inhibition of the adenylylation of liver plasma membrane-bound...

Biol. Chem. Hoppe-SeylerVol. 374. pp. 133-141. February 1993

Inhibition of the Adenylylation of Liver Plasma Membrane-BoundProteins by Plant and Mammalian Lectins

Estcban SAN JosiV1. EduardoViLLALOBoa, Hans-J. GABiush and Antonio VILLALOBO"

:| Ins t i tu tode Invcstigacioncs Biomedieas, Conscjo Superior dc Investigaciones Cientificas. Madrid, Spain andh Institut für Pharmazeutische Chemie der Philipps-Universität.Abteilung Glykobiochcmie und AngcwandteTumorlektinologic. Marburg. Germany

(Received 6 August 1992)

Summary: Liver plasma membrane contains fourmajor (130-kDa, 120-kDa, 110-kDa and 100-kDa)sialic acid-containing glycopolypeptides that are ableto undergo adenylylation, as well as phosphorylation(San Jose et al."(1990) J. Biol. Chem. 265; 20653-20661). To gain insight into the regulation of theseprocesses, lectins are employed to probe the extent ofinfluence of their interaction with membrane frac-tions for these reactions. We demonstrate that the -galactoside-specific lectins from bovine heart andmistletoe at low concentrations inhibit the adenylyla-lion of this set of plasma membrane glycopolypep-tides.The extent of phosphorylation of these polypep-tides is also reduced although to a lesser degree. Con-canavalin A, too, inhibits the adenylylation of theplasma membrane glycopolypeptides, althoughhigher concentrations of this lectin were required,whereas wheat germ lectin has only a very small in-hibitory effect. The adenylylable polypeptides were

isolated by concanavalin A-agarose chromatographyupon elution with mannose. In agreement with thisresult, control experiments with a panel of neoglyco-proteins indicate that mannose residues appear to berequired for the concanavalin -induced inhibition ofthe adenylylation. Neoglycoproteins containing man-nose 6-phosphate, lactose, fucose, or sialic acid in-stead of mannose lack the ability to protect theadenylylation from the inhibitory action of con-canavalin A. In contrast, none of the above-men-tioned neoglycoproteins, nor asialofetuin, norgalac-tose-containing saccharides protect the adenylylationagainst the inhibitory effect of both the mistletoe andbovine heart lectins, emphasizing the importance ofeither high affinity carbohydrate ligands in the overallprocess, or other ligand sites for the lectins beside car-bohydrates to affect the regulation of the adenylyla-tion system.

Key terms: Adenylylation, phosphorylation, lectins, glycoproteins.

A great majority of integral plasma membrane pro-teins from eukaryotic cells are glycosylated. Theglycosylation of plasma membrane proteins should

not just be considered as a biologically inert structuraladdition to the protein backbone. Conversely, impor-tant cellular functions, such as intercellular recogni-

Abbreviarions:Hepes, 4-(2-hydroxyethyl)-l-piperazinethanesulfonic acid; SDS, sodium dodecyl sulfatc; PMSF, phenylmcthanesulfonyl fluoride;ML-I, mistletoe lectin I; 14k-bh. bovine heart 14 kDa lectin: ConA, concanavalin A; WGA, wheat germ lectin (agglutinin); Lac,lactose; -Mel, a-D(+)-melibiose (6-O-a-D-galactopyranosyl-D-glucose); Man, mannose; Gal, o(+)-galactose; 4-ß-Gal-Man,4-O-ß-galactopyranosyl-D-mannopyranose; 6-ß-Gal-Gal, 6-O-/3-galactopyranosyl-D-galactose; 4-a-Gal-Gal, 4-O-a-D-galacto-pyranosyl-D-galactopyranose; 3-ß-GaI-Ara, 3-0-ß-n-galactopyranosyl-D-arabinose; 4-ß-Gal-Fru, 4-O-ß-D-galactopyranosyl-D-fructofuranose (lactulose); ASF, asialofetuin; Lac-BSA, neoglycoprotein containing ß-lactose; Man-BSA, ncoglycoprotein con-taining -D-mannose; Man-6-P-BSA. neoglycoprotein containing a-D-mannosc-6-phosphatc; Fuc-BSA, neoglycoprotein contain-ing -L-fucose; Sia-BSA, neoglycoprotein-containing sialic acid; BSA, bovine serum albumin; TGF-y2. transforming growth fac-tor-y2; poly(Glu:Tyr), co-polymer of glutamic acid and tyrosine.

Copyright © by Walter de Gruyter & Co · Berlin · New YorkBrought to you by | CSIC - Unidad Coordinacion BibliotecasAuthenticated | 161.111.180.191

Download Date | 6/26/13 12:38 PM

134 E. San Jose. E.Villalobo. B_-J.jClabiusand A.Villalobo Vol. 374 (1993)

tion and cell adhesion processes, certain hormonal ac-tions, and immunological recognition events can bemediated by the carbohydrate residues of proteins.The diverse sequences and complex branching of thecarbohydrate residues in glycoproteins are likely tocontain important informational clues that have to bedeciphered for appropriate action of soluble or mem-brane-bound ligands. Suitable ligands may well belectins that play a physiological role in the cell and/orcan be involved in pathogenic actions'1'3'.Lectins were first recognized as proteins from plantsby their characterized ability to bind to the carbo-hydrate residues of glycoconjugates'^.The observa-tion that lectins are present in animal cellsi:>*6J has un-doubtedly contributed to the realization that lectins,in addition to being useful biochemical tools171, are in-volved in important cellular functions.Lectins mediate their cellular actions by yet unknownsignal transduction mechanisms, which function be-tween their plasma membrane-bound glycoproteinreceptor(s) and intracellular target(s). Regulation ofresponse to hormones and gangliosides is supposed toencompass such protein-carbohydrate recognitionmechanisms^8"10!. The reversible interconversion ofenzymes is at the crossroads of signal transduction inmultiple cellular systems. To investigate the role ofthis type of molecular recognition event, we haveexplored the action of a mammalian lectin, namelythe /3-galactoside-specific lectin of 14-kDa, as well asthree plant lectins on the function of a recently de-scribed'11' rat liver plasma membrane-bound adenyly-lation system(s). In this report, we present evidencethat the adenylylation of four major plasma mem-brane-bound glycopolypeptides can be inhibited bythese lectins. Remarkably, the phosphorylation of thesame polypeptides is affected by lectins to a far lesserextent. No significant effects of these lectins on thephosphorylation of other liver plasma membrane pro-teins were detected, when crude or further purifiedmembrane fractions were analysed.

Hepes and divinyl sulfone were from Merck, carbohydrate-freebovine serum albumin used for synthesis of neoglycoproteins wasfrom Biomol. Sepharose 4B from Pharmacia, and molecular massstandards for electrophoresis were purchased from. Bio-Rad. Allother chemicals used in this work were of analytical grade.

Preparation of liver plasma membrane fractionsCrude and further purified liver plasma membrane fractions frommale Sprague-Dawley albino rats (250-300 g) were prepared fol-lowing the method of Brown et al J12' as modified by us'11·13'.The en-richment we obtained in plasma membrane enzymatic markers (5'-nucleotidase, alkaline phosphatase or phosphodiesterase), with re-spect to the crude homogenate was 15-fold ± I (average ± SE) inthe crude fractions (27 preparations) and 41-fold ± 6 in the furtherpurified fractions (10 preparations). The crude fractions (20 to 50mg of protein per preparation) were obtained from the first sucrosegradient centrifugal ion. and the further purified fractions (5 to 10mg of protein per preparation) were the light fractions obtainedfrom the second sucrose gradient centrifugation. Crude plasmamembrane fractions were used for most of the adenylylation exper-iments, whereas the further purified plasma membrane fractionswere used for most of the phosphorylation experiments, becausethe extent of phosphorylation of the polypeptides under study wasbetter observed in the further purified membranes.The legends tothe figures give detailed information on which type of preparationhas been used in each series of experiments.

Adenylylation and phosphorylation experimentsUnless indicated otherwise, our assays were carried out as follows:an adequate amount of plasma membranes (28 to 115 μ-g of protein)was incubated at 37 °C for l min (phosphorylation experiments) or5 min (adenylylation experiments) in a total volume of 100 or 200μ/ of a medium containing 20mM Na-Hepes (pH 7.4), 0.1% (w/v)Triton X-100, and 10 μΜ of either [γ-32Ρ]ΑΤΡ (2-3 μ€Ί) (phos-phorylation experiments) or [a-32P]ATP (1.5-4 μ,Ο) (adenylyla-tion experiments). Triton X-100 was added to permeabilize mem-brane vesicles, in order to allow the binding of ligands to both sidesof the membrane. Some phosphorylation experiments were per-formed in the presence of 6mM MgCli. Lectins were incubated withthe membranes at 4 °C for 20 to 30 min prior to the assay to allowbinding to membrane ligands. Some of the adenylylation experi-ments were performed in the presence of ΙΠΙΜ EDTA, since thismetal-chelating agent yields 20 to 40-fold increases in the adenyly-lation level due to the inhibition of ATPases, therefore increasingthe availability of ATP for the adenylylation process. The reactionwas stopped by addition of ice-cold trichloroacetic acid to a finalconcentration of 10% (w/v), and the pellet of precipitated proteinswas first neutralized with a minimum volume of 1.25MTris/HCl (pH8.8), the solution was gently stirred with a glass-rod to dissolve theprotein precipitate and processed for electrophoresis and au-toradiography. We found that the inhibitory action of lectins wasnot significantly affected by the concentration of plasma mem-brane protein in the assay system within the range used.

Materials and Methods

ChemicalsThe radiolabeled compounds [γ-32Ρ]ΑΤΡ (triethylammonium salt)3000 Ci x mmor1) (1 Ci = 37 GBq), [a-32P]ATP (tetra[triethylam-monium] salt (3000 Ci x mmol"1)) were purchased from New Eng-land Nuclear and X-Omat AR x-ray blue-sensitive films were fromEastman Kodak. Concanavalin A, concanavalin A-agarose, wheatgerm lectin, 4-nitrophenyl glycosides, ATP (sodium salt), lactose(a and β enantiomeric forms), o(+)-mannose, a-D(+)-melibiose,D(+)-galactose, 4-O- -galactopyranosyl-D-mannopyranose, 6-0-j8-galactopyranosyl-D-galactose.4-O-a-D-galactopyranosyI-D-galacto-pyranose, 3-O-/3-D-galactopyranosyl-D-arabinose, and 4-O-/3-D-galactopyranosyl-D-fructofuranose were obtained from Sigma.

Synthesis of neoglycoproteinsSugar-free bovine serum albumin was used as carrier protein for theattachment of glycosides and coupled to the p-isothiocyanate de-rivatives of p-aminophenyl glycosides to yield neoglycoproteinscontaining on average 30 ± 4 carbohydrate moieties per carriermolecule, as described in detail elsewhere^14'.

Purification of lectins from mistletoe and bovine heartFractionation of extracts from dried mistletoe leaves and frombovine heart on lactose-Sepharose 4B, obtained by divinyl sulfoneactivation of the matrix and subsequent coupling of the ligand, elu-tion by 0.3M lactose, and quality controls of the purified lectinshave been described in detail previously1'51.

Brought to you by | CSIC - Unidad Coordinacion BibliotecasAuthenticated | 161.111.180.191


Vol. 374 (1993) Adenylylation Inhibition of Membrane-Bound Proteins by Lectins 135

Preparation oflectin-bearing matricesDivinyl sulfone-activated Sepharose 4B (10 ml) was incubated with3 mg lectin dissolved in 8 ml of 0.2M ^HPOyKH^PO., (pH 8.6),0.9% (\v/v) NaCl. for 12 h at 4 °C. The bovine heart lectin wasstabilized by carboxymethylation prior to coupling according toPowell and Whitney'16'. The coupling yield was determined by thedye-binding assay with albumin as standard'17', and the capacity ofthe matrix was checked with asialofetuin, as described recently'18'.

Affinity chromatography with immobilized mistletoeand bovine heart lectinsPurified plasma membranes (43 mg of protein) were solubilized ina medium containing 20mMTris/HCl (pH 7.8). 0.5M NaCl. 4mM 2-mercaptoethanol. 2mM EDTA, lOOmM lactose. 0.5% (w/v)TritonX-100. 0.1% (w/v) Na-deoxycholate. and protease inhibitors(O.lmMPMSF. 5/ig x m/"1 leupeptin.andSMg x m/'1 amipapain).After dialysis overnight against 20mM Tris/HCl (pH 7.8), 150m,MNaCl, 2mM EDTA, 2mM 2-mercaptoethanol, and 0.02% (w/v)Tri-ton X-100. to remove the sugar, the extract (half volume each) waspassed through the mistletoe lectin 1-Sepharose or the bovine heart14-kDa lectin-Sepharose columns (5-m/ bed volume containing280Mg of protein x m/~ l) equilibrated with 20mMTris/HCl (pH 7.8),ISOm.M NaCl, ImM dithiothreitol, and 0.02% (w/v) Triton X-100.After extensive washing with this buffer the elution was performedwith the same buffer containing 0.3M lactose.The eluted fractionswere dialysed first against 20mM Tris/HCl (pH 7.8), containing2mMEDTA. 4niM 2-mercaptoethanol. and 0.05% (w/v) Triton X-100,and thereafter against the same buffer containing only 7.5mMTris/HC1 (pH 7.8). After chloroform extraction to remove the TritonX-100, the samples were lyophilized and stored at - 60 °C untilused. The resin with immobilized mistletoe lectin yielded 65 μ-g ofprotein, and the bovine heart lectin-Sepharose column 105 μ§ ofprotein according to the dye-binding assay.

Concanavalin A-agarose chromatographyPurified plasma membranes (5 to 7.5 mg of proteins) were sol-ubilized for 10 min at 4 °C in 1.5 to 3 ml of a medium containing25niM Na-Hepes (pH 7.4), and 1% (w/v)Triton X-100, and ccn-trifuged at 100000 x gmax for 15 min.The resulting supernatant waspassed through a concanavalin A-agarose column (5 m/bed volumecontaining 14 mg of protein x m/'1), equilibrated with 25mM Na-Hepes (pH 7.4), ImM CaCl2 (when added), and 1% (w/v)TritonX-100. Thereafter, the bound proteins were eluted with a buffercontaining 25mM Na-Hepes (pH 7.4), 1% (w/v)Triton X-100, ImMCaCl2 (when added), and 50mM mannose.The eluted fractions (0.8ml each) were collected and stored at - 70 °C until use.

Of her analytical proceduresSlab gel electrophoresis (30 to 70 μ-g of protein per lane) was per-formed according to the method of Laemmli1191 at 12 m A overnightin linear gradient 5-20% (w/v) polyacrylarnide gels in the presenceof 0.1% (w/v) SDS at pH 8.3. Alternatively, slab gels with 10%(w/v) polyacrylarnide in the runninggel and3% (w/v) in the stackinggel were used.The gels were stained with Coomassie Brilliant BlueR-250, and dried under vacuum at 70 °C on Whatman 3MM filterpaper. Alternatively, silver staining'20' was employed for visualiza-tion of protein bands when less than 10 /xg were applied per lane.The x-ray films were exposed in the dark at - 20 °C for appropriateperiods of time (1 to 5 days) to obtain autoradiographs.The inten-sities of the labeled bands on the autoradiographs were measuredin a scanning photodensitometer. We demonstrated that the opticaldensities of the labeled bands in the autoradiographs were linearlyproportional to the amount of 32P bound to the proteins in the ex-perimental conditions used.The method of Lowry et al.1211 was generally used to determine pro-tein concentrations after precipitating the proteins with 10% (w/v)trichloroacetic acid, using bovine serum albumin as a standard, the

only exception being the determination of the amount of protein inthe eluted fractions from the lectin chromatography columns withthe dye-binding assay of Redinbaugh and Campbell'17'.

Results and Discussion

Effects of lectins on the adenylylation of plasmamembrane-bound glycoproteinsWe have previously demonstrated that the incubationof isolated rat liver plasma membrane fractions with[a-32P]ATP or [2,5',8-3H]ATP results in the labelingof four major glycopolypeptides of 130-kDa, 120-kDa, 110-kDa and 100-kDa1111. This labeling repre-sents the adenylylation of these proteins. Further-more, we have demonstrated that this adenylylationis a reversible process, and that the 120-kDa and 110-kDa polypeptides form a heterodimer linked by disul-fide bridge(s)'nl In addition, we demonstrated thepresence of a 86-kDa phosphodiesterase in thesemembranes that forms a catalytic intermediate^111.We have suggested that the adenylylation of thesepolypeptides probably represents posttranslationalmodifications of a plasma membrane system(s) thatpossibly plays a regulatory role of yet unknown na-ture1111. It is interesting to refer to prokaryotes, whereseveral enzymes have been described as being regu-lated by adenylylation122"271. However, there is no de-finitive information available about similar regula-tory systems in eukaryotic cells.While attempts to clarify the nature of the adenyly-lated proteins are currently being undertaken, eluci-dation of the regulation of this type of modificationsimilarly warrants attention. To initiate the study ofthis aspect, we focused on a class of proteins that canspecifically interact with the carbohydrate part ofmembrane glycoproteins which are adenylylated,namely lectins. Lectins are supposed to be involved inbiosignaling processes, as emphasized by their induc-tion of mitogenesis upon binding to the membrane ofresponsive cells^1. In this study we used a mammalian

-galactoside-specific 14-kDa lectin of rather ubi-quitous nature128'291 isolated from bovine heart, aswell as several plant lectins.In Fig. 1 (panel A) the [c*-32P] ATP-labeled patterns ofthe plasma membrane polypeptides in the absence(none) and presence of wheat germ lectin (WGA),concanavalin A (ConA), mistletoe lectin I (ML-I),and the 14-kDa bovine heart lectin (14k-bh) are pre-sented. A strong inhibition of the adenylylation of the130-kDa, 120-kDa, 110-kDa and 100-kDa polypep-tides in the presence of mistletoe and bovine heart lec-tins, and to a lesser extent in the presence of con-canavalin A was observed. In comparison, wheatgerm lectin exhibits a slight inhibitory effect. As can



136 E. San Jose. E.Villalobo. H.-J. Gahiusand A.Villalobo Vol. 374 (1993)

ωcoc

200 - .·.·"'.:·

116.3 ^

45

31

W 3$; *$$ ̂

150

125

100cgl 75>̂s

l 50

25

-//-


wheat germ lectin do not inhibit more than 20% theadenylylation of these polypeptides. Therefore, thefailure of wheat germ lectin to exert a significant ef-fect on the adenylylation suggests that binding to ac-cessible W-acetylglucosamine or /V-acetylneuraminicacid moieties in glycoconjugates is not too relevantfor the net adenylylation process.We also present in Fig. 1 (panel C) the effects of theselectins on the catalytic intermediate of the 86-kDaphosphodiesterase. It can be observed that both, mis-tletoe and bovine heart lectins are strong inhibitors ofthe phosphodiesterase. We calculate an apparent K\ ofapproximately 10 /*g x m/~l for both lectins. In con-trast, 500 /xg x m/"1 of wheat germ lectin only inhibitsthe formation of this catalytic intermediate by 25%.With respect to concanavalin A, an increase in thelevel of this intermediate of more than 2-fold is detect-able by addition of up to 500 μ£ x m/"1 to the assay.

We have also demonstrated that a fraction of theadenylylable polypeptides bind to concanavalin A-agarose in the absence as well as in the presence of cal-cium ions, and are eluted in the presence of mannose(results not shown).To assess the enzymatic activities of the purified mate-rial, we have carried out phosphorylation experi-ments using the fractions eluted with mannose from

£S

138 E. San Jose. E. Villalobo, H.-J. GabiusandA. Villalobo Vol. 374 (1993)

the concanavalin A-agarose column. In Fig. 3 the pat-terns of labeling obtained with [a-32P]ATP (panel A),and [γ-32Ρ]ΑΤΡ (panel B) are compared.It can be seen that the four major polypeptides (130-kDa. 120-kDa. 110-kDa, and 100-kDa) are adenyly-lated (labeling by [a-32P] ATP), not only in their mem-brane-bound form (panel A, lane I), but also in theirsolubilized form obtained from the concanavalin A-agarose column (panel A, lane 2). We performed theadenylylation assays of the polypeptides with themannose-eluted fractions after they were eluted fromthe concanavalin A-agarose column. Therefore, it ispossible that the polypeptides under study are under-going an autoadenylylation process. Alternatively, we

200 ^

66.2 ^

45 ».

31 »,

21.5 ^

14.4 ^

200

116.397.4

66.2

45

31

21.5

14.4

Fig. 3. A protein kinase activity is present in the concanavalinA-agarose-elutcd fractions.(Panel A) Purified plasma membranes (40/xg of protein) (lane1), and the peak from the mannose-eluted fractions from theconcanavalin A-agarose column (80μ/) (lane 2) were assayedat 37°C for 5min in 200μ/ of a medium containing 15mM Na-Hepes (pH 7.4), ΙΙΏΜ EDTA, and ΙΟμΜ [α-32?]ATP. (Panel B)Purified plasma membranes (40/ig of protein) (lane 1), and thepeak from the mannose-eluted fractions from the concanavalinA-agarose column (80μ,/) (lanes 2 to 4) were assayed at 37°Cfor i min in ΙΟΟμ/ of a medium containing 15m.\i Na-Hepes(pH 7.4). 50μ§χηιΓ1 histone (lane 3 only), KX^gxm/'1

poly(Glir.Tyr) (lane 4 only), and ΙΟμΜ [γ-32Ρ]ΑΤΡ. The reac-tions were stopped with 10% (w/v) of ice-cold trichloroaceticacid, and the precipitated protein processed forelectrophoresisand autoradiography as described in Materials and Methods.A representative autoradi graph is presented. Similar resultswere obtained when the assays were carried out in the presenceof 6rmi MgG2.

may have copurified by the concanavalin A-agarosecolumn the polypeptides that can be adenylylated andan exogenous adenylylation system, responsible forthe adenylylation of these polypeptides.Fig. 3 shows that the four polypeptides under studyare also phosphorylated (labeling by [γ-32Ρ]ΑΤΡ) intheir membrane-bound form (panel B, lane 1). Intheir solubilized form eluted with mannose from theconcanavalin A-agarose column, however, the onlylabeled band observed corresponds to the 130-kDapolypeptide (panel B, lane 2). We can not excludethat the other lower molecular mass polypeptidescould be detected by prolonged exposure of the filmor may be phosphorylated under other experimentalconditions. However, these experiments suggest thatthe phosphorylation systems for the 120-kDa, 110-kDa, and 100-kDa polypeptides on one hand, and thephosphorylation system for the 130-kDa polypeptideon the other, are different.We also show that the fractions eluted from the con-canavalin A-agarose column are able to phosphory-late exogenous substrates, such as histones (panel B,lane 3). However, phosphorylation of a polymer ofglutamic acid and tyrosine was not detected (panel B,lane 4). Phosphorylated poly(Glu:Tyr) should havemigrated in these gels as a broad smear approximatelybetween 120-kDa and 25-kDa, as demonstrated byphosphorylation of this co-polymer by an epidermalgrowth factor receptor preparation (results notshown). In over-exposed autoradiographs, how-ever, we detected trace levels of phosphorylatedpoly(Glu:Tyr). Nevertheless, it is apparent that thefractions eluted from the concanavalin A-agarose col-umn contain a protein kinase activity that does not ap-pear to be significantly directed to tyrosine residues.

Mistletoe and bovine heart lectins bind to differentliver plasma membrane glycoproteinsThe mistletoe and bovine heart lectins are known tointeract with -galactoside-containing glycoproteins.However, other types of interactions of these lectinswith their target proteins, including protein-proteininteractions are not excluded. Indeed, the mamma-lian lectin has been described as a cell growth reg-ulator, and as TGF-y2, independent of its carbo-hydrate-binding activity, underscoring its bifunc-tional nature'31'32^ The patterns of polypeptideseluted with lactose from two columns prepared by co-valent binding of each lectin to a matrix are shown inFig. 4. They clearly show that both lectins interactwith different glycoproteins from liver plasma mem-brane. Identical monosaccharide specificity of twolectins is thus not necessarily an indication of identicalaffinities to tissue ligands.



Vol. 374 (1993) Adenylylation Inhibition of Membrane-Bound Proteins by Lcctins 139

97.4

66

4536

29

14.3

containing lactose residues do not prevent the inhibi-tory action of these lectins.There are two main possible explanations of these re-sults: i) the affinity of these lectins for the 0-galac-toside residues and their unknown subterminal exten-sion of the target proteins and/or the adenylylationsystem(s) is far higher than for the lactose-containingneoglycoprotein, free lactose or the other free j8-galactosides used; or alternatively ii) we are dealingwith protein-protein interactions. We can exclude thatthe simple sugars used could not compete for lectinsbound to more complex protein carbohydrate ligandssince it has been demonstrated that galactose effec-tively reduces binding of the lectins to the carbo-hvdrate chains of asialofetuin^33l

Fig. 4. Mistletoe and bovine heart lectins ha\fe different targetsfor liver plasma membrane polypeptidcs.Polypeptides isolated from the 14-kDa bovine heart lectin affini-ty chromatography column (lane a), and from the mistletoe lec-tin I affinity chromatography column (lane b) upon elution withlactose as described in Materials and Methods. The sampleswere analysed by polyacrylamide gel electrophoresis in thepresence of SDS, followed by silver staining.

Neoglycoproteins and asialofetuin fail to protectthe adenylylation process from the inhibitory actionof mistletoe and bovine heart lectinsDue to the proven ability of the mammalian lectin tointeract with proteins independently of its capacity toact as sugar receptor, we tested the above-mentionedneoglycoproteins, as well as asialofetuin, for theirpossible protective effects on the mistletoe andbovine heart lectins-induced inhibition of adenylyla-tion. Remarkably, none of these neoglycoproteins,nor asialofetuin, prevent the inhibition of theadenylylation produced by mistletoe lectin or bovineheart lectin (results not shown). In all cases both lec-tins were still able to inhibit the adenylylation of theplasma membrane glycopolypeptides even in thepresence of l mg x m/~l of the neoglycoproteins or ofasialofetuin in the assay system.To corroborate these results with the neoglycopro-teins, we have assayed these lectins in the absenceand presence of galactose, mannose and several free

-galactosides, as well as α-galactosides, up to a con-centration of 20mM.Lactose, melibiose, 3- -Gal-Ara, 4- -Gal-Fru, 4-/3-Gal-Man, 6-0-Gal-Gal, and 4-a-Gal-Gal, as well asgalactose and mannose fail to prevent the inhibitoryaction of mistletoe lectin or bovine heart lectin (re-sults not shown). Overall, these experiments supportour previous results, in which the neoglycoprotein-

The phosphorylation of the plasma membraneglycoproteins that can be adenylylaied is alsoinhibited by lectins but to a lesser extent thanadenylylationPhosphorylation of the proteins that are adenylylatedis better observed in highly purified plasma mem-brane fractions. There fore, we used these fractions toperform phosphorylation experiments. Fig. 5 pre-sents the effect of mistletoe lectin and bovine heartlectin on the phosphorylation (label by [γ-32Ρ] ATP) ofproteins in a further purified plasma membrane frac-tion (panel B).The phosphorylation of the 130-kDa,120-kDa, 110-kDa, 100-kDa polypeptides that can beadenylylated, and the phosphorylation of the 86-kDaphosphodiesterase are slightly inhibited by both lec-tins. As a control, we present in panel A of the samefigure the effect of identical concentrations of bothlectins on the adenylylation of the polypeptides understudy. These lectins inhibit the adenylylation of thesepolypeptides far more efficiently than their phos-phorylation. Furthermore, these lectins do not signifi-cantly affect the phosphorylation of other proteinsobserved in crude membrane fractions (results notshown).The levels of phosphorylation of the 130-kDa, 120-kDa, 110-kDa and 100-kDa plasma membrane poly-peptides that can be adenylylated, and the phospho-rylation of the 86-kDa phosphodiesterase are inhi-bited by progressively higher concentrations of differ-ent lectins. In Fig. 5, panel C the inhibitory action ofmistletoe lectin (open triangles), bovine heart lectin(filled triangles), and concanavalin A (open circles)are presented. We calculated K\ values for mistletoeand bovine heart lectins of 80 and 100 μ% x m/'1, re-spectively, from these plots. In contrast, a concentra-tion of concanavalin A as high as 500 /xg x m/*"1 onlyinhibits the phosphorylation of these polypeptidesby 25%.



140 E. San Jose, E.Villalobo, H.-J. Gabiusand A. Villalobo Vol. 374 (1993)

ML-I 14k-bh- +

ML-I 14k-bh- +

200 -

116.397.4 ^

66.2 ^

45 -

31 ~-

21.5 -

14.4 ^

150

125

I1 100

t 75ox:8 50

25

ο1-^Ο

Fig.5. Comparative effects of lectins on the ade-nylylation and phosphorylation of the 130-kDa.120-kDa. 110-kDa and 100-kDa polypeptides thatcan be adenylylated, and on the phosphorylation ofthe 86-kDa phosphodiesterase.(Panel A) Crude plasma membranes (40/ig of pro-tein) were assayed at 37°C for 5 min in 100μ,/ of amedium containing 15mMNa-Hepes(pH 7.4).0.1%(w/v) Triton Χ-10θ" ΙΟμΜ [α-32Ρ]ΑΤΡ, and in the ab-sence (-) or presence ( + ) of 100μgx m/~ ' mist-letoe lectin \ (ML-I), or 100/tg x m/"1 bovine heart14-kDa lectin (14k-bh). (Panel B) Purified plasmamembranes (92μg of protein) were assayed forImin at 37°C in ΙΟΟμ/ of a medium containing15niM Na-Hepes (pH 7.4), 0.1 % (w/v) Triton X-100 ,̂ΙΟμΜ [γ-32Ρ]ΑΤΡ, and in the absence (-) or pres-ence ( + ) of the indicated lectins as above. Thereaction was stopped with 10% (w/v) ice-cold tri-chloroacetic acid, and the precipitated proteins pro-cessed for electrophoresis and autoradiography asdescribed in Materials and Methods. Representa-tive an to radiographs are presented. (Panel C). Aplot representing the action of different concentra-tions of lectins on the average phosphorylation ofthe 130-kDa, 120-kDa, 110-kDa and 100-kDa ade-nylylable polypeptides and the phosphorylation ofthe 86-kDa phosphodiesterase is presented. The as-says were carried out as above with further purifiedplasma membrane fractions, except for the follow-ing amount of membrane proteins: 28μg (open cir-cles), 28μg (open triangles), 92μg (filled trian-gles), and the indicated concentrations of concana-valin A (open circles), mistletoe lectin I (open trian-gles), or bovine heart 14-kDa lectin (filled trian-gles).

10 100[Lectin] ttg/m/l

1000

We must emphasize that we do not yet know whetherthe adenylylation system(s) is(are) distinct from thetarget adenylylated proteins. Similarly, it is not clearwhether all observed phosphorylation in these pro-teins are autophosphorylation processes or whetheran exogenous protein kinase(s) is(are) involved.Therefore, the chain of events in the mechanism ofaction of lectins on the adenylylation and phosphory-lation of the plasma membrane polypeptides understudy can at present not be explained at a molecularlevel.The finding that the adenylylation and the phosphory-lation of these polypeptides are not equally inhibitedby lectins, as inferred by dissimilar K\ values, suggeststhat the systems responsible for both posttransla-

tional modifications of these plasma membrane pro-teins are indeed different. Furthermore, it suggeststhat these lectins inhibit the two processes by differ-ent mechanisms.Nevertheless, it is reasonable to keep in mind that theadenylylation-deadenylylation system and/or thepolypeptides that can be adenylylated (if different)from the plasma membrane could be physiologicaltargets for the mammalian lectin. Since triggering re-sponses by a lectin may require occupation of otherstructures in addition to carbohydrate-binding sites,as demonstrated for wheat germ lectin-induced syn-thesis of decay-accelerating factor in human endothe-lial cells1341, description of such ligands will be helpfulto clarify interactions of a potent lectin within the




membrane. Thus, further work should be done toelucidate whether or not the interaction of lectinswith the adenylylation and phosphorylation sys-tem(s) involved in the posttranslational modificationof these plasma membrane glycoproteins belongs tothe physiological steps involved in the still elusivemechanism of signal transduction by these carbo-hydrate-binding proteins.

The expert technical assistance of K.P. Hellmann is gratefullyacknowledged. This work was supported in part by grants (toA.V.) from the Direction General de Investigation Cientifica yTecnica (PB 89-0079). and from the Consejeria de Education dela Comunidad de Madrid (C174-90 and 366/92) Spain, grants(toH.-J.G.)fromtheDr. M. Sclieel-SiifmngfiirKrebsforschungand the BMFT program Alternative Methoden der Krebsbe-kämpfung, Germany, and the Acciones Imegradas (42A) be-tween Germany and Spain (to H.-J.G. andA.V.). E.S.J. is therecipient of a predoctoral fellowship from the Departamentode Education, Universidades e Investigation del GobiernoVasco.

References

1 Liener, I.E.. Sharon. N. & Goldstein, I.J. (eds.). (1986)The Lectins. Properties, Function and Applications inBiology and Medicine, Academic Press, New York.

2 Olden, K. & Parent, J.B. (eds.). (1987) Vertebrate Lectins,Van Nostrand Reinhold Co., New York.

3 Cabins, H.-J. & Gabius, S. (eds). (1991) Lectins andCancer, Springer-Verlag, Heidelberg.

4 Lis, H. & Sharon, N. (1986) Annu. Rev. Biochem. 55,35-67.

5 Gabius, H.-J. (1991) Biochim. Biophys. Acta 1071, 1-18.6 Barondes, S.H. (1984) Science 223, 1259-1264.7 Sharon, N. & Lis, H. (1987) Trends Biochem. Sei. 12,488-

491.8 Calvo, P.O. & Ryan, R.J. (1985) Biochemistry 24, 1953-

1959.

9 Nagai, Y. & Tsuji, S. (1988) in Lectins and Glycoconjugatesin Oncology (Gabius, H.-J. & Nagel, G. ., eds.) pp. 59-66. Springer-Verlag, Heidelberg.

10 Sairam, M.R. (1989) FASEBJ. 3, 1915-1926.11 San Jose, E., Benguria, A. & Villalobo, A. (1990) /. Biol.

Chem. 265, 20653-20661.12 Brown, A.E., Lok, M.P. & Elovson, J. (1976) Biochim.

Biophys. Acta 426, 418-432.13 Church, J.G., Ghosh, S., Roufogalis, B.D. & Villalobo,

A. (1988) Biochem. Cell Biol. 66, 1-12.14 Gabius, H.-J. & Bardosi, A. (1991) Prog. Histochem.

Cvtochem. 22, 1-63.15 Gabius. H.-J. (1990) Anal. Biochem. 189, 91-94.16 Powell, J.T. & Whitney, PL. (1984) Biochem. J. 223, 769-

776.17 Redinbaugh, M.G. & Campbell, W.H. (1985) Anal.

Biochem. 147,144-147.18 Gabius, H.-J., Wosgien, B., Hendrys, M. & Bardosi, A.

(1991) Histochemistry 95, 269-277.19 Laemmli, U.K. (1970) Nature (London) 227, 680-685.20 Blum, H., Beier, H. & Gross, H.J. (1987) Electrophoresis

8, 93-99.21 Lowry, O.H., Rosebrough, N.J., Farr, A.L. & Randall,

R.J. (1951) J. Biol. Chem. 193, 265-275.22 Shapiro, B.M., Kingdon, H.S. & Stadtman, E.R. (1967)

Proc. Nad. Acad. Sei. USA, 58, 642-649.23 Shapiro, B.M. & Stadtman, E.R. (1968) J. Biol. Chem.

243, 3769-3771.24 Kustu, S., Hirschman, J., Burton, D., Jelesko, J. &

Meeks, J.C. (1984) Mol. Gen. Genet. 197, 309-317.25 Kustu, S., Hirschman, J. & Meeks, J.C. (1985) Curr. Top.

Cell. Regul. 27, 201-213.26 Niles, E.G. & Westhead, E.W. (1973) Biochemistry 12,

1723-1729.27 Chelala, C. A., Hirschbein, L. & Torres, H.N. (1971) Proc.

NatLAcad. Sei. USA, 68, 152-154.28 Caron, M., Bladier, D. & Joubert, R. (1990) Int. J.

Biochem. 22,1379-1385.29 Harrison, F.L. (1991) J. Cell Sei. 100, 9-14.30 Barondes, S.H. (1988) Trends Biochem. Sei. 13,480-482.31 Wells, V. & Malluci, L. (1991) Cell 64,91-97.32 Yamaoka, K., Ohno, S., Kawasaki, H. & Suzuki, K.

(1991) Biochem. Biophys. Res. Commun. 179,272-279.33 Lee, R.T., Gabius, H.-J. & Lee, Y.C. (1992) J. Biol.

Chem. 267,23722-23727.34 Bryant, R.W., Granzow, C.A., Siegel, M.I., Egan, R.W.

& Billah, M.M. (1991) /. Immunol. 147, 1856-1862.

E. San Jose, E. Villalobo and A. Villalobo*, Instituto de Investigaciones Biomedicas, CSIC,Arturo Duperier 4, E-28029 Madrid, Spain:H.-J. Gabius, Institut für Pharmazeutische Chemie der Universität, Abteilung Glykobiochemie und AngewandteTumorlektinologie,MarbacherWeg 6, W-3550 Marburg, Germany.

* To whom correspondence should be sent.



inhibition of the adenylylation of liver plasma membrane-bound...

Documents