for the lost volume. However, with- drawal of the precipitate was found to be unnecessary, as it remained at the top of the column and did not affect band res-

0 IO 20 30 40 50

FRACTION NUMBER olution. The column was run at room temperature with cold

FIG. 2. Sephadex G-150 chromatography of the 14-3-2-contain- (8”) water flowing. After 72 hours the final stabilized potential

ing DEAE-fraction. The sample (120 mg) was applied to a col- was 600 to 700 volts at 4 to 5 ma. Fractions collected from the

umn (1.5 X 100 cm) in a volume of 2 ml. The flow rate was ad- column were monitored by an LKB Uvicord, consistently pro-

justed to 2 ml per hour. ducing the A2s0 trace which is illustrated in Fig. 3. The isoelectric point of the rat brain 14-3-2 in 2 M urea is 5.1 & 0.1.

tein. The above solution was applied on a DEAE-52 column (2.6 x 70 cm) and eluted as described under “Materials and Methods.” A flow rate of 30 ml per hour was found to give the best resolution. Fig. 1 illustrates the Azso tracing consistently obtained. The peak labeled 14-3-2 was identified initially by its reaction with beef 14-3-2 antiserum. The fractions corresponding to this peak were isolated, pooled, and dialyzed against 50 volumes of double distilled water with three changes. This solution was then lyophilized. This fraction usually contained 100 to 150 mg of protein.

Gel F&a&m-The lyophilized DEAE-fraction was resuspended in 2 ml of 10 mM Tris-phosphate, pH 7.5, and applied to a column (1.5 X 100 cm) packed with Sephadex G-150 (superfine). The column eluate was passed through the Uvicord monitor (280 nm) with a typical tracing shown in Fig. 2. The second larger peak contained the 14-3-2. The fractions corresponding to this peak were pooled, and usually contained about 80 mg of protein.

1soeZectric Focusing-One-half of the sample obtained from Sephadex G-150 chromatography was added slowly to the light and heavy Ampholine solutions. Each Ampholine solution contained 2% pH 4 to 6 ampholytes, 2 M urea, and 0.25% P-mercaptoethanol. The heavy solution contained 479” sucrose and the

The isoelectric point in the absence of urea was lower, with the 14-3-2 peak usually appearing at pH 4.7. Focusing in the absence of urea produced very low yields (1 to 3 mg of 14-3-2) whereas incorporation of 2 M urea apparently increases the solu- bility of 14-3-2 and raised the yield to 15 to 20 mg per run. The omission of /3-mercaptoethanol also decreased the yield to about 5 mg per run.

The fractions corresponding to 14-3-2 were pooled and dialyzed against 100 volumes (three changes) of 10 mM Tris-phosphate, pH 7.5. All of the ampholytes were not removed after dialysis, since a very rapidly migrating band was observed on polyacrylamide electrophoresis of the dialyzed focusing fraction. This band was not observed in the sample applied on the electrofocusing column. Complete removal of the nondialyzable ampholyte was effected by lyophilizing the above fraction, re- suspending it in 1 ml of 10 mM Tris-phosphate, pH 7.5, and applying it to a Sephadex G-150 (superfine) column (1.5 X 20 cm). The sharp symmetrical peak obtained at the expected elution volume was pooled and divided into 250~~1 aliquots and stored at ,-20”. Routine yields of pure 14-3-2 ranged from 15 to 20 mg. The preparation was stable for at least 3 months when stored at -20”. Table I summarizes the purification procedure. The value of 0.14% for the amount of 14-3-2 obtained represents the quantity obtained in the purification and


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

1887

TABLE I Summary of rat brain l.J-s-2 purijication scheme

The protein content of each fraction was estimated by assum- ing that the Ei.$ was equal to 1.0 for each crude fraction. The Ei$ was determined to be 0.5 for the pure 14-3-2 fraction.

SOLUBLE FRACTION

40 10 60% (NH4)2SO4 FRACTION

DEAE CELLULOSE FRACTION

SEPHADEX G-150 FRACTION

ISOELECTRIC FOCUSING FRACTION

PASSED THRU G-150

VOLUME PROTEIN ml mg

1200 13,200

250 3750

120 125

5 100

10 20

PER CENT SOLUBLE PRO1

100

28

0 90

0 75

014

FIG. 4. Polyacrylamide gel electrophoresis patterns of various fractions in the 14-3-2 purification procedure. The gels were 7.5$& polyacrylamide Tris-glycine at pH 8.9 and were run at 3 ma per tube for 3 hours. The samples from left to right contain: 0.1 mg of crude soluble fraction, 0.1 mg of P-60 fraction, 0.1 mg of DEAE fraction, 10 and 50 pg of the product. The gel containing 50 pg of 14-3-2 was obtained from a separate experiment.

t -1.5

2 - -1.7

-1.9 1

-2.1 1 -2.3 /-

:::FL_,J 44 46 48 50 52

r2 FIG. 5. Sedimentation equilibrium of native rat brain 14-3-2

(In A versus r* plot). Rat brain 14-3-2 (0.3 mg per ml) was centrifuged at 10,000 rpm for 24 hours at 20”. The protein solution contained 10 mM Tris-phosphate, pH 7.5, and 0.5 M NaCl. The term In A refers to the natural log of the absorbance at 280 nm.

not the actual per cent of 14-3-2 present in the cell; losses occur during the procedure. In Fig. 4, the 7.5% polyacrylamide gel patterns of various fractions during the procedure are shown.

Molecular Weight-The molecular weight of the native 14-3-2 was determined by sedimentation equilibrium. A molecular weight of 78,000 was observed in Tris-NaCl, pH 7.5. The plot of In A versus r2 was a straight line, indicating a homogeneous population of molecules (Fig. 5). Some samples of rat 14-3-2 which were allowed to stand for prolonged periods at room temperature displayed a small degree of polydispersity due to aggregation. The beef brain 14-3-2 was also run under these conditions and yielded a curve plot of In A versus r2, indicating a heterogeneous system. Molecular weight values ranged from 45,000 at the top of the cell to 80,000 at the bottom of the cell. Sedi- mentation velocity experiments indicated a sharp boundary for rat brain 14-3-2 and a broad boundary for the beef protein. The beef protein displayed an S value of 4.8 and the rat protein a value of 5.6.

Subunit Analysis-The rat brain protein was kept in 6 M guanidine hydrochloride containing 0.02% glutathione for 1 hour at 50”. The protein concentration was 0.3 mg per ml. Sedimentation equilibrium was then performed on this sample. The In A versus r2 is shown in Fig. 6. A homogeneous component of molecular weight 39,000 was observed indicating that the native protein was probably a dimer. If glutathione was omitted from the sample a heterogeneous population of molecules was observed with molecular weight values ranging from 40,000 at the top of the cell to 80,000 at t.he bottom. These data indicate that sulfhydryl group-dependent aggregation is occurring, with the monomer and dimer existing in equilibrium.

Sodium dodecyl sulfate polyacrylamide gel electrophoresis of rat brain 14-3-2 using lysozyme, chymotrypsin, ovalbumin, and serum albumin as markers yielded a molecular weight of 48,000, as can be seen in Fig. 7. The rat brain and beef brain 14-3-2 were not resolvable on sodium dodecyl sulfate gels, indicating that they had very similar molecular weights in this system.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

t b -2.944 8 I I I 1

46

FIG. 6. Sedimentation equilibrium of denatured rat brain 14-3-2 (In A versus r2 plot). The sample containing 0.3 mg per ml of rat brain 14-3-2, 6 M guanidine hydrochloride, and 0.020/, glutathione was centrifuged at 18,900 rpm for 70 hours.

80 I , I ( 1 1 ’ 1 ’ 1 ’

7o--\ 0 SERUM ALBUMIN (68,OOOl

60

RELATIVE MOBILITY Amino Acid Composition FIG. 7. Molecular weight determination using 0.1% sodium

dodecyl sulfate polyacrylamide gels. Five micrograms o.f rat 14-3-2 and each marker nrotein were treated electrophoretlcally for 4 hours at 9 ma per tube. Relative mobilities were determined from gels containing markers and 14-3-2 as well as from gels where each protein was run individually.

The amino acid composition of both the rat and beef brain 14-3-2 was determined, with the results shown in Table II. The beef and rat proteins are very similar, both containing a high percentage of acidic amino acids. The amino acid mole per cent for the rat and beef proteins are also similar to those obtained by Bennett and Edelman for antigen (Y (1). Since sulfhy- dry1 groups appeared to be very important for maintenance of the proper quaternary structure of 14-3-2, it was of interest to determine the total number of cysteine residues present in the

1 The abbreviation used is : TEMED, N ,N, N’ , N’-tetramethyl- ethylenediamine; DTNB, 5,5’-dithiobis(2-nitrobenzoic acid).

One sharp band was always obtained for the rat brain 14-3-2 in this gel system, further attesting to the homogeneity of the preparation. The sodium dodecyl sulfate gel results were not recon- cilable with the sedimentation data, which indicated a molecular weight of 39,000. Recent reports have indicated that the

FIG. 8. Subunit analysis of rat 14-3-2 on 0.1% sodium dodecyl sulfate and 8 M urea polyacrylamide gels. Ten micrograms of 14-3-2 were run in each gel system. The sodium dodecyl sulfate gel (left) was run at 9 ma per tube for 4 hours and the 8 M urea gel at 3 ma per tube for 3 hours.

TEMEDr present in sodium dodecyl sulfate gels affects the molecular weight determination of some bram protems (24). Sodium dodecyl sulfate gel electrophoresis was performed using one-tenth the usual amount of TEMED but the 14-3-2 molecular weight remained at 46,000 to 48,000. The molecular weights obtained are in good agreement with those obtained by Bennett and Edelman (15) for antigen 01 and beef brain 14-3-2. This apparent molecular weight 10,000 discrepancy is consistent and appears to be due to anomalous behavior of 14-3-2 on sodium dodecyl sulfate gels. This type of behavior has also been observed with other proteins (25-27).

In an effort to determine whether each of the 14-3-2 subunits had the same net charge, the protein was electrophoretically treated in 8 M urea gels. Fig. 8 shows the pattern obtained for both sodium dodecyl sulfate and 8 M urea gel electrophoresis of 14-3-2. Electrophoresis in 8 M urea at three different pH values (pH 3.2, 3.5, and 4.0) all produced one protein band, indicating that the constituent subunits of the dimer had a similar or identical net charge.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

1889

TABLE II Amino acid composition of rat and beef i&M

Fifty micrograms each of beef and rat 14-3-2 were used for the analysis of tryptophan as well as for the analysis of the remaining amino acids. Data are presented as mole per cent with cysteine being omitted from the calculations.

Amino acids Beef I Rat .-

Lysine ..................... Histidine. ................. Arginine .................. Aspartic acid .............. Threonine ................ Serine ..................... Glutamine ................. Proline .................... Glycine .................. Alanine. Valine ..................... Methionine ................ Isoleucine ................. Leucine ................... Tyrosine .................. Phenylalanine ............. Tryptophan .............

6.13 6.70 1.58 1’.62 4.90 4.47

12.08 11.99 3.85 3.66 4.73 4.67

11.38 11.38 4.03 3.86 9.11 9.35

11.38 11.79 6.83 6.91 1.40 1.02 5.60 5.69 9.63 9.96 2.45 2.23 3.15 3.25 1.75 1.42

molecule. This was done using a spectrophotometric assay employing DTNB. The results indicated that there are 6 cysteine residues per 78,000 molecular weight or 3 residues per subunit. The cysteine determination was also done under non- denaturing conditions to determine the number of exposed sulfhydryl groups in the native conformation. DTNB did not react with native 14-3-2 in the first 60 min of the assay, indicating that little or none of the component sulfhydryl groups were exposed under these conditions.

Rat and beef brain 14-3-2 were similar in several of their properties, such as amino acid composition, antibody cross-reactivity, and mobility in sodium dodecyl sulfate polyacrylamide gels. In order to obtain further information on the degree of similarity between the beef and rat proteins their mobilities were compared on 7.5% polyacrylamide Tris-glycine gels at pH 9.0. Fig. 9 shows the results obtained from such an experiment. It can be seen that the beef protein has a greater electrophoretic mobility than that of the rat. Furthermore, when coelectrophoretically treated both proteins are completely resolved from one another, indicating that they differ significantly in net charge.

Antiserum Preparation

Antisera to rat brain 14-3-2 were prepared from both rabbits and goats. Two injections of 1 mg of protein each were sufficient to elicit antibody formation. The ability of goats to produce a precipitating antibody for 14-3-2 is fortunate, as large volumes of serum and hence large quantities of antibody can be obtained. Fig. 10 shows an Ouchterlony double diffusion plate obtained with the rabbit antiserum in the center well. It can be seen that only one precipitation arc is seen with both pure 14-3-2 (Wells 1

and 2) and the P-60 fraction (Wells S to 6). These patterns indicate that the antisera are specific in their reaction with 14-3-2.

Other preliminary data were obtained concerning the physical and chemical properties of rat brain 14-3-2. The CD spectrum indicates that the protein contains 29% helix. Although Ca2+

FIG. 9. Comparison of beef and rat 14-3-2 on Tris-glycine polyacrylamide gels. The gels from Zefl to right contained 5 pg of beef 14-3-2, 5 pg of rat 14-3-2, and 5 rg each of the rat and beef protein. The gels were run at 3 ma per tube for 3 hours.

FIG. 10. Double immunodiffusion experiment (rat 14-3-2 versus rabbit antiserum). The c&ter well contained 20 ~1 of rabbit antiserum. Wells 1 and 2 contained 3 rg each of pure rat 14-3-2; Wells 3 and 4, 20 ~1 of the P-60 fraction; and Wells 6 and 6 have a 1:3 dilution of the P-60 fraction.


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

1890

has been shown to exert profound effects on the conformation of the S-100 protein (1) no effect was seen on the electrophoretic mobility or the CD spectrum of the 14-3-2 protein. The cation was also without significant effect on the number of exposed sulfhydryl groups as determined by the DTNB method. Collec- tively these findings suggest that Ca2+ does not significantly affect t,he conformation of rat brain 14-3-2. The periodic acid- Schiff staining method (28) (gel electrophoresis) was also performed with negative results, indicat,ing that 14-3-2 contained no carbohydrate moieties. A phosphoproteinstaining procedure was also employed (29), which also gave negative result,s indicating the absence of phosphorus in 14-3-2.

DISCUsSION

dodecyl sulfate electrophoresis which imparts a negative charge to all proteins and discriminates solely on the basis of size. Taken in conjunct,ion with the sedimentation data, it appears that rat brain 14-3-2 is a dimer consisting of identical subunits.

14-3-2 is an excellent antigen in both rabbits and goats. A quantity of 2 mg given in l-mg injections one month apart is sufficient to produce high antibody titres in both the above ani- mals. The results were identical both with and without the in- corporat’ion of methylated serum albumin in the samples used for injection. A precipitation arc was observed on Ouchterlony plates with a 1:50 dilution of the rabbit antiserum and a 1: 10 dilution of the goat serum. Higher titres were obtained with subsequent booster injections in the goat. Since 14-3-2 anti- bodies can be produced in goats, large quantities of blood (500 ml)

A purification procedure has been devised for rat brain 14-3-2 can be obtained, which makes it feasible to prepare the purified utilizing column isoelectric focusing as the final step. The pro- 14-3-2 antibody. Immunological evidence (Fig. 10) supports cedure consistently yields a homogeneous product as judged by the contention t,hat rat brain 14-3-2 was isolated in pure form electrophoresis on three different gel systems (Figs. 4, 7, and 8) because only one precipitin arc was seen on Ouchterlony plates as well as it,s behavior in sedimentation experiment,s (Figs. 5 and when 14-3-2 antiserum was reacted with a crude P-60 fract,ion. 6). The procedure employed by us is different than that used Sulfhydryl group analysis indicated that the protein contained by Bennett and Edelman to purify antigen cy. In their procedure 6 cyst’eine residues per dimer or 3 residues per monomer unit. a pH 5.0 precipitation step was used to precipitate prot,ein other The native protein had no exposed sulfhydryl groups, suggesting than 14-3-2. In our system treatment of any 14-3-2 fraction in that t’he conformation of the molecule was such that the cysteine this manner led to a great loss of the protein. All washed pH residues were not accessible to DTNB. It is possible that di- 5.0 precipitates analyzed contained a dark protein band on sulfide bridges contribute to the subunit int,eraction and are t’hus polyacrylamide gels which corresponded to 14-3-2. These frac- buried in the center of the molecule. This idea is supported by tions were also highly reactive wit,h anti-14-3-2 serum on Ouch- the fact that dimers only appeared in the 6 M guanidine HCl terlony plates. sedimentation equilibrium experiments when a sulfhydryl-re-

Comparison of the product obtained by us and t,hat of Bennett ducing agent, was omitted, indicating that dimer formation was and Edelman also reveals some distinct differences. The rat probably to some degree disulfide-dependent. brain 14-3-2 was homogeneous during sedimentation in a Tris- Rat brain and beef 14-3-2 are very similar, but apparent,ly not NaCl system, displaying a molecular weight of 78,000 while the identical. The antiserum prepared against beef 14-3-2 cross- (Y antigen preparation contained two components of molecular reacts very well with rat 14-3-2 and vice versa, indicating that the weight 45,000 and 72,000. The antigen LY preparation also exists antigenic determinant for the two proteins is probably very as a single component of molecular weight 60,500 in 6.3 M guani- similar. The amino acid compositions of beef and rat 14-3-2 are dine, while we observed a heterogeneous mixture of species under also remarkably similar (Table II). Moreover, t,he mobility of these conditions. The species observed probably represent a 14-3-2 from these two sources on sodium dodecyl sulfate gels and mixture of monomers and dimers. Both preparations yield a 8 M urea gels was identical. Coelectrophoresis of the two dena- single species of 39,000 molecular weight when sedimented in 6 M tured proteins on these gel systems produced one band. These guanidine under sulfhydryl-reducing conditions, indicating that results support the idea that the monomer unit of both beef and both are probably composed of the same basic unit. rat 14-3-2 is structurally very similar. However, when the un-

The 14-3-2 protein as isolated behaves as a stable dimer with denatured beef and rat protein are compared on Tris-glycine gels little tendency to aggregate into high molecular weight species. (Fig. 9) complete resolution of the proteins is achieved. This Sephadex G-150 chromatography of the pure preparation always clearly indicates t,hat beef and rat 14-3-2 have significantly differ- produced a single symmetrical peak, further indicating the lack ent net charges under these conditions. The probable explana- of any significant aggregation. The sedimentation data strongly tion is that one or several amino acids differ between the two pro- support the idea that the protein is a dimer because the dena- teins. This difference is probably reflected in a conformational tured 14-3-2 molecular weight is a multiple of the native molecu- alteration which leads to a difference in the exposed charged lar weight. However, the molecular weight of 48,000 obtained groups on each molecule. The hypothesis of a conformational on sodium dodecyl sulfate polyacrylamide gels is difficult to difference is also supported by sedimentation data which showed explain as it does not fit the sedimentation data. It does not that the beef brain protein had an S value of 4.8 while the rat appear that the TEMED present in the gels is affecting the brain protein had an S value of 5.6. This large difference in S molecular weight determination, as the same results are obtained value between rat and beef 14-3-2 may be misleading, since the when one-tenth the usual amount of TEMED is used. Similar beef protein yielded a broad interface during sedimentation ve- molecular weight results were obtained by Bennett and Edelman locity runs. It should be mentioned that direct comparisons with t’he rat antigen a! and beef 14-3-2 (15). The possibility between beef and rat 14-3-2 should be made with caution because exists that 14-3-2 behaves differently in sodium dodecyl sulfate it appears that the 14-3-2 protein from beef has a greater tendency than other proteins, which would account for the decreased toward nonspecific aggregation than that obtained from the rat. mobility of the protein on sodium dodecyl sulfate gels (25-27). The difference observed in sedimentation and electrophoretic

The results obtained in the 8 M urea polyacrylamide gel ex- patterns may be caused in part by these differences. perimcnts support the idea that rat brain 14-3-2 is composed of The remarkable similarity between beef and rat 14-3-2 that one monomer type. This gel system discriminates denatured has been demonstrated indicates that evolution has conserved proteins according to both size and charge, contrary to sodium this protein to a high degree, since these two species are phylo-


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

genetically removed from one another. This suggests that the specific function of this protein is ubiquitous and is essential to the survival of the organism. In attempting to determine the function of the brain specific protein designated 14-3-2 we have adopted the tactic of initially studying its chemical and physical properties in the anticipation that this would yield information contributing to this goal. Other ongoing studies in our labora- tory attempting to characterize brain specific proteins functionally include investigation of factors regulating the in vitro syn- thesis of 14-3-2 and S-100 (30), as well as the axoplasmic transport of 14-3-2.2 It is our hope that such a broad based attack on the problem of brain specific protein function will be successful and lead to a better understanding of brain function at the molecular level.

Acknowledgments-We express sincere appreciation to Dr. Chunyen Lai and Dr. Miloslav Boublik of The Roche Institute of Molecular Biology for performing the amino acid analysis and CD spectra, respectively.

1.

2.

3.

4.

5.

6.

REFERENCES

BENNETT, G. S., AND EDELMAN, G. M. (1968) J. Biol. Chem. 243, 6234-6241

MOORE, B. W., AND MCGREGOR, D. (1965) J. Biol. Chem. 240, 1647-1653

TANAKA, T., HARANO, Y., MORIMURA, H., ‘ZND MORI, It. (1965) Biochem. Biophys. Res. Commun. 21, 55-60

MOORE, B. W., AND PI;:RI~z, V. J. (1968) in Physiological and Biochemical Aspects of Nervous Integration (CARLSON, F. D., ed) pp. 343-359, Prentice-Hall, Englewood Cliffs, N. J.

SHOOTER, E. M., AND EINSTEIN, E. R. (1971) Annu. Rev. Bio- them. 40, 63H48

SCHNEXDER, D. J., ed (1973) Proteins of the Nervous System, Raven Press, New York

2 Manuscript in preparation.

7.

8.

9.

10.

11.

12.

13.

14. 15. 16. 17. 18.

19.

20. 21. 22.

23.

24.

25.

26.

27. 28.

29.

30.

1891

PEREZ, V. J., OLNISY, J. W., CICERO, T. J., MOORIS, B. W., AND BAHN, B. A. (1970) J. Neurochem. 17, 511-519

CICERO, T. J., COWAN, W. M., AND MOORI,:, B. W. (1970) Brain Res. 24, l-10

DANNIES, P. S., AND LEVINE, L. (1969) Biochem. Biophys. Res. Commun. 37, 587-592

DANNIES, P. S., AND LEVINE, L. (1971) J. BioZ. Chem. 246, 6276-6283

CALISSANO, P., MOORI<:, B. W., AND FRIESICN, A. (1969) Bio- chemistry 8, 4318-4326

HYDICN, H., AND LANGE, P. W. (1970) Exp. Cell Res. 62, 125. 132

MOORE, B. W. (1973) in Proteins of the Nervous System (SCHNF,I- DF:R, D. J., ed) pp. l-12, Raven Press, New York

Moo~tr:, B. W. (1972) Int. Rev. Neurobiol. 16, 215-225 B~;NNETT, G. S. (1974) Bruin Res. 68, 365-369 ORNSTEIN, L. (1964) Ann. N. Y. Acad. Sci. 121, 321-349 DAVIS, B. J. (1964) Ann. N. Y. Acad. Sci. 121, 404-427 WRAY, W., AND STUBBLEFIELD, E. (1970) Anal. Biochem. 38,

454-460 WEBER, K., AND OSBORN, M. (1969) J. BioZ. Chem. 244, 4406-

4412 ELLMAN, G. I,. (1959) Arch. Biochem. Biophys. 82, 7&77 HAB~;I<B, A. F. S. A. (1972) Methods Enzymol. 26, 457-464 SPACKMAN, D. H., STICIN, W. H., AND MOORE, S. (1958) Anal.

Chem. 30, 1190-1196 LIU, T.-Y., AND CHANG. Y. H. (1971) J. BioZ. Chem. 246, 2842-

2848 ALLISON, J. H., AGILBWAL, H. C., AND MOORIS:, B. W. (1974)

Anal. Biochem. 68, 592-601 DUNKER, A. K., AND RUECKERT, R. R. (1969) J. BioZ. Chem.

244, 5074-5080 TUNG, J.-S., AND KNIGHT, C. A. (1971) Biochem. Biophys.

Rrs. Commun. 42, 1117-1121 KATZMAN, R. L. (1972) Biochim. Biophys. Acta 266, 269-272 KAPITANY, R. A., AND ZEBROI~SKI, E. J. (1973) Anal. Biochem.

66, 361-369 GREEN, M. It., PASTETVKA, J. V., AND PE.~COCK, A. C. (1973)

Anal. Biochem. 66, 43351 ZOMZELY-NEURATH, C., YORK, C., AND MOORI,:, B. W. (1973)

Arch. Biochem. Biophys. 166, 58-69


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

P J Marangos, C Zomzely-Neurath, D C Luk and C Yorkprotein.

rat brain. Purification, subunit composition, and comparison to the beef brain Isolation and characterization of the nervous system-specific protein 14-3-2 from

1975, 250:1884-1891.J. Biol. Chem.

http://www.jbc.org/content/250/5/1884Access the most updated version of this article at

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

http://www.jbc.org/content/250/5/1884.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/content/250/5/1884

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;250/5/1884&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/250/5/1884

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=250/5/1884&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/250/5/1884

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/250/5/1884.full.html#ref-list-1

http://www.jbc.org/

Isolation and Characterization of the Nervous System-specific ...

Documents

Transcript of Isolation and Characterization of the Nervous System-specific ...