inhibition and ultraviolet-induced chemical modification ... · mechanismof chlorpromazine...

Plant Physiol. (1992) 100, 1927-19330032-0889/92/100/1 927/07/$01 .00/0

Received for publication April 21, 1992Accepted July 24, 1992

Inhibition and Ultraviolet-induced Chemical Modification ofUDP-Glucose:(1,3)-,B-Glucan (Callose)

Synthase by Chlorpromazinel

Mechanism of Chlorpromazine Binding to the Plant Plasma Membrane

Robert W. Harriman, Ai-Ping Shao, and Bruce P. Wasserman*Department of Food Science, New Jersey Agricultural Experiment Station, Cook College, Rutgers University,

New Brunswick, New Jersey 08903-0231

ABSTRACT

UDP-glucose:(1,3)-ft-glucan (callose) synthase (CS) from storagetissue of red beet (Beta vulgaris L.) was strongly inhibited by thephenothiazine drug chlorpromazine (CPZ). In the absence of ultra-violet irradiation, CPZ was a noncompetitive inhibitor with 50%inhibitory concentration values for plasma membrane and solubi-lized CS of 100 and 90 um, respectively. Both the Ca2"- and Mg2"-stimulated components of CS activity were affected. CPZ inhibitionwas partially alleviated at saturating levels of Ca2", but not Mg2",suggesting that CPZ interferes with the Ca2"-binding site of CS.Binding experiments with ["4C]CPZ, however, showed strong non-specific partitioning of CPZ into the plasma membrane, providingevidence that perturbation of the membrane environment is prob-ably the predominant mode of inhibition. Ultraviolet irradiation at254 nm markedly enhanced CPZ inhibition, with complete activityloss following exposure to 4 Mm CPZ for 2 min. Inhibition followeda pseudo-first order mechanism with at least three CPZ bindingsites per CS complex. Under these conditions, [3HJCPZ was cova-lently incorporated into plasma membrane preparations by a freeradical mechanism; however, polypeptide labeling profiles showedlabeling to be largely nonspecific, with many polypeptides labeledeven at [3H]CPZ levels as low as 1 gM, and with boiled membranes.Although CPZ is one of the most potent known inhibitors of CS, itsuse as a photolabel will require a homogeneous CS complex orestablishment of conditions that protect against the interaction ofCPZ with specific binding sites located on various polypeptidecomponents of the CS complex.

UDP-Glc-(1,3)-13-glucan2 CS (EC 2.4.1.34) is a PM-locatedenzyme responsible for the biosynthesis of the (1,3)-f:-linkedglucan callose. Although callose occurs naturally in such

1 This research was supported in part by grants from the NationalScience Foundation (DCB-8907202), the Charles and Johanna BuschFoundation, and the New Jersey Agricultural Experiment Stationwith State and Hatch Act Funds. This is New Jersey AgriculturalExperiment Station publication No. D-10558-1-92.

2 Abbreviations: UDP-Glc, uridine diphosphate glucose; AChR,nicotinic acetylcholine receptor; CPZ, chlorpromazine; CS, callosesynthase; NDP, nucleotide diphosphate; PM, plasma membrane;IC50, concentration required for 50% inhibition; B/F ratio, bound/free ratio.

specialized tissues as sieve plates and pollen tubes, CS activityis ubiquitous and callose synthesis can be induced by wound-ing (3). The regulation of CS activity has been intensivelystudied (3, 16, 27). One hypothesis is that callose is synthe-sized by a deregulated form of cellulose synthase that resultsfrom damage to cellular membranes (3, 11).

Callose synthesis requires divalent cations and is activatedby Ca2" and Mg2". Little is known about the mechanism bywhich these cations interact with CS. Phenothiazines, suchas the psychoactive drug CPZ, have been found to inhibitcation-dependent enzymes (8, 14, 15, 28). For example, CPZwas used to differentiate between Mg2"-dependent andMg2+-independent phosphatidate phosphohydrolase activi-ties in rat lung (28). Kumar et al. (15) showed that trifluper-azine, a CPZ analog, was an inhibitor of protein synthesis inEhrlich ascites tumor cells and that this inhibition was over-come by calmodulin. CPZ labeling of the AChR with [3H]-CPZ has allowed the identification of specific amino acidsthat comprise the membrane-spanning region and are in-volved in the transport of monovalent cations (6, 7, 23). Inthe AChR, [3H]CPZ has been shown to react with Ser248 ofthe a-subunit; Ser254 and Leu257 (fl subunit); Thr253, Ser257,and Leu260 (y subunit); and Ser262 (5 subunit) (6, 7, 23). Inplants, CPZ has been shown to mediate membrane break-down in potato tuber homogenates (18). CPZ effectivelyinhibited tracheary-element differentiation in suspension-cultured Zinnia, presumably by acting as a calmodulin inhib-itor (24). However, Kauss et al. (13) showed that inhibitionof soybean CS by trifluperazine, calmidazolium, and poly-myxin B was not mediated by calmodulin.These studies led us to hypothesize that CPZ and its

analogs would be inhibitors of CS. Because CPZ is highlyaromatic, its use as a direct photoaffinity label was investi-gated with the objective of developing a new tool for probingthe catalytic mechanism and molecular structure of CS. Thisstudy demonstrates that CPZ is a highly potent inhibitor ofCS and provides a characterization of the interaction of CPZwith CS and the plant PM.

MATERIALS AND METHODSMaterialsRed beets (Beta vulgaris L.) were obtained from local mar-

kets. UDP_[14C]Glc (specific activity 220 mCi/mmol) was a1927 www.plantphysiol.orgon April 7, 2020 - Published by Downloaded from

Copyright © 1992 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

Plant Physiol. Vol. 100, 1992

product of ICN Radiochemicals (Irvine, CA). [3H]CPZ (spe-cific activity 30 Ci/mmol) was obtained from New EnglandNuclear (Boston, MA). Digitonin was from ICN Biochemicals(Cleveland, OH). CPZ, its analogs, and all other chemicalswere obtained from Sigma.

Preparation of Membranes and Solubilized CS

Microsomal membranes were prepared from red beet stor-age tissue by differential centrifugation (30), with PM isolatedby sucrose gradient centrifugation (26, 30) or aqueous two-phase partitioning (32). Solubilized CS was prepared in 0.6%CHAPS, 1 mm EDTA, 1 mm EGTA, 7.5% glycerol, and 50mM Tris-HCl, pH 7.5, by the two-step solubilization proce-dure (Mg2" precipitation) (30, 32). KI-washed CS was pre-pared from two-phase partitioned membranes as described(31).

CS Assay

Assays were performed in mixtures of 100 ,uL containingenzyme (1.7-5.8 ,ug of protein, as specified), effectors (5 mMMgCl2, 0.5 mm CaCl2, 5 mm cellobiose, 0.01% digitonin), andsubstrate (1 mm UDP-["4C]Glc, 0.12 mCi/mmol) in 50 mMTris-HCl, pH 7.5. CPZ was added to assay mixtures asindicated. Reactions proceeded for 30 min at 300C andethanol-insoluble glucan was measured (30). One unit ofactivity is defined as that which catalyzes the incorporationof 1 nmol of glucose per min into ethanol-insoluble glucan.IC50 values were determined for PM-bound CS (5.8 ,g).

Scatchard Analysis

Scatchard plots were generated from data obtained asdescribed (20) using 5.4 ,g of KI-washed CS in 0.01% digi-tonin and 50 mm Tris-HCl, pH 7.5. KI-washed CS wasincubated with 1.6 x 105 dpm of [3H]CPZ and [14C]mannitolfor 15 min at 200C (four replicates). The samples were thenchallenged with various levels of cold CPZ for 15 min at200C and centrifuged onto 0.22-,um nitrocellulose filters(MSI, Westboro, MA) for 15 s in a Beckman Microfuge B.The filters were wetted with 50 ML of 10 mm KOH to which0.5 mL of 1% Triton X-100 was added, and the mixture wasshaken for 30 min at 350 rpm. The samples were countedwith a Beckman LS3801 liquid scintillation counter usingdual-label dpm programs. To account for liquid absorbed tothe filters, bound [14C]mannitol was subtracted from the total[3H]CPZ bound to the membrane.

UV-Induced CPZ Inhibition

UV-induced CPZ inhibition experiments were conductedby irradiation of 60-ML mixtures of PM and CPZ in 50 mMTris-HCl, pH 7.5, on ice, at 254 nm (9000 ,W/cm2) with aSpectroline model R-51 lamp. Protein and CPZ levels andexposure times are given under each experiment. To deter-mine residual CS activity, effector levels were adjusted to theconcentrations specified under 'CS Assay' with 30 ,L of aneffector mix, and reactions were then initiated by addition ofUDP-["4C]Glc. For protection experiments, ascorbate was pre-pared by titration of ascorbic acid to pH 7.5 with NaOH.

Direct Photolabeling with [3H]CPZSamples of 25 ML containing PM (3 Mug) and 1.6 Mm [3H]-

CPZ (30 Ci/mmol) were irradiated at 0°C as described above.To remove unbound [3H]CPZ, samples were brought to 15%TCA, held 15 min, and spotted on GF/A glass fiber filters(Whatman). Filters were washed sequentially with 10% TCAand 100 Mm CPZ (20 mL), 10% TCA (20 mL), 70% ethanol(20 mL), and acetone (10 mL). Covalently bound [3H]CPZwas then determined by scintillation counting.

Labeling patterns were determined by SDS-PAGE andfluorography (5). Instead of spotting TCA-precipitated sam-ples on glass fiber filters as above, the samples were centri-fuged at 13,600g for 15 min at 40C, and the pellets resus-pended in electrophoresis sample buffer (5). Electroblottingwas performed for 1.5 h on a Hoefer Transphor model TE-50 with power lid. The blots were dried, sprayed withEn3Hance (New England Nuclear), and exposed to Kodak X-Omat film.

Protein Determination

Protein was determined by Coomassie blue dye-bindingwith BSA as standard (1).

RESULTS

Initial Characterization

CPZ and propanolol, both of which are amphiphilic andcationic, have been used to inhibit various cation-bindingenzymes (8, 14, 15, 28). Both were found to inhibit CS;however, CPZ was more effective, with an IC50 of 100 Mfor PM-bound and 90gM for solubilized CS (Fig. 1), comparedwith an IC50 of 2 mm for propanolol (Table I). The inhibitoryproperties of structural analogs of CPZ were also examined(Table I). Only analogs such as triflupromazine and chloro-prothixene containing the positively charged methyl aminegroups and electronegative R2 moeities gave inhibition com-parable to CPZ. Analogs such as phenothiazine, which con-

100

PPla Membrane

\ \ * Solubilized

40

20

00 200 400 600 800 1000

CHLORPROMAZINE (gM)

Figure 1. CPZ inhibition of PM-bound and solubilized CS. Assaymixtures of PM and CS solubilized by CHAPS (26) contained 5.8and 1.75 ,ug of protein, respectively.

1928 HARRIMAN ET AL.

www.plantphysiol.orgon April 7, 2020 - Published by Downloaded from Copyright © 1992 American Society of Plant Biologists. All rights reserved.


CHLORPROMAZINE INHIBITION OF PLASMA MEMBRANE CALLOSE SYNTHASE

Table I. Structures and IC50 Values of Chlorpromazine and Related

R,

Chlorpromazine -CH2CH2CH2N(CH3)2 -CLPhenothiazine -H -HAcepromazine -CH2CH2CH2N(CH3)2 --CPromazine -CH2CH2CH2N(CH3)2 -HPropionylpromazine -CH2CH2CH2N(CH3)2 -CCTriflupromazine -CH2CH2CH2N(CH3)2 -CF:Chlorprothixene =CHCH2CH2N(CH3)2 -Cl

Ethopropazine CH3 -H--CH2CHN(C2H5)2

Promethazine CH3 -HT-CH2CHN(CH3)2

Trimeprazine -HC HNCH H3 H

Propanolol

Chlorpropamide

OH

OCH2CHCH2NHCH(CH)2

Ci .S02NHCONHCH2CH2CH3

(110, 68, and 18 nmol min-' mg-' at 0, 70, and 140 $M CPZ,respectively), indicating that CPZ acts as a noncompetitiveinhibitor of CS.

Influence of CS Effectors

R2 ICSO Investigations of CPZ's mode of action with respect to theAM

divalent cations Mg2+ and Ca24 were performed in two dif-100 ferent ways. The first set of experiments involved testing

>2000 CPZ inhibition in the absence of Mg24 and/or Ca24 (Fig. 3).KCH3 300 Consistent with previous studies (5, 10), removal of Mg2+

400 (assays conducted in the presence of Ca2+ only) decreased)CH2CH3 400 CS activity by approximately 20%, whereas the removal of3 100 Ca2` (assays conducted with Mg2+ and EGTA) decreased CS

150 activity by 70 to 80%. Figure 3 shows that CPZ strongly

650 inhibited both the Ca2+- and Mg2+-dependent activities in a

500

400

2000

100

80

60

40

20

0

100>2000

tain the heterotricyclic aromatic rings but lack the positivelycharged amine and electronegative R2 group, showed noinhibition of CS activity, whereas weak inhibition occurredwith amines containing sterically bulky groups as is seen withethopropazine and promethazine. A plot of UDP-Glc con-centration versus activity at various CPZ levels (Fig. 2) showslittle change in Km (150 Mm) but a significant decrease in Vm.

120

100

60

0240 40 600 80 10

UDP-Glc (jIM)Figure 2. Noncompetitive inhibition of CS by CPZ. Reaction mix-tures containing PM (5.8 Mg of protein) were assayed at CPZ levelsof 0, 70, 140, and 210 Mm for 5 min at 30°C at the indicated levelsof UDP-Glc.

001

-o-

-JJ

naw

+2 +2 +2 +2Ca/Mg -Mg -Ca

COMBINATION OF C

-Both

CATIONS

Figure 3. Effect of divalent cation combinations on CS inhibitionby CPZ. CPZ was added to assay mixtures at 100 (A), 150 (B), or200 gM (C). PM (5.8 Mg) CS were assayed in the presence andabsence of various combinations of Mg2", Ca24, and CPZ. Fullreaction mixtures served as controls, and effectors that were omit-ted are listed under Combination of Cations with -Mg24 referringto the Ca2"-stimulated component and -Ca24 to the Mg24-stimu-lated component. Assays lacking Ca24 contained 1 mm EGTA,whereas cation deficient assays (-Both) contained 1 mm EDTA/EGTA.

1929




concentration-dependent manner. At 100 and 150 gM CPZ(Fig. 3, A and B), the Ca2"-stimulated component of CSactivity (assayed in the absence of Mg2") was inhibited to agreater extent than Mg2+-dependent activity (assayed in theabsence of Ca2+). At 200 MM (Fig. 3C), CPZ inhibition of eachcation-stimulated component of CS activity was nearly com-plete, and differential effects could no longer be discerned.Inhibition of cation-independent CS activity (assayed withEGTA and EDTA), which represented 15% or less of fullyexpressed activity, was generally variable due to low levelsof recoverable activity. Residual divalent cation-independentactivity is probably due to cations endogenously bound toCS, since this activity declined by 50% when isolated PMswere washed with EDTA and EGTA (not shown). Equivalentlevels of inhibition were observed irrespective of the order ofaddition of CPZ and divalent cations to CS assay mixtures.Thus, pretreatment of membranes with either Mg2+ or Ca2+did not affect the magnitude of inhibition.The second set of experiments was designed to determine

whether increasing the levels of Mg2+ or Ca2+ in assaymixtures could overcome CPZ inhibition of CS activity. Here,a differential effect was seen. With Mg2+ in the presence of0.5 mM Ca2+, no change in the ratio of inhibition was detectedin either PM or solubilized fractions as levels of Mg2+ wereraised (Fig. 4, C and D). With Ca2` in the presence of 5 mmMg2+, CS activity was maximal at 0.5 mim and declined asCa2+ levels increased (Fig. 4, A and B). When CPZ was added,the greatest amount of inhibition was seen at low levels ofCa2+. As Ca2+ levels were increased beyond 1 mm, CS becameless sensitive to CPZ inhibition. The observation that Ca2+but not Mg2+ overcomes CPZ inhibition suggests that CPZinterferes with the Ca2+-stimulated component of CS. Alter-natively, Ca2+ may compensate for changes in membranestructure caused by CPZ.The effects of digitonin and PM sidedness were investi-

gated using PM vesicles prepared by aqueous two-phase

24

20

0m 16

3 12Eo5 8E

4

I-

8C4

S2 6cIL

uA 40U2

10 . , ., ., . . I, , 240A. Plasma Membrane C.0 -o- ~~~~Control

-0°>* 200 ILM CPZ Plasma Membrane 160Control

to~~ ~ ~ ~ ~ ~~4-200pMCPZ 120

10 *0

10 40

0~~~~~0

B. Solubilized D.0O -0-- control 0

-- 100 PaM CPZ

0 Colublized 00

D0 Contrd0W0 l4-* 100PMCPZ

.10 0 2 00 2 4 6 S 1 0 0 2 4 6 B 10 12

CALCIUM (mM) MAGNESIUM (mM)

Figure 4. Effect of divalent cation levels on CPZ inhibition of CS. Aand C, PM (5.8 Mg); B and D, solubilized CS (1.75 Mg). Assays were

conducted with Ca2" (A and B) and Mg2" and 1 mm EGTA (C andD) as described in "Materials and Methods."

so'

Fr60

-l

4A 4

200 20 40 60 80

HEAT INACTIVATED PM (gg)

Figure 5. Competition for CPZ by heat-inactivated PM. Assayscontained PM (5.8 ,ug protein), 200 gM CPZ, and a heat-inactivated(55°C for 30 min) PM preparation at the indicated levels. Relativeactivity is based upon duplicate controls lacking CPZ and heat-inactivated PM. The heat-inactivated PM fraction contained nodetectable levels of CS activity.

partitioning. As determined by assays conducted in the ab-sence or presence of 0.01% digitonin, approximately 85% ofthe vesicles were in the right-side-out orientation. Inside-outvesicles were generated using a freeze-thaw treatment (19),yielding approximately 50% of the vesicles in the inside-outorientation. Both vesicle preparations were equally inhibitedby CPZ, indicating no differential effect due to membranesidedness (not shown). The IC50 of CPZ was unaffected byremoval of 0.01% digitonin from assay mixtures (not shown).Thus, CPZ appears to interact with the PM at either face.

Other Factors Affecting Interaction of CPZ with CS

Previous studies employing EM and electron spin reso-nance have demonstrated that CPZ binds strongly to mam-malian membranes (9, 14, 25). The interaction of CPZ withthe plant PM was investigated in several ways. First, wesought to determine whether addition of heat-inactivatedPMs to the CS assay mixtures could reverse CPZ inhibition.In this experiment (Fig. 5), PM was preincubated with CPZand all effectors. Increasing levels of heat-treated PM werethen added, and reactions were initiated by addition of UDP-Glc. CPZ inhibition was overcome in a linear manner toabout 20 Mg of heat-denatured protein, with near completerestoration of activity occurring at 50 ,g. These results suggestthat CPZ-binding to the PM is reversible and has a strongnonspecific component.

Binding of [3H]CPZ to the PM was quantified by Scatchardanalysis (Fig. 6). The shape of the plot is atypical in that theB/F ratio showed an increase (peaking at a B/F ratio of 0.77at 8 Mm cold CPZ) rather than a decline as the concentrationof cold CPZ was raised. Binding was also abnormally high,with 77% of the total added CPZ becoming bound at satu-ration (8.7 Mm CPZ). This suggests that binding of CPZ to thePM is a cooperative process. When low levels of CPZ interactwith the PM, permeabilization of the membrane probablyoccurs. This leads to exposure of additional CPZ-binding

1 930 HARRIMAN ET AL.




0-I1-

0(ICwi

0 2000 4000 6000 8000

B (nmoles)

Figure 6. Scatchard analysis for CPZ binding to KI-washed CS. KI-washed CS (5.4 Mg) in 0.01% digitonin and 50 mm Tris-HCI, pH 7.5,was incubated with 1.6 x 105 dpm of [3H]CPZ and [14C]mannitoland then challenged with various levels of unlabeled CPZ at 20°C.

sites, which results in continued CPZ binding until saturationoccurs. A high-affinity binding component correlatable to aspecific CPZ-binding protein was not found.

UV-lnduced Inhibition and Chemical Modification of CS

Enhanced Inhibition by UV Irradiation

Exposure to UV irradiation dramatically increased the sen-sitivity of CS to CPZ. In the presence of 4 AM CPZ, CS wascompletely inactivated following 2 min of irradiation (Fig. 7).Control experiments showed that PM exposed to 2 min ofUV irradiation in the absence of CPZ retained full activity.However, longer irradiations resulted in a gradual loss ofactivity, with 17% remaining after 20 min (not shown).Inactivation followed a pseudo-first order rate, with calcula-

100

CHLORPROMAZINE (gM)

Figure 7. UV-enhanced inhibition of CS by CPZ. PM-bound CS(2.7 Mg) in 60 ML of 50 mm Tris-HCI, pH 7.5, and CPZ were irradiatedfor 2 min at OC.

TIME (min)

Figure 8. Kinetics of UV-induced CPZ inactivation. PM-bound CS(2.7 Ag) in 60 AL of 50 mm Tris-HCI, pH 7.5, and CPZ were irradiatedwith UV light for various durations at 0°C (see 'Materials andMethods'). The inset showing reaction order was plotted from t1/2values according to the equation: log 1000/tl/2 = nlog[CPZ] - logk2, where n = reaction order (12).

tions indicating the binding of three CPZ molecules for eachunit of CS activity (Fig. 8).

Protection ExperimentsThe aromatic nature of CPZ and the enhanced inactivation

by UV irradiation suggest a free radical mechanism of inhi-bition. Therefore, free radical quenchers such as ascorbateshould protect against inactivation. Ascorbate protected in aconcentration-dependent manner (Fig. 9), with complete pro-tection afforded at 6 mm, supporting a free radical mechanismfor CPZ attack. Other widely utilized antioxidants such asbutylated hydroxyanisole, tocopherol, and propyl gallatewere not tested due to their low water solubility.

so

4 40

0~~~~~~~~~~~~ ~~~UDP-GicLLI 20~ ~ ~ ~ ---CDP-GIc

-U-UDP-GIcNAc

u.o .5 1.0 1.5 2.0

PROTECTANT (gM)

Figure 9. Protection of UV-enhanced CPZ inhibition of CS. Ascor-bate and NDP-sugars were added to 60-ML mixtures of 2 AM CPZ,PM (2.7 Mg), and 50 mm Tris-HCI, pH 7.5, and irradiated for 5 minprior to assay.

ILm

0

0wC.

OC

:i

1931




To investigate further CPZ specificity for substrate andeffector binding sites, protection experiments with NDP-sugars and divalent cations were conducted. Inclusion ofUDP-Glc (in both the absence and presence of divalentcations) protected against UV-induced CPZ inactivation in aconcentration-dependent manner. To determine whetherprotection by UDP-Glc was specific, other NDP-sugars werethen tested. All of the other NDP-sugars tested, such as CDP-Glc (Fig. 9), UDP-GlcNAc (Fig. 9), UDP-glucuronic acid (notshown), and GDP-Glc (not shown), protected CS in a dose-dependent manner similar to ascorbate and UDP-Glc. Glu-cose, however, did not protect. Thus, protection by UDP-Glcwas nonspecific. We believe that protection by the NDP-sugars is due to direct quenching of CPZ radicals by thearomatic purine and pyrimidine moieties, rather than tospecific blocking of CPZ-binding sites (See 'Discussion').

Because studies of the AChR have suggested that CPZlodges in the cation-selective pore (6, 7, 23), we sought todetermine whether the presence of nascent glucan couldprotect against CPZ inhibition. Therefore, CPZ was addeddirectly to assay mixtures actively synthesizing glucan. How-ever, regardless of UDP-Glc levels or incubation time, pro-tection above that which was seen with ascorbate and theNDP-sugars was not observed (not shown). Inclusion of Ca2"during UV irradiation did not protect against inactivation.

Direct Photolabeling with [3H]CPZ

As expected, UV irradiation of CS resulted in covalentincorporation of [3H]CPZ into TCA-precipitable material.However, polypeptide profiles following SDS-PAGE andfluorography were complex, with many polypeptides becom-ing labeled (not shown). Labeling was effectively protectedagainst by addition of ascorbate; however, boiled and un-boiled samples yielded virtually identical patterns. Theselabeling experiments and the Scatchard binding data (Fig. 6)yield results that are highly consistent with one another. Bothexperiments confirm the occurrence of strong nonspecificabsorption of CPZ by the plant PM.One possible mechanism of UV-induced inhibition by CPZ

is the promotion of cross-linking of membrane proteins.Merville et al. (17) and Piette et al. (21) showed photosensi-tized cross-linking of erythrocyte membrane proteins; how-ever, 300 gMCPZ was needed. Our labeling studies employedonly 1.6 FM CPZ and resulted in no detectable cross-linkingof membrane proteins (not shown); therefore, it is highlyunlikely that UV-induced inhibition by CPZ occurs throughprotein cross-linking.

DISCUSSION

Phenothiazine drugs, notably CPZ, have played an impor-tant role in the characterization of cation-dependent enzymes(8, 14, 15, 28) and receptors (6, 7, 23). Therefore, CPZ wasused to probe CS, which is strongly activated by Ca2" andMg2 . The results obtained here show that CPZ is a potentinhibitor of CS, which may be partly due to its interactionwith its Ca2+-binding site. However, during attempts to labelspecifically CS polypeptides with [3H]CPZ, it became evidentthat CPZ is tightly absorbed by the PM, resulting in the

nonspecific labeling of numerous polypeptides. Thus, thesites of action of CPZ with respect to the CS complex prob-ably consist of a combination of the divalent cation site aswell as undefined sites located at the CS boundary-lipidinterface or within the enzyme's hydrophobic milieu.

In many ways, the effects shown by CPZ on plant andmammalian enzymes are quite similar. For example, CPZ wasa potent noncompetitive inhibitor of CS with an IC5o of 100Mm (Figs. 1 and 2). This is the same order of magnitude as

phosphatidate phosphohydrolase, Mg2' ATPase, and Na+-K+ ATPase, which have IC50 values of 200, 120, and 60 gM,respectively (2, 4). A second similarity relates to effects ofCPZ on Ca2' binding. Saturating levels of Ca2+, but not Mg2+,overcame CPZ inhibition (Fig. 4). This finding is similar tothat found in rat brain cytosolic phosphatidate phosphohy-drolase, where Ca2' also specifically counteracted the effectof CPZ inhibition, possibly by displacement of CPZ from itsbinding site (14), and heart microsomal fractions, where CPZdecreased Ca2' binding and uptake by the membranes (4).

This study suggests that CPZ binding to the PM has astrong nonspecific component. This finding is in accordancewith earlier studies showing that cationic amphiphilic drugssuch as CPZ partition into the membrane bilayer, formingnoncovalent complexes with both proteins and lipids (22,33). Rosso et al. (25) showed that partitioning of CPZ intothe erythrocyte membrane was accompanied by a redistri-bution of spin-labeled phospholipids. Because boundary lip-ids of red beet CS are thought to be important for enzymeactivity (26, 27), a major component of CS inhibition by CPZinhibition may be derived from disruption of CS-phospho-lipid interactions. Kauss et al. (13) noted calmidazolium andtrifluperazine inhibition of the Ca2+-mediated stimulation aswell as the enhancement in specific activity by trypsinizationwas not due to Ca2+-calmodulin interference but rather tononspecific interactions, possibly with phospholipids.

In the presence of UV irradiation, CPZ is covalently incor-porated into PM proteins, and CS is rapidly inactivated by afree-radical mechanism. Protection against this inactivationby UDP-Glc suggested that CPZ could be binding directly atthe substrate-binding site. However, the subsequent findingthat a range of other NDP-sugars also protected providedevidence for an alternate mechanism of protection. Becauseglucose did not afford any protection, it is likely that CPZradicals are quenched directly by the aromatic purine andpyrimidine components of NDP-sugars. This finding dem-onstrates that protection experiments in UV photolabelingexperiments using uridine-containing probes must be inter-preted with caution. To establish whether protection occursby specific blockage of UDP-Glc binding sites, it is importantto compare the effects of both uridine and non-uridine-containing nucleotides. A result showing preferential protec-tion by the uridine-containing protectants would argue infavor of a specific interaction of ligand with the enzyme'ssubstrate-binding site.

In conclusion, CPZ is a potent inhibitor of CS in plants.However, it appears to have multiple effects and its apparentlack of specificity for a given protein limits its use as aphotoprobe that could be used to identify polypeptide sub-units. In contrast with the AChR, where CPZ has proved tobe a valuable tool for establishing the molecular architecture

1 932 HARRIMAN ET AL.




favor of a specific interaction of ligand with the enzyme'ssubstrate-binding site.

In conclusion, CPZ is a potent inhibitor of CS in plants.However, it appears to have multiple effects and its apparentlack of specificity for a given protein limits its use as a

photoprobe that could be used to identify polypeptide sub-units. In contrast with the AChR, where CPZ has proved tobe a valuable tool for establishing the molecular architectureof subunits spanning the transmembrane region surroundingits ion channel (6, 7, 23), similar types of experiments cannotbe conducted on CS until it is purified to homogeneity andits subunits unambiguously elucidated. Although CPZ hasthe potential for providing analogous structural informationin the CS system, its use as a tool for identifying subunits inpartially pure preparations seems limited.

ACKNOWLEDGMENTS

Assistance with Scatchard binding experiments provided by Dr.Gerald A. Berkowitz is deeply appreciated. We thank Ms. Ayong Wufor conducting the membrane sidedness experiment.

LITERATURE CITED

1. Bradford MM (1976) A rapid and sensitive method for thequantitation of microgram quantities of protein using theprinciple of protein-dye binding. Anal Biochem 72: 248-254

2. Brindley DN (1985) Intracellular translocation of phosphatidatephosphohydralase and its possible role in the control of glyc-erolipid synthesis. J Prog Lipid Res 22: 115-133

3. Delmer DP (1987) Cellulose biosynthesis. Annu Rev Plant Phys-iol 38: 259-290

4. Dhalla NS, Lee SL, Takeo S, Panagia V, Bhayana V (1980)Effects of chlorpromazine and imipramine on rat heart sub-cellular membranes. Biochem Pharmacol 29: 629-633

5. Frost DJ, Read SM, Drake RR, Haley BE, Wasserman BP(1990) Identification of the UDP-Glc-binding polypeptide ofcallose synthase from Beta vulgaris L. by photoaffinity labelingwith 5-azido-UDP-glucose. J Biol Chem 265: 2162-2167

6. Giraudat J, Dennis M, Heidmann T, Haumont PY, Lederer F,Changeux JP (1987) Structure of the high-affinity bindingsite for noncompetitive blockers of the acetylcholine receptor:[3H]chlorpromazine labels homologous residues in the ,B and-y chains. Biochemistry 26: 2410-2418

7. Giraudat J, Galzi JL, Revah F, Changeux JP, Haumont PY,Lederer F (1989) The noncompetitive blocker [3H]chlorprom-azine labels segment M2 but not segment Ml of the nicotinicacetylcholine receptor a-subunit. FEBS Lett 253: 190-198

8. Hait WN, Lee GL (1985) Characteristics of the cytotoxic effectsof the phenothiazine class of calmodulin antagonists. BiochemPharmacol 34: 3973-3978

9. Hartmann J, Glaser R (1991) The influence of chlorpromazineon the potential-induced shape change of human erythrocyte.Biosci Rep 11: 213-221

10. Hayashi T, Read SM, Bussell J, Thelen M, Lin FC, Brown RMJr, Delmer DP (1987) UDP-Glucose:(1,3)-,B-glucan synthasesfrom mung bean and cotton. Differential effects of Ca2" andMg2+ on enzyme properties and on macromolecular structureof the glucan product. Plant Physiol 83: 1054-1062

11. Jacob SR, Northcote DH (1985) In vitro glucan synthesis bymembranes of celery petioles: the role of the membrane indetermining the type of linkage formed. J Cell Sci Suppl 2:1-11

12. Kasamo K (1988) Essential arginyl residues in the plasma mem-brane H+-ATPase from Vigna radiata L. (mung bean) roots.Plant Physiol 87: 126-129

13. Kauss H, Kohle H, Jeblick W (1983) Proteolytic activation andstimulation by Ca2" of glucan synthase from soybean cells.FEBS Lett 158: 84-88

14. Koul 0, Hauser G (1987) Modulation of rat brain cytosolic

phosphatidate phosphohydrolase: effect of cationic amphi-philic drugs and divalent cations. Arch Biochem Biophys 253:453-461

15. Kumar RV, Panniers R, Wolfman A, Henshaw EC (1991)Inhibition of protein synthesis by antagonists of calmodulinin Ehrlich ascites tumor cells. Eur J Biochem 195: 313-319

16. Ma S, Gross KC, Wasserman BP (1991) Developmental regu-lation of the (1,3)-,B-glucan (callose) synthase from tomato.Possible role of endogenous phospholipases. Plant Physiol 96:664-667

17. Merville MP, Piette J, Decuyper J, Calberg-Bacq CM, Van DeVorst A (1983) Phototoxicity of phenothiazine derivatives. II.Photosensitized cross-linking of erythrocyte membrane pro-teins. Chem-Biol Interact 44: 275-287

18. Moreau RA, Isett TF, Piazza GJ (1985) Dibucaine, chlorprom-azine, and detergents mediate membrane breakdown in potatotuber homogenates. Phytochemistry 24: 2555-2558

19. Palmgren MG, Askerlund P, Fredrikson K, Widell S, Som-marin M, Larsson C (1990) Sealed inside-out and right-side-out plasma membrane vesicles. Plant Physiol 92: 871-880

20. Peters J, Longhi D, Berkowitz G (1991) Preparation and use ofliposomes with incorporated chloroplast envelope proteins tostudy ion transport (abstract No. 207). Plant Physiol 96: S-37

21. Piette J, Decuyper J, Merville-Louis MP, Van de Vorst A(1986) Biomolecular photoalterations mediated by phenothi-azine derivatives. Biochimie 68: 835-842

22. Ondrias K, Stasko A, Misik V, Reguli J, Svajdlenka E (1991)Comparison of perturbation effect of propanolol, verapamil,chlorpromazine and carbisocaine on lecithin liposomes andbrain total lipid liposomes. An EPR spectroscopy study. Chem-Biol Interact 79: 197-206

23. Revah F, Galzi JL, Giraudet J, Haumont PY, Lederer F, Chan-geux JP (1990) The noncompetitive blocker [3H]chlorproma-zine labels three amino acids of the acetylcholine receptor ysubunit: implications for the a-helical organization of regionsMII and for the structure of the ion channel. Proc Natl AcadSci USA 87: 4675-4679

24. Roberts AW, Haigler CH (1990) Tracheary-element differentia-tion in suspension-cultured cells of Zinnia requires uptake ofextracellular Ca2". Experiments with calcium-channel blockersand calmodulin inhibitors. Planta 180: 502-509

25. Rosso J, Zachowski A, Devaux PF (1988) Influence of chlor-promazine on the transverse mobility of phospholipids in thehuman erythrocyte membrane: relation to shape changes.Biochim Biophys Acta 942: 271-279

26. Sloan ME, Rodis P, Wasserman BP (1987) CHAPS-solubiliza-tion and functional reconstitution of ,B-glucan synthase fromred beet root. Plant Physiol 85: 516-522

27. Sloan ME, Wasserman BP (1989) Susceptibility of UDP-glucose:(1,3)-fl-glucan to inactivation by phospholipases and trypsin.Plant Physiol 89: 1341-1344

28. Walton PA, Possmayer F (1989) The effects of Triton X-100 andchlorpromazine on the Mg24-dependent and Mg24-independ-ent phosphatidate phosphorylase activities of rat lung.Biochem J 261: 673-678

29. Deleted in proof30. Wasserman BP, Frost DJ, Lawson SG, Mason TL, Rodis P,

Sabin RD, Sloan ME (1989) Biosynthesis of cell wall polysac-charides: membrane isolation, in vitro glycosyl transferaseassay and enzyme solubilization. In H-F Linskens, JF Jackson,eds, Modem Methods of Plant Analysis, New Series Vol 10.Springer-Verlag, Heidelberg-New York, pp 1-11

31. Wasserman BP, Wu A, Harriman RW (1992) Probing themolecular architecture of (1,3)-13-glucan (callose) synthase:polypeptide depletion studies. Biochem Soc Trans 20: 18-22

32. Wu A, Harriman RW, Frost DJ, Read SM, Wasserman BP(1991) Rapid enrichment of UDP-glucose: (1,3)-il-glucan (cal-lose) synthase from Beta vulgaris L. by product entrapment.Entrapment mechanisms and polypeptide characterization.Plant Physiol 97: 684-692

33. Yamaguchi T, Watanabe S, Kimoto E (1985) ESR spectralchanges induced by chlorpromazine in spin-labeled erythro-cyte ghost membranes. Biochim Biophys Acta 820: 157-164

1933



inhibition and ultraviolet-induced chemical modification ... · mechanismof chlorpromazine...

Documents