CDP-choline: 1,2-diacylglycerol cholinephosphotransferase from rat liver microsomes. II. Photoaffinity labeling by radioactive CDP-choline analogs

36 0 1992 Elsevier

BBALIP

Science

Bwchrmrcu ct Biophysiw Actu, I 124 ( 1092) 36-44 Publishers B.V. All rights reserved OOOS-27~0/92/$05.00

53832

CDP-choline: 1,2-diacylglycerol cholinephosphotransferase from rat liver microsomes. II. Photoaffinity labeling by radioactive CDP-choline analogs Kozo Ishidate, The Medicul Research Institute,

(Revised

Key words:

Ritsuko

Matsuo

and Yasuo Nakazawa

Tokyo Medic& und Dentul lJnic,er,rity. Chiyodu-ku, (Received Y July 199 I) manuscript received 29 September

Cholinephosphotransferase;

CDP-choline;

Tokyo (Jupun)

1991)

Photoaffinity

labeling;

(Rat liver)

Photoaffinity labeling of cholinephosphotransferase from rat liver microsomes directly by its substrate, 13*P1CDPcholine or by a synthetic photoreactive CDP-choline analog, 3’(2’)-0-(4-henzoyDhenzoy1 [32P1CDP-choline (BB[32P1CDP-choline), was examined for the possible identification of its molecular form on subsequent SDS-PAGE followed by 32P-autoradiography. When the partially purified cholinephosphotransferase was photoirradiated in the presence of [32P1CDP-choline, a considerable amount of 32P-radioactivity was incorporated into the TCA-insoluble component. This incorporation was dependent on irradiation time, Mg2+ or Mn2+ -requiring and inhibited strongly inhibited the ultraviolet irradiation-dependent by the presence of Ca2 +. Either CDP-choline or CDP-ethanolamine incorporation of 32P-radioactivity into the TCA-insoluble component in a dose-dependent manner, whereas neither phosphocholine or 5’XDP had any effect on this process. These results strongly suggested that the observed 32P-incorporation from [32P1CDP-choline into the protein component could he a consequence of the covalent interaction between cholinephosphotransferase and its substrate, [32P1CDP-choline. Two polypeptides, 25 kDa and 18 kDa, with high 32P-radioactivity were clearly identified on a SDS gel after the direct photoaffhtity labeling with [32P1CDP-choline for more than 5 min of ultraviolet irradiation. On the other hand, when BB-[32PlCDP-choline was used as a photoafftnity ligand, a single polypeptide with apparent molecular size of 55 kDa could he rapidly photolabeled within 2.5 min, then this band gradually lost its 32P-radioactivity with increasing time of ultraviolet irradiation. Thus, the overall results strongly indicated that cholinephosphotransferase in rat liver microsomes exists most likely as a 55 kDa polypeptide (or subunit) and that 25 kDa and 18 kDa peptides identified after the direct photoaffinity labeling with [32Pl CDP-choline were probably the photo-cleavage products of cholinephosphotransferase during the prolonged ultraviolet irradiation, both of which could contain the catalytic domain of the original enzyme protein(s).

Introduction Photoaffinity labeling with radiolabeled substrates or ligands has been used extensively with mononucleotide-requiring enzymes or proteins for the identification and mapping of active site regions. In these studies, photoreactive aryl azide derivatives of purine

Abbreviations: SDS-PAGE, sodium gel electrophoresis: BB-CDP-choline, CDP-choline.

dodecylsulfate-polyacrylamide 3’(2’)-O-(4-benzoyllbenzoyl

Correspondence: K. Ishidate, The Medical Research Institute, Tokyo Medical and Dental University, 2-3-10 Kanda-surugadai, Chiyoda-ku. Tokyo 101, Japan.

nucleotides are most efficiently employed [1,2]. Although in principle, pyrimidine nucleotide-requiring enzymes and proteins should also be amenable to study by a similar approach, this has not been examined extensively probably because of the difficulty in synthesizing azide derivatives on pyrimidine bases such as cytosine and uracil. In the past few years, direct photoactivation with underivatized nucleotide phosphates has also proved a successful method for labeling a number of enzymes and proteins [3-E]. This method has an advantage that the affinity between enzymes and substrates retains to be intact, whereas photosensitivity is relatively low when compared to that with photoreactive derivatives.

37 On the other hand, a potent photoreactive derivative of CTP, 3’-O-(4-benzoyl)benzoyl CTP has been developed recently and used successfully for the labeling of a CTP-requiring enzyme, CMP-N-acetylneuraminic acid synthetase [16]. As discussed elsewhere [17], it has now been found that the purification by conventional methods usually effective for the purification of a number of membrane-bound enzymes would not be very effective for cholinephosphotransferase from rat liver microsomes mostly because of its unusual sensitiveness to the presence of detergents. Thus, we next attempted to identify the cholinephosphotransferase protein on SDS-PAGE after either a direct photoaffinity labeling of partially purified enzyme from rat liver microsomes with 32Plabeled CDP-choline itself or a labeling with its synthetic derivative, 3’(2’)-0-(4-benzoyl)benzoyl [ j2 PICDP-choline (BB-CDP-choline). If cholinephosphotransferase were specifically identified on a SDS gel, then an antibody could be raised against the eluted protein. Alternatively, purification of the “‘P-labeled protein could be achieved without any care of the enzyme activity during the subsequent process. In this report, we describe first the nature of a specific binding occurred during the ultraviolet irradiation of partially purified rat liver cholinephosphotransferase preparation in the presence of “2P-labeled CDP-choline, then the autoradiographic identification on a SDS gel of several polypeptides possibly derived from a cholinephosphotransferase protein of rat liver microsomes, after the photoaffinity labeling with either 32P-labeled CDP-choline or its photoreactive derivative, BB-[ 32P]CDP-choline. Materials

and Methods

Chemicals [ y- “‘PIATP (spec. act.: 4500 Ci/mmol,

radiochemical purity: > 98%) was obtained from ICN Radiochemicals (Irvine, CA) and used without further purification. All other chemicals were of analytical grade. Choline kinase was partially purified from rat liver cytosol through pH 5.1 treatment, ammonium sulfate fractionation, DEAE-cellulose column chromatography followed by Sephadex G-200 gel filtration and stocked in - 20°C as described elsewhere [18]. The specific activity of the partially purified enzyme was approx. 200 nmol/min per mg protein. Partial purification of cholinephosphotransferase rat liter microsomes

from

Cholinephosphotransferase was solubilized from rat liver microsomes by the presence of 6 mM (0.25%, 0.25 mg/mg protein) sodium deoxycholate (DOC) and partially purified through a Superose 12HR column as described elsewhewre [17]. DOC was removed by

overnight dialysis against Buffer B [17] and a pellet fraction after the high-speed centrifugation (200 000 x g for 30 min) was resuspended in a small amount of Buffer B, then stored in -20°C until use. Preparation of “‘P-labeled CDP-choline [ ‘* P]CDP-choline was synthesized chemically from 5’-CMP and [ j2 Plphosphocholine essentially according to the method of Pontoni et al. [ 191. [“*P]Phospho-

choline was prepared enzymatically from choline and [Y-“~P]ATP through the reaction by choline kinase. The typical incubation mixtures for choline kinase reaction contained 0.1 M Tris-HC1 (pH 9.0>, 3.3 mM MgCl,, 5 mM choline chloride, 2.5 mM [y-“*P]ATP (spec. act.: 3.3 mCi/pmol) and partially purified enzyme from rat liver cytosol (1 mg protein) in a final volume of 0.6 ml. The incubation was conducted for 60 min at 37°C and stopped by the addition of 40% trichloroacetic acid (TCA) solution to a final TCA concentration of 5%. The mixtures were kept on ice for 30 min with occasional mixing, then centrifuged at 1000 x g for 5 min. The pellet was washed once with 500 ~1 of 5% TCA and recentrifuged in a same man ner. The resulting supernatants were pooled and washed three times with the same volume each of diethyl ether to remove most of TCA. After the pH of the aqueous phase was adjusted to 8.0 with 1 M KOH, the whole content was applied onto a small Dowex 1X8 column (1 x 13 cm, HCOO- form) and the column was washed with five volumes (approx. 50 ml) of H,O. Then, [“‘Plphosphocholine was eluted out with 0.1 M HCOOH. The unreacted [ y- “2P]ATP was still retained on the column at this stage and it could be recovered by elution with 4 M HCOOH plus 0.8 M HCOONH,. The 0.1 M HCOOH-eluted fractions with high ‘2P-radioactivity were pooled and lyophilized into a 15 ml polypropylene tube. The yield of [ “2P]phosphocholine ranged 50-60% as estimated from the starting specific radioactivity of [ y- 32P]ATP. The dried [“*P]phosphocholine was added by 100 ~1 of 2 M tosylchloride in dimethylformamide (DMF) and incubated for 15 min at room temperature. The mixtures were transferred carefully to another polypropylene tube which had contained 100 pmol (32 mg) 5’CMP (free acid). Another 100 ~1 of DMF was used for washing out the former tube, then added to the reaction mixtures. The mixtures were stood for 2 h at room temperature with occasional mixing, then added with 300 ~1 of H,O to stop the reaction. The content was washed three times with the same volume each of ethyl acetate and the final aqueous phase was subjected to TLC (Kiesel gel 60 from Merck, Dermstadt, Germany) separation of a product [32P]CDP-choline from unreacted [32Plphosphocholine as well as from a large excess of 5’-CMP. A developing solvent system of MeOH/H,O/AcOH (60 : 25 : 15, v/v) was employed

38

Synthesis of BB-[“2P]CDP-choline

: Fig. I. Isolation of [” PICDP-choline through HPLC. The lyophilized, crude [“P]CDP-choline fraction from the TLC plate was dissolved in a small amount of 25 mM HCOONH, (pH 4.4) and an aliquot (usually 100 in ~1) was applied onto a TSK-DEAE-2SW column. An isocratic elution with 25 mM HCOONH, (pH 4.4) resulted in complete separation of [3’P]CDP-choline (peak B) from either [32P]phosphocholine (peak A) or 5’CMP (peak C). ( -), relative absorbance at 254 nm; (o-----l), relative “‘P-radioactivity.

[19]. The area of CDP-choline spot (estimated by the simultaneous run of the authentic standard on each side as well as the center of the plate) was marked under ultraviolet irradiation (254 nm), and scraped off carefully from the plate into a 50 ml polypropylene tube. [ ‘* PICDP-choline was extracted several times with 0.5% NH,OH solution and the total extract was lyophilized to dryness. The dried [ j2 P]CDP-choline fraction was dissolved in a small amount of 25 mM HCOONH, (pH 4.4) and subjected to further purification by HPLC (Hitachi model L-6200) equipped with an anion-exchange column (TSK-DEAE-2SW, from Tosoh, Japan). An isocratic elution with 25 mM resulted in complete separation of HCOONH, [ ‘2P]CDP-choline from [ 32P]phosphocholine or 5’CMP, either of which had still contaminated to the fraction (Fig. 1). The fractions corresponding to [“‘PICDP-choline were pooled and lyophilized to dryness. The purity was checked further by rechromatography on TSK-DEAE-2SW with a different elution system (a linear gradient elution of O-O.75 M triethylamine acetate (TEAA), pH 7.8) and estimated over 99%. Finally, the purified sample was dissolved in a small amount of 5% EtOH and stored in -20°C. The specific activity of [“‘PICDP-choline thus obtained was usually in a range of 1-5 pCi/pg, which was totally dependent on the starting specific radioactivity of [y‘*P]ATP and the duration required for the above synthetic process.

Both the chemical synthesis of BB-CDP-choline and the method of its identification were followed to those for 3’-0-(4-benzoyl)benzoyl CTP reported by Abeijon et al. 1161. Briefly, the mixture of l,l-carbonyldiimidazole (0.43 M) and 4-benzoylbenzoic acid (0.14 M) in DMF was stirred for 15 min at room temperature, then 22 ~1 portion of the mixture was added to the aqueous solution (50 ~1) of [‘*P]CDP-choline (50-100 PCi, 30-60 nmol). The mixtures were vortexed for 30 min, then stood overnight in a light-shield tube. The content was dried in a vacuum centrifuge and the remnant was washed several times with cold acetone (100 ~1 each). The resulting precipitates were dissolved in 50 ~1 of H,O, then applied on a TSK-DEAE-2SW column. Two major ultraviolet absorbing peaks with high radioactivities were detected with a linear gradient elution of O-O.75 M TEAA (pH 7.8): The first peak was co-eluted with authentic CDP-choline and the other was eluted thereafter (Fig. 21, and the latter peak was estimated to be of BB-[ 32PICDP-choline based on several criteria described below. The yield of BB[ 32P]CDP-choline ranged 9- 12% and the remaining was recovered quantitatively as unreacted [ 32PICDPcholine through the HPLC. The purified, putative BB[ “*P]CDP-choline fraction was lyophilized, dissolved in a small amount of 5% EtOH and stored at -20°C in a light-shield tube. For the synthesis of non-radioactive BB-CDPcholine, the initial incubation was scaled up to 50-100

15

IIIJIIIIIIIIIII/

,A’,

/I,<

/111111,,,~//~~,,1,,,/

vj ::

”

::

Y

Fig. 2. HPLC pattern of the reaction products of [“PICDP-choline after the incubation with 4-benzoyl benzoic acid. The reaction mixtures were dried, washed with cold acetone, dissolved in a small amount of H,O, then applied onto a TSK-DEAE-2SW column. The elution was conducted with a linear gradient of O-O.75 M TEAA. Other details are described in the text. (A), [ “P]CDP-choline; (B), a ). relative absorbance putative BB-[“PICDP-choline peak. (at 254 nm; CO---O). relative “P-radioactivity: (- - - - - -). TEAA gradient CM).

39 times. A subsequent workup was done exactly as described for the radioactive compound. The criteria for the identification of BB-CDP-choline were as follows: (1) A single ultraviolet absorbing spot with an R, of 0.60 was detected following chromatography of an aliquot of this ultraviolet absorbing fraction on a TLC cellulose plate in a developing solvent system of n-butanol/AcOH/H,O (5 : 2 : 3, v/v) [ 161. In this TLC system, 4-benzoylbenzoic acid and CDPcholine had R, values of 0.95 and 0.18, respectively; (2) The putative BB-CDP-choline was shown to be photoreactive. An aliquot was applied on a TLC plate, ultraviolet irradiated (wave length > 366 nm) for up to 30 min at a distance of 2 cm, then chromatoSiaphed in a solvent system described above. Following development, the extent of photo-cleavage materials which remained at the origin of the plate was determined. Such an observation showed that the ultraviolet reacting material remained at the origin of the plate increased with increasing time of irradiation. Essentially no ultraviolet reacting material remained at the origin of the plate when nonradiated control or CDP-choline was subjected to similar irradiation; (3) The ultraviolet spectrum of BB-CDP-choline at pH 7.0 showed a maximum absorption at 263 nm. Under the same conditions, the maximum absorption of CDP-choline was 271 nm and for 4-benzoylbenzoic acid it was 262 nm; and (4) Both the putative BB-CDP-choline and CDPcholine were treated overnight with 0.5 M KOH, then the degradation products were compared in the above TLC system. Besides a spot corresponding to benzoylbenzoic acid derived from BB-CDP-choline, there was no significant difference in TLC patterns between the two compounds. All of these results strongly suggested that a reaction product of CDP-choline and 4-benzoylbenzoic acid described above was BB-CDP-choline, although evidence for the benzoylbenzoic acid group being attached to the 3’-hydroxy group of the ribose [16] was not confirmed at the present investigation. Photoaffinity labeling of cholinephosphotransferase Partially purified cholinephosphotransferase (100 pg protein) was incubated for 15 min at 0°C in the medium (complete system) containing 50 mM Tris-HCl (pH 8.0), 20% glycerol, 2 mM dithiothreitol (DTT), 1 mM EGTA, 10 mM MgCl,, 5 mM NaF (a phosphatase inhibitor) and either [‘2P]CDP-choline (1 PCi, 1.0-3.0 nmol) or BB-[‘2P]CDP-choline (0.3 PCi, 0.3-0.9 nmol) in a final volume of 100 ~1. The mixtures were then placed on a Parafilm layered over an ice-cold aluminum block, and photo-irradiated for 2.5-30 min at a distance of 7 cm from the ultraviolet lamp (254 nm, approx. 3000 FW/cm’). After the irradiation, the mixtures were transferred back carefully into the original polypropylene microtube and added with 100 ~1 of 5

mM cold CDP-choline or BB-CDP-choline. TCA was added to a final concentration of 10% and the mixtures were kept on ice for 1 h with occasional mixing, then centrifuged for 2 min at 10000 rpm with a microcentrifuge. The resulting pellet was washed three times with 200 ~1 each of 5% TCA, and once with 200 ~1 of cold acetone. The final precipitates were dissolved in 100 ~1 of SDS sample buffer of Laemmli [20] and warmed up to 65°C for 30 min. An aliquot was used for the determination of “‘P-radioactivity with 1 ml H,O and 10 ml ACS-II (Amersham) liquid scintillation cocktail. The remaining was subjected to slab SDS-PAGE (a 5-15% linear polyacrylamide gradient gel) following to the condition of Laemmli [20]. The gel was stained with Coomassie R-250, destained, dried and immediately autoradiographed with Fuji AIF New RX film. Results Direct photoaffinity labeling of cholinephosphotransferase by [“2P]CDP-choline When the partially purified cholinephosphotransferase preparation from rat liver microsomes was preincubated and photo-irradiated in the medium (complete system) containing [ 32PICDP-choline, the incorporation of ‘2P-radioactivity into the TCA-insoluble materials increased linearly with time up to 30 min of ultraviolet irradiation as shown in Fig. 3. There detected some radioactivity which was independent upon ultraviolet irradiation and this was probably due to a non-covalent binding between cholinephosphotransferase and [“PICDP-choline because this did not occur at all in the absence of Mg2+. MgZf, as well as Mn2+ has been known to be a prerequisite cation for the ekpression of cholinephosphotransferase activity

4i

Fig. 3. Photo-irradiation-dependent [“PICDP-choline binding to the TCA-insoluble components of partially purified cholinephosphotransferase. The partially purified cholinephosphotransferase (100 yg protein) was incubated for 15 min at 0°C in the medium (complete system) containing 50 mM Tris-HCI (pH 8.0), 20% glycerol, 2 mM DTT, 1 mM EGTA, 10 mM MgCI,, 5 mM NaF and 1 PCi (1.8 nmol) [‘*P]CDP-choline in a final volume of 100 ~1. In some incubations, MgCI, was omitted from the complete system. The mixtures were then placed on an ice-cold aluminum block and photo-irradiated for the indicated period of time at a distance of 7 cm from the ultraviolet lamp. Other experimental details were described in the Materials and Methods. (a), complete system; (A), Mg*+ omitted.

40 121-251. It was shown that photoirradiation-dependent binding of [‘2P]CDP-choline also required the presence of Mg2+ (Fig. 3), which suggested that the observed binding (supposed to be covalent) could be a result of the enzyme-substrate interaction. It may be possible to speculate that the observed “P-incorporation into TCA-insoluble components could be a result of 32P-phosphatidylcholine production from [ “2P]CDP-choline and endogenous diacylglycerols during the preincubation, and the resulting ‘* P-phosphatidylcholine could be oxidized during ultraviolet irradiation to cause its cross-linking with macromolecules. However, this could not likely be the case because our partially purified cholinephosphotransferase preparation was found to be free of diacylglycerol, and, if any, a trace amount of 32P-phosphatidylcholine could be removed out from the TCA-precipitate by washing with acetone (see the Experimental section). It was found, at the same time, that photo-irradiation caused a severe inactivation of cholinephosphotransferase in either the partially purified enzyme preparation or microsomal membranes (Fig. 4). This inactivation was probably due to the cleavage of peptide bonds upon ultraviolet irradiation, because the result of SDS-PAGE clearly indicated that a number of polypeptide bands disappeared with increasing time of ultraviolet irradiation, while the staining of the portions of small molecular size components as well as of large aggregates significantly increased (data not shown). Intact membranes (microsomes) were shown to be relatively stable against inactivation by photo-irradiation, which may indicate the possible function of membrane phospholipids in protecting the enzyme activity against the radiation.

Fig. 4. Effect of ultraviolet-irradiation on cholinephosphotransferase activity. Both the partially purified cholinephosphotransferase (100 pg protein) and intact microsomes (300 kg protein) from rat liver were photo-irradiated as described in Fig. 3, except that the final incubation volume was 200 ~1 and [“‘PICDP-choline was omitted from the medium. At the indicated time, an aliquot was taken out for the assay of cholinephosphotransferase activity [17]. The specific activities of non-irradiated controls were: 120 nmol/min per mg protein for the partially purified enzyme and 32 nmol/min per mg partially purified enzyme; protein for microsomes. (0 -o), A ), microsomes. CPT, cholinephosphotransferase. (A-----

Next, the effect of several factors on ultraviolet irradiation-dependent binding of [” P]CDP-choline was investigated. The removal of either DTT or EGTA from the medium did not significantly affect the binding efficiency. The pH range between 6 and 8.5 gave essentially the same degree of binding, whereas pH at 9.0 resulted in about half of the maximum binding at pH 8.0. The inclusion of 1,2-diacylglycerol (0.25-1.0 mM) caused a significant elevation of “‘P-incorporation into the TCA-precipitable fraction, but this was mainly due to the elevation of photo-irradiation-independent binding. Thus, the stimulation of apparent j2P-binding by diacylglycerol could be explained most likely in this case by the production of j2P-labeled phosphatidylcholine during the incubation, which could have been co-precipitated with the protein component by the presence of TCA and could not be removed completely by a single acetone washing. The photoirradiation-dependent binding, on the other hand, did not change significantly by the addition of 1,2-diacylglycerol. As described above, [ 32P]CDP-choline binding had an absolute requirement for Mg’+. The replacement of Mg2+ (5 mMI by 5 mM Mn2+ resulted in 50-600/o reduction of the photo-irradiation-dependent binding of [“2P]CDP-choline, which had been expected from our previous observation [17] that Mn2+ could replace in part for Mg2+ in cholinephosphotransferase reaction with the partially purified enzyme preparation. Fig. 5A shows the dependency of photo-irradiationdependent binding on Mg2+ concentration in the medium. As low as 0.1 mM Mg2+ was found to give 80% of the maximum binding, and this dependency seemed to be very similar to that for cholinephosphotransferase reaction (Fig. 50. It has been shown that Cal+ was a potent inhibitor of cholinephosphotransferase from various tissue sources [21,22,25-281, and so was the case for partially purified cholinephosphotransferase from rat liver microsomes (Fig. 5D). Our assay mixtures for cholinephosphotransferase activity contained 10 mM MgC12 and 0.25 mM EGTA in the present investigation. When the Ca2+ concentration exceeded that of EGTA, a strong inhibition of cholinephosphotransferase reaction occurred. The photoirradiation-dependent binding of [ “P]CDP-choline to the protein component also was severely inhibited by the presence of PM Ca2+ (Fig. 5B), which indicated strongly that Ca2+ could inhibit cholinephosphotransferase reaction by inhibiting the formation of enzymesubstrate (CDP-choline) complex. It was also indicated that Mg2+, or Mn2+, was absolutely required for the binding of enzyme to its substrate, CDP-choline in cholinephosphotransferase reaction. To our knowledge, this is the first direct evidence for the role of by Mg2+ (Mn*‘) as well as for the site of inhibition Ca2+ in cholinephosphotransferase reaction.

41

h

I-

s 2 a V

02

04

L 02

08

04

tCa2+ln4 8

/?i+

-

32

CCa*+ln-M

%”

3:

[ “2P]CDP-choline binding and Fig. 5. Comparison of ~g2~-dependency and inhibition by Ca2+ between ultraviolet irradiation-dependent cholinephosphotransferase activity. In (A) and (B): The partially purified cholinephosphotransferase (50 pg each) was preincubated and photo-irradiated for 15 min as described in Fig. 3, except that the concentration of EGTA in this experiment was 0.125 mM and that of MgCI, was changed as indicated (A). In (0 and CD): The assay medium for cholinephosphotransferase activity contained 7.5 mM Tris-HCI (pH 8.5), 10 mM (D) or the indicated amount (C) of MgCI,, 0.25 mM EGTA, 8 mM DTT, 2 mM 1,2-diacylglycerol (in final 0.02% Tween-20 emulsion), 0.5 mM CDP-[.J4e-‘4C]choline (spec. act., 0.1 Ci/mol) and the partially purified enzyme preparation (10 pg protein) in a final volume of 0.2 ml. Other details for cholinephosphotransferase assay are described elsewhere [17]. CPT, cholinephosphotransferase.

Although the results described above strongly suggested that the observed photo-irradiation-dependent binding most likely occurred between [32P]CDP-choline and cholinephosphotransferase protein, the possibility still existed that the binding might be nonspecific. To overcome this possibility, we next attempted to examine the effect of several CDP-choline structural analogs on the [32P]CDP-choline binding. As shown in Fig. 6, the binding was inhibited in a dose-dependent manner by the presence of PM ranges of cold CDP-choline.

Fig. 6. Specific inhibition of ultraviolet-irradiation-dependent [32P]CDP-choline binding by either CDP-choline or CDP-ethanolamine. The preincubation (15 min at 0 o C) and photo-irradiation (15 min) conditions were the same as in Fig. 3, except that 3 nmol 13’P]CDP-choline (1 PCi) was used as an affinity ligand in this experiment. (01, CDP-choline; (A 1, CDP-ethanolamine; (01, phosphocholine (P-choline) (W ),5’-CDP.

CDP-ethanolamine was found to be also a strong inhibitor against [32P]CDP-choline binding up to 50% inhibition, but the remaining binding activity was not influenced by increasing concentrations of CDPethanolamine. This result might be explained for that there could be two [“‘PICDP-choline binding components existed in the partially purified cholinephosphotransferase preparation; one was specific for CDPcholine and the other had an affinity for either of CDP-bases equally. In this point, it should be noted that there could exist two phosphotransferases in yeast; one was specific for cholinephosphotransferase reaction whereas the other had catalytic activities for both cholinephosphotransferase and ethanolaminephosphotransferase reactions [29,30]. Finally, it was found that neither phosphocholine or 5’-CDP had any significant effect on the photo-irradiation-dependent [32P]CDP-choline binding (Fig. 6). Ident~~cat~~n of ~3~P~~DP-choline- or 3B-t3~P~~DPcholine-labeled p~~ypeptides on SDS-PAGE followed by 32P-autoradiagraphy

The goal of the present study was to identify the protein component(s) to which [ 32PICDP-choline bound covalently by photoaffinity labeling. For this a portion of the photo-irradiated, TCA-precipitated fraction was subjected to SDS-PAGE, and the dried gel was then

42

97K > 55K > 36K >

j

abcdefghi Fig.

7. Identification

of 3’P-labeled

after the photoaffinity transferase

with

[‘I

labeling of partially

P]CDP-choline.

lized samples from the experiment slab SDS

gel (5-ISV

linear

staining with Coomassie autoradiographed

at

polypeptides purified

aliquot

gradient

the gel was dried

with

Fuji

AIF

in the Materials a number

New

standards globulin

from (I70

dehydrogenase inhibitor

kDa,

Boehringer reduced).

(55 kDa),

(20 kDa).

CDP-choline:

CDP-choline: ethanolamine:

Mannheim

film.

co-migrated

to the

size components

molecular

sizes of MW

b (07 kDa),

lactate dehydrogenase

Other

The band at

Biochemica;

phosphorylase

After

a?-macroglutamate

(36 kDa) and trypsin

The origin of each sample was: lane a. complete

system; lane b. complete c. 30 FM

RX

of small molecular

should co-exist. Arrows indicate the apparent

gel).

and immediately

and Mcthoda.

the bottom of each lane indicates the radioactivity top of the gel where

cholinephospho-

of the SDS-solubi-

shown in Fig. h was applied on a

polyacrylamide

R-250,

-70°C

details were described

An

on SDS-PAGE

site, i.e., a CDP-choline binding domain. We thought that either of these polypeptides may be too small for a subunit (if any) of cholinephosphotransferase, because the yeast enzyme has recently been deduced to bc composed of a single polypeptide of MW 46305 [3 I]. In addition, the direct photoaffinity labeling with [32P]CDP-choline was found to require more than IO min of ultraviolet irradiation to give a sufficient amount ot radioactivity into these polypeptides. As described carher, a longer period of ultraviolet irradiation caused a significant cleavage of polypeptides into smaller pcptide fragments as well as into large aggregates. and this could also be a case for cholinephosphotransferase in the above investigation. Then, we next examined the photoaffinity labeling of cholinephosphotransferase by a more photoreactive derivative of CDP-choline, BB-[ 3’P]CDP-cholinc. In a preliminary experiment, BB-CDP-choline was shown to be a competitive inhibitor for CDP-choline (K,: 2 mM vs. K,,, for CDP-choline: 235 PM) in cholincphospho-

system without lane d, 60 PM

lane f. IS0 FM lane h, 60 FM

CDP-ethanolamine;

ultraviolet CDP-choline;

CDP-choline;

irradiation;

lane g, 30 PM

CDP-ethanolamine;

lane

lane e. 90 ,uM CDP-

lane i. YO FM

lane j, I50 PM CDP-ethanolamine.

examined with 3’P-autoradiography. As shown in Fig. 7, two radiolabeled bands of approximate molecular sizes of 25 kDa and 18 kDa were clearly identified on the autoradiogram (lane a>. Neither of these bands could be detected without photo-irradiation (lane b), indicating the photo-induced incorporation of ” P-radioactivity from [ 3’P]CDP-choline into those polypeptides. In addition, the intensity of photo-irradiation-dependent 32P-labeling of these polypeptides decreased in a dose-dependent manner by the presence of cold CDP-choline in the incubation mixtures before the ultraviolet irradiation (Fig. 7, lane c-f). The presence of CDP-ethanolamine also resulted in the deinto both creased incorporation of “2P-radioactivity peptide bands though with relatively low integrity when compared to that of CDP-choline (lane g-j>. The results from both Figs. 6 and 7 suggested strongly that either 25 kDa or 18 kDa polypeptide band on SDS-PAGE could be at least a part of the peptide fragments derived from rat liver cholinephosphotransferase protein which contained its catalytic

J(* --

a

--20K

0-L.

b

cd

Fig. 8. Identification the photoaffinity purified bated

labeling

f

cholinephosphotransferase

(IO0

with either

h

on SDS-PAGE

pg

The

protein)

after

partially

was preincu-

[“P]CDP-choline

or BB-[“P]CDP-choline

in a same manner

g

polypeptidc

with BB-[3’P]CDP-choline.

and photoirradiated

nmol) (lanes a-d) e-h)

e of “P-labeled

(I

PC?. I

(0.3 PCi, 0.3 nmol) (lanes

as in Fig. 3. The TCA-insoluble

fraction was

dissolved in 100 ~1 of SDS sample buffer [20] and an aliquot (50 ~1 each) was applied gradient

gel).

Materials

and

on a slab SDS gel (S-15%

Other

experimental

Methods.

indicates the radioactivity of the MW

standards

right. The incubation without ultraviolet

The

details

band

co-migrated

at the

irradiation;

of

in the

each

lane

run were pointed

at the

of each sample were: lanes a and e,

lanes b and f. ultraviolet

2.5 min; lanes c and g, ultraviolet ultraviolet

described

bottom

to the top of the gel. Positions

from a simultaneous conditions

linear polyacrylamide were

irradiated

irradiated

irradiated

for

for 5 min: lanes d and h,

for IO min.

43 transferase reaction with rat liver microsomes (data not shown). The result of photoaffinity labeling by BB[“2P]CDP-choline is shown in Fig. 8, together with that by [ 32PICDP-choline in comparison. The autoradiogram with BB-[‘2 PICDP-choline clearly showed that only a single polypeptide with apparent MW of 55 kDa was radiolabeled through the ultraviolet irradiation for as short as 2.5 min (Fig. 8, lane f). A longer period of irradiation over 5 min caused a considerable loss of radioactivity in this band and some of the radioactivity seemed to appear into smaller molecular bands instead. It should be noted that BB-[‘2P]CDP-choline itself could not be stable for the prolonged ultraviolet irradiation, thus might cause a number of nonspecific bindings to macromolecules in the later periods. The autoradiogram with [“2P]CDP-choline, on the other hand, showed a 25 kDa polypeptide band radiolabeled after as long as 5 min of ultraviolet irradiation (Fig. 8, lane c) and this band was found to get much more ‘2P-radioactivity after 10 min of irradiation (lane d). The 18 kDa polypeptide band in this experiment moved to near the top of the gel and was masked by a large amount of “2P-radioactivity co-migrated with the smaller molecular size components. Thus, the overall results from SDS-PAGE-autoradiographic investigation strongly indicated that cholinephosphotransferase in rat liver microsomes could exist in its approximate molecular size (or subunit) of 55 kDa. In addition, an extensive ultraviolet irradiation most likely cleaved a 55 kDa cholinephosphotransferase into two polypeptide fragments, 25 kDa and 18 kDa, both of which could contain the catalytic domain of the original enzyme protein. Discussion Recently, Cornell has described in her review on cholinephosphotransferase [251 that photoaffinity labeling by a certain radioactive CDP-choline analog may be a shorter way to obtain molecular information of cholinephosphotransferase from animal origins than classical purification of the enzyme activity following to its solubilization from membranes. We are now rather inclined to agree on her proposal because we have found that, as discussed elsewhere [17], the purification by conventional methods did not work very effectively for cholinephosphotransferase from rat liver microsomes, and that photoaffinity labeling by [“2P]CDPcholine analogs could result in the identification of specifically s2 P-labeled polypeptides on SDS-PAGE. In the present investigation, we attempted, for the first time, to identify a cholinephosphotransferase protein in rat liver microsomes on SDS-PAGE following to the photoaffinity labeling either directly by [32PlCDP-choline or by its photoreactive derivative, BB-[ 32PlCDP-choline. The result from binding experi-

ments of [“2P]CDP-choline to the TCA-insoluble components in partially purified cholinephosphotransferase preparation from rat liver microsomes demonstrated that there did occur some ultraviolet irradiation-dependent [“PICDP-choline binding for which the presence of Mg2+ or Mn2+ was absolutely required. This binding was strongly inhibited by the presence of a trace amount of Ca2+. In addition, the binding was shown to be specific in the sense that the addition of cold CDP-choline inhibited the binding in a dose-dependent manner whereas neither phosphocholine or 5’-CDP could significantly affect the binding. Thus, the characteristics of ultraviolet irradiation-dependent [ j2 P]CDP-choline binding to the TCA-insoluble component were found to be almost equivalent to those of cholinephosphotransferase reaction, suggesting strongly that the observed photoaffinity binding could be the result of the specific interaction between cholinephosphotransferase and one of its substrate, CDP-choline. Two polypeptide bands of apparent molecular sizes of 25 kDa and 18 kDa were clearly identified on a SDS gel through “P-autoradiography after the photoaffinity labeling of cholinephosphotransferase preparation with [‘2P]CDP-choline. Neither of these bands could be detected without ultraviolet irradiation, indicating that the incorporation of “2P-radioactivity into these polypeptides could be due to a photo-induced covalent binding of [32P]CDP-choline to its specific acceptor, cholinephosphotransferase. On the other hand, the labeling with BB-[.72P]CDPcholine, a potent photoreactive derivative of CDPcholine, resulted in a rapid appearance of 55 kDa polypeptide band with strong autoradiographic intensity on SDS-PAGE. This band, however, disappeared with increasing duration of photo-irradiation, indicating that either this polypeptide or BB-CDP-choline: peptide complex was not stable against the prolonged ultraviolet irradiation and gradually cleaved into smaller peptide fragments. The molecular information of cholinephosphotransferase has so far been reported only for a yeast system 1311, where a genetic complementation analysis to a mutant lacking cholinephosphotransferase activity provided its possible molecular size of 46305, which has been deduced from the leading nucleotide sequence of the structure gene. Our identification by photoaffinity labeling with BB[ 32P]CDP-choline followed by SDS-PAGE and j2Pautoradiography suggested strongly that cholinephosphotransferase in rat liver microsomes most likely exists in 55 kDa polypeptide (or subunit), which seems to be of somewhat higher molecular size than that in yeast membranes. Attempts to isolate the 32P-labeled 55 kDa polypeptide as well as its putative photo-cleavage products, 25 kDa and IS kDa peptide fragments labeled with

44 [“P]CDP-choline, are now in active progress laboratory for the production of antibodies.

in this

Acknowledgement This work was supported in part by Grant-in-aids for Scientific Research from the Ministry of Education, Science and Culture, Japan (No. 01771971) and from the Meiji Milk Products, Japan. We wish to thank Ms. Yukiko Ozaki, a student from the College of Pharmacy, Science University of Tokyo, for her excellent technical assistance throughout the experiments. References 1 Czarnecki. J., Geahlen. R. and Haley. B. (1979) Method. Enzymol. 56, 642-653. 2 Hoyyer. P.B.. Owens. J.R. and Haley, B.E. (1980) Ann. N.Y. Acad. Sci. 346, 2X0&301. 3 Antonoff, R.S. and Ferguson, J.J., Jr. (1974) J. Biol. Chem. 240. 3319-3321. 4 Yue, V.T. and Schimmel, P.R. (1977) Biochemistry, 16. 467% 4684. 5 Maruta, H. and Korn, E.D. (1981) J. Biol. Chem. 256, 4999502. 6 Caras, I.W. and Martin, D.W.. Jr. (1982) J. Biol. Chem. 257. 950X-0512. 7 Eriksson, S., Caras, I.W. and Martin, D.W., Jr. (1982) Proc. Natl. Acad. Sci. USA, 79. X1-85. H Caras, I.W., Jones, T., Eriksson, S. and Martin. D.W., Jr. (1983) J. Biol. Chem. 25X. 306443068. Y Lee. R.W.H.. Suchanek. C. and Huttner, W.B. (1984) J. Biol. Chem.. 259. 11153-11156. IO Biswas, S.B. and Kornberg, A. (1984) J. Biol. Chem. 259, 79907993.

11 Eriksson, S., Sjoberg, B.-M., Jornvall, H. and Carlquist, M. (1986) J. Biol. Chem. 261, 187X1X82. 12 Moriyama, Y. and Nelson, N. (1987) J. Biol. Chem. 262. 1472314729. I3 Kierdaszuk, B. and Eriksson. S. (IYXX) Biochemistry. 27. 4Y524956. I4 Stroop, S.D., Charbonneau. H. and Beavo, J.A. (19x9) J. Biol. Chem., 264, 1371X-13725. 15 Jansson, 0. and Eriksson. S. (1990) Biochem. J. 269. 2Oll205. 16 Abeijon, C.. Capasso. J.M., Tal, D., Vann. W.F. and ~iirschberg. C.B. (1986) J. Biol. Chem. 261, 11374-l 1377. I7 Ishidate, K., Matsuo, R. and Nakazawa. Y.. submitted for publication. IX Ishidate, K., Nakagomi, K. and Nakazawa, Y. (lY84) J. Biol. Chem., 25Y. 14706-14710. I9 Pontoni, G.. Manna. C.. Salluzzo, A., del Piano, L.. Galletti. P., De Rosa. M. and Zappia. V. (19x5) Biochim. Biophys. Acta. X36. 222-232. 20 Laemmli. U.K. t 1970) Nature (Land.), 227. 6X0-6X5. 21 Kennedy, E.P. and Weiss S.B. (1956) J. Biol. Chem. 222, 103-214. 22 Coleman, B. and Bell, R.M., J. Biol. Chem. (1977) 252. 3050-3056. 23 Kanoh, H. and Ohno. K., Method. Enzymol. (IYXI) 71, 536.546. 24 0. K.-M., Siow. Y.L. and Choy, P.C. (19X9) Biochem. Cell Biol. 67. 6X0-686. 25 Cornell, R. (1989) in Phosphatidylcholine Metabolism (Vance. D.E., ed.), Chapter 4, pp. 47-64, CRC Press, Boca Raton. 26 Weiss. S.B.. Smith, SW. and Kennedy. E.P. (1958) J. Biol. Chem. 23 I) 55fl4. 27 Taniguchi, S.. Morikawa. S.. Hayashi. H., Fujii, K., Mori, H., Fujiwara, M. and Fujiwara, M. (1086) J. Biochem. IOO. 4X5-4’) I. 2X Alemany, S.. Varela, 1. and Mato. J.M. (19X2) Biochem. J. 208. 453-457. 29 fijelmstad. R.H. and Bell. R.M. (IYX7) J. Biol. Chem. 262. 39OY-3Y17. 30 Hjelmstad, R.H. and Bell. R.M. (1988) J. Biol. C’hem. 263. 19748~19757. 31 Hjelmstad. R.H. and Bell, R.M. (1000) J. Biol. Chem. 265. 375.5-1764.