ORIGINAL ARTICLE

Magdalena I. Tchorbadjieva Æ Ralitsa I. Kalmukova

Ivelin Y. Pantchev Æ Stanimir D. Kyurkchiev

Monoclonal antibody against a cell wall marker protein for embryogenicpotential of Dactylis glomerata L. suspension cultures

Received: 21 February 2005 / Accepted: 9 May 2005 / Published online: 15 July 2005� Springer-Verlag 2005

Abstract We identified and isolated a monoclonal anti-body (MAb 3G2) raised against extracellular proteinsfrom microcluster cells of orchard grass (Dactylisglomerata L.) embryogenic suspension culture. MAb3G2 recognized with high specificity an antigen ionicallybound within the primary cell wall and in the culturemedium of microcluster cells. Two-dimensional poly-acrylamide gel analysis and blotting of proteins onPVDF membrane showed that MAb 3G2 detected asingle polypeptide of apparent molecular mass of48 kDa and an isoelectric point (pI) of 5.2, designatedEP48. A transient expression during somatic embryo-genesis was observed for EP48. Indirect immunofluo-rescence showed that this protein highly accumulated inthe cell walls of some single cells, microclusters andpartly in proembryogenic masses (PEMs), but not inglobular embryos of the embryogenic cell line and mi-croclusters from the non-embryogenic cell line. Signalintensity varied between individual cells of the samepopulation and in successive stages of somatic embryodevelopment. Screening of several D. glomerata L.embryogenic and non-embryogenic cell lines with MAb3G2 indicated the presence of ECP48 in only embryo-genic suspension cultures at early stages of embryodevelopment long before morphological changes havetaken place and thus it could serve as an early markerfor embryogenic potential in D. glomerata L. suspensioncultures.

Keywords Cell wall Æ Dactylis Æ Extracellular proteins ÆMonoclonal antibody (3G2) Æ Protein markers ÆSomatic embryogenesis

Abbreviations Dicamba: 3,6-Dichloro-o-anisicacid Æ ELISA: Enzyme-linked immunosorbentassay Æ Em: Microclusters from embryogenic cellline Æ IEF: Isoelectric focusing Æ MAb: Monoclonalantibody Æ NEm: Microclusters from non-embryogeniccell line Æ PEMs: Proembryogenic masses Æ pI: Isoelectricpoint Æ SH-0: Schenk and Hildebrandt (1972)medium Æ SH-30: SH medium, supplemented with30 lM dicamba

Introduction

The capacity for somatic embryogenesis is a remarkableproperty of plant cells. Somatic embryogenesis is theprocess by which somatic cells develop into plantsthrough characteristic morphological stages thus ren-dering it a good model to study early plant development.The molecular basis of this unique developmentalpathway, particularly the transition of somatic cells intoembryogenic ones is still the least understood (forreview, Feher et al. 2003). Somatic embryogenesis in cellsuspension cultures provides an alternative way to ad-dress this problem. Conditioned medium harbors acomplex array of molecules, mainly derived from cellwalls, which are associated with developmental changes(Mordhorst et al. 1997). Suspension cultures secrete intothe medium glycoproteins that play an important role insomatic embryogenesis by their ability to stimulate (DeVries et al. 1988; Kreuger and Van Host 1993; Eg-ertsdotter and Von Arnold 1995; Domon et al. 2000) orinhibit (Gavish et al. 1992; Maes et al. 1997) somaticembryo development. The role of these proteins has notbeen elucidated fully, but for some of them like lipidtransfer protein EP2, acidic endochitinase EP3 and

M. I. Tchorbadjieva (&) Æ R. I. Kalmukova Æ I. Y. PantchevFaculty of Biology, Department of Biochemistry,Sofia University, 8 Dragan Zankov str,1164 Sofia, BulgariaE-mail: magd@biofac.uni-sofia.bg

S. D. KyurkchievInstitute of Biology and Immunology of Reproduction,Bulgarian Academy of Sciences,73, Tsarigradsko shaussee, 1113 Sofia, Bulgaria

R. I. KalmukovaMIT, Center for Cancer Research,40 Ames St, E17-110, Cambridge, MA, 02139 USA

Planta (2005) 222: 811–819DOI 10.1007/s00425-005-0027-9

arabinogalactan proteins (AGPs) it has been shown thatthey play a key role in plant somatic embryogenesis.Expression of lipid transfer protein EP2 is a well-knownearly marker of somatic embryo induction (Sterk et al.1991) and embryo differentiation as it is linked to theformation of the protoderm layer in developing somaticembryos (Thoma et al. 1994; Toonen et al. 1997). Anextracellular 32 kDa acidic endochitinase was identifiedable to rescue somatic embryogenesis in the mutantcarrot cell line ts11 (De Jong et al. 1992). Van Hengel etal. (2001) hypothesized that chitinase-modified AGPsare extracellular molecules capable of controlling ormaintaining the embryogenic competent cell state.

Several antibodies were prepared against diverseAGPs and were used to mark specific cell types (forreviews, Knox 1997; Willats et al. 2000). An AGP epi-tope recognized by the JIM8 antibody was originallydescribed as a marker of the very early transitional stageof cultured carrot cells after embryogenic induction(Pennell et al. 1992). Subsequently, it was shown thatmost embryos develop from cells lacking the JIM8 epi-tope (Toonen et al. 1996). Finally, it was found that theJIM8 epitope marks a specific cell type, that upon celldivision asymmetrically transferred the JIM8 epitope toa JIM8� embryogenic and JIM8+ apoptotic cell type. Itwas further demonstrated that the JIM8 epitope repre-sents a soluble signal produced by JIM8+ cells tostimulate embryo development of JIM8� cells (McCabeet al. 1997).

Most of the important crops and grasses are re-calcitrant for in vitro culturing which hampers thedevelopment of reliable regeneration techniques. Thebetter understanding of fundamental processes, whichtrigger and control somatic embryogenesis, will lead toregeneration protocols that are more rational. Thecharacterization and functional analysis of proteinmarkers for somatic embryogenesis offers a possibility todetermine the embryogenic potential of plant cells inculture long before any morphological changes havetaken place as well as to gain further information on themolecular basis of induction and differentiation of plantcells.

The investigations on the presence, role and use ofsecreted proteins as markers of embryogenic potential inmonocots were scarce (Nielsen and Hansen 1992; Stirnet al. 1995; Tchorbadjieva and Odjakova 2001). Dactylisglomerata L. is the first cereal grass to complete embryodevelopment and germination directly in a single liquidmedium (Conger et al. 1989). Thus, its suspension cul-tures present a useful model system to study the wholeprocess of somatic embryogenesis from single cells toplants. Previously, we have observed that distinct mor-phological structures from D. glomerata L. embryogenicsuspension cultures secrete proteins in a stage-specificmanner (Tchorbadjieva et al. 2004). In the present pa-per, we describe the production of a monoclonal anti-body against a cell wall protein secreted by the earliestmorphological structures (microclusters) in somaticembryogenesis in order to use it to monitor the expres-

sion of embryogenic potential in D. glomerata L. sus-pension cultures.

Materials and methods

Plant material and suspension cultures

Callus-derived suspension cultures from three embryo-genic (E1, E2, E3) and three non-embryogenic (NE1,NE2, NE3) cell lines of orchard grass (Dactylis glomerataL.) were initiated according to Conger et al. (1989) andmaintained in a liquid SH-30 medium essentially aspreviously described (Tchorbadjieva and Odjakova2001). The non-embryogenic suspension cultures NE1,NE2 and NE3 were initiated from segregated non-embryogenic sectors of the calli used for induction of E1,E2 and E3 lines, respectively. Viability of cultured plantcells was determined by the Evan’s Blue exclusionmethod of Gaff and Okong’o-Ogola (1971).

Fractionation of suspension cultures

Fractions of globular embryos, PEMs, microclustersand single cells from the embryogenic suspension cul-tures were collected by passing the cultures consecutivelythrough a series of 230-, 104-, 60- and 30 lm sieves,respectively. Each of the fractions was rinsed with SH-0medium and subcultured in a fresh SH-30 medium at adensity of 1 ll packed cell volume per ml medium. After7 days in culture, the culture media were separated fromthe cells and were used as a source of extracellularproteins. The single cells and microclusters from thenon-embryogenic cultures, retained on the 30- and 60-lm sieves, respectively, were maintained in the samemanner.

Protein preparation

Suspension-cultured cells at day 7, after transfer, werecentrifuged at 500 g for 5 min. The culture medium wasrecovered by passing the supernatant through a Milli-pore 0.22 lm filter. Extracellular proteins in the mediumwere precipitated by the addition of 2.5 volumes ofethanol. After overnight at 4�C, the precipitate wascollected by centrifugation (12,000 g at 4�C for 30 min),vacuum-dried and stored at �70�C or dissolved in waterfor immediate use (De Vries et al. 1988). The cell wallproteins were isolated from sedimented living cellsaccording to O’Neill and Scott (1987). After washingtwice with fresh SH-0 medium, the cells were incubatedin 0.1 mM CaCl2 on ice for 25 min, centrifuged and thesupernatant was passed through a Millipore 0.22 lmfilter. The proteins were precipitated as described forextracellular proteins. Intracellular soluble proteins wereobtained by grinding the cells in extraction buffer con-taining 50 mM Tris–HCl, pH 7.5, 1 mM EDTA, 5 mM

812

MgCl2 and 1 mM PMSF. The homogenate was filteredthrough nylon gauze, centrifuged at 15,000 g at 4�C for30 min and the supernatant was recovered. The proteincontent was determined according to Bradford (1976).The LMW-SDS marker kit (Amersham Biosciences)consisted of phosphorylase b (97 kDa), bovine serumalbumin (66 kDa), ovalbumin (45 kDa), carbonic an-hydrase (30 kDa), trypsin inhibitor (20.1 kDa) and alactalbumin (14.4 kDa).

Immunization procedures and hybridoma preparation

A set of four 60-day-old BALB/c mice was immunizedwith extracellular proteins from microclusters (Em) ofthe E1 embryogenic cell line. 100 lg in 100 ll wereemulsified with the same volume of complete Freund’sadjuvant for the first injection and with incompleteFreund’s adjuvant for subsequent challenges. Intrave-nous boost of 100 lg protein in saline was given 3 daysprior to spleenectomy. Sera were collected after eachboost and screened for reaction with extracellular pro-teins from Em by enzyme-linked immunosorbent assay(ELISA). Spleenocytes (108) cells were fused with5·107 P3U-1 myeloma cells. HAT selection and allsubsequent steps were carried out according to standardprotocols (Liddell and Cryer 1991). Selected cell popu-lations were cloned at least three times by the method oflimiting dilution to obtain monoclonality. A mouse -MAb-isotyping kit (Sigma) was used to determine thesubclass of the antibodies.

Screening by ELISA

The screening procedure consisted of two subsequentELISA tests and a dot-blot binding assay. In the firstELISA, the ELISA microtitre plates (NUNC maxisorb)were coated with 100 ll per well of 200 lg ml�1 extra-cellular proteins from Em, resuspended in buffer C(50 mM sodium carbonate buffer, pH 9.6) overnight at4�C. Plates were blocked for 1 h with 200 ll PBS, con-taining 1% bovine serum albumin (BSA) at room tem-perature. Plates were washed with tap water and 100 llper well hybridoma supernatants were added and incu-bated at room temperature for 2 h. Plates were washedagain, goat anti-mouse IgG-alkaline phosphatase (Bio-Rad) was applied at a 1/2000 dilution in PBS containing1% (w/v) BSA at 100 ll per well and the plates wereincubated at room temperature for 2 h. Followingexhaustive washing in tap water, bound alkaline phos-phatase activity was measured using p-nitrophenylphosphate as substrate (50 ll of a 1 mg ml�1 solution in100 mM sodium bicarbonate pH 9.5, containing 1 mMMgCl2). The alkaline phosphatase was inactivated after30 min by addition of 3 N NaOH (50 ll). Absorbanceswere read on Dynatech MR 700 ELISA reader at405 nm. In the second ELISA, antibodies reactingpreferentially with extracellular proteins from Em were

selected. The hybridoma supernatants were testedsimultaneously for reactions with extracellular proteinsfrom the Em and microclusters (NEm) of the non-embryogenic NE1 cell line by ELISA as described above.As a third screening step dot-blot assays were per-formed. Various plant antigens (1 lg) in 2 ll aliquotswere applied to nitrocellulose. Dot-blots were developedessentially as described by Smallwood et al. (1994) ex-cept that alkaline phosphatase - conjugated secondaryantibody was used.

Immunodot-binding assay

The nature of the antigenic epitope recognized by MAb3G2 was determined in an immunodot-binding assay.Extracellular proteins from the Em at a concentration of2 mg ml�1 were resuspended in an equal volume of125 mM Tris-HCl, pH 6.8, 4 % SDS with or without10% b-mercaptoethanol, boiled for 3 min in a waterbath and aliquots (2 ll) were applied on nitrocellulosemembrane. The membrane was air-dried at room tem-perature for 30 min. After blocking for 1 h in TBST-0.5(10 mM Tris–HCl, pH 7.4, 150 mM NaCl, 0.5% Tween20), the membrane was incubated for 2 h in MAb 3G2(culture supernatant at a 1/10 dilution in TBST-0.05).After an extensive wash in TBST-0.05 (10 mM Tris–HCl, pH 7.4, 150 mM NaCl, 0.05% Tween 20) themembrane was incubated for 2 h with a 1:2000 dilutionof alkaline phosphatase-conjugated goat anti-mouseantibody (Bio-Rad), washed with TBST-0.05 anddeveloped using 0.03% nitroblue tetrazolium and0.015% 5-bromo-chloro-3-indolyl phosphate in AP-buffer (100 mM Tris–HCl, pH 9.5, 100 mM NaCl,1 mM MgCl2).

Analysis by electrophoresis and Western blotting

For gel electrophoresis, proteins were incubated for15 min at 37�C in 62.5 mM Tris–HCl, pH 6.8, 2% (w/v) SDS and 5% (w/v) b-mercaptoethanol and separatedby SDS-PAGE according to Okadjima et al. (1993)using a 12 % acrylamide separating gel and a 4%acrylamide stacking gel. Gels were run at 200 V con-stant voltage in a Mini-Protean II electrophoresis cell(Bio-Rad). Isoelectric focusing (IEF) was carried outon 0.75 mm thick acrylamide slab gels (5%), using 2%(v/v) Ampholine carrier ampholytes, pH 3.5 to 9.5(Pharmacia Biotech). IEF gel electrophoresis (Phar-macia Multiphor II) was performed according to themanufacturer’s procedure. Isoelectric points (pI) weredetermined by using a set of broad range (pI 3.5-9.5)IEF standards (Pharmacia Biotech). For the seconddimension, lanes from IEF were excised, equilibrated inequilibration buffer (62.5 mM Tris–HCl, pH 6.8,50 mM DTT, 2% (w/v) SDS and 0.01% (w/v)Bromphenol Blue) for 20 min and then were placedon 10 cm·8 cm·0.075 cm slab gels for separation

813

according to Okadjima et al. (1993). Proteins werestained with silver nitrate (Blum et al. 1987). UponSDS-PAGE or 2-D PAGE separation, the proteinswere transferred to PVDF membranes for 1 h using asemi-dry apparatus (Hoefer Semi-Phor) in the pH 9.9carbonate buffer system of Dunn (1986) after firstincubating the gel in 50 mM Tris–HCl, pH 7.4 for30 min. The membrane was further incubated for 16 hat 37�C in PBS, containing 0.02% NaN3, followed byblocking for 1 h in TBST-0.5. The membrane wasincubated with MAb 3G2 (culture supernatant at a 1/10 dilution in TBST-0.05, 1 M D-glucose and 10 %glycerol) for 3 h at room temperature (Birk andKoepsell 1987). After an extensive wash in TBST-0.05,the antibody binding was detected with the alkaline-phosphatase-linked second antibody as described abovefor the dot-blot assay.

Indirect immunofluorescence microscopy

Immunofluorescent immersion labeling of intact sus-pension-cultured cells was carried out as described byWillats et al. (1999b). Briefly, cells were fixed in 4%paraformaldehyde in 50 mM PIPES, 5 mM MgSO4 and5 mM EGTA, pH 6.9 overnight at 4�C. Cells were wa-shed several times in SH-0 medium and, after a 30 minblocking in TBST-0.5, cells were incubated in primaryantibody (hybridoma supernatant 3G2, diluted 1/10 inTBST-0.5) for 2 h, then washed thrice with TBST-0.05prior to incubation with fluorescein isothyocyanate(FITC)-labeled goat anti-mouse antibody (1:20 dilu-tion)(Sigma) in TBST for 2 h. Cells were washed thrice,mounted in PBS/glycerol-based anti-fade solution (Cit-ifluor) and examined with microscope equipped withepifluorescence illumination. Images were recorded on400 ASA color slide film (Kodak).

Results

Selection of hybridomas and characterization of MAbs

The monoclonal antibody MAb 3G2 was derived sub-sequent to immunization with a preparation of extra-cellular proteins from the conditioned medium oforchard grass embryogenic suspension-cultured micro-cluster cells. Approximately 600 hybridoma clones werescreened on ELISA plates coated with extracellularproteins from microclusters (Em) of the E1 embryogeniccell line or microclusters (NEm) from the NE1 non-embryogenic cell line to identify antigens preferentiallyrepresented in the medium of the embryogenic micro-cluster cells. Thirty-four hybridomas that producedantibodies highly reactive with Em antigens were furthertested under the same conditions and one stablehybridoma (3G2) was finally selected, cloned and usedfor further experiments. With the use of an isotyping kit,the MAb was found to belong to the IgG1 class.

Detection of the antigens in culture medium

A dot-blot assay of the antigens denatured with SDSonly or with SDS and b-mercaptoethanol showed thatthe immunoreaction with MAb 3G2 was stronglyreduced after SDS treatment and was fully abolishedafter the combined SDS and b-mercaptoethanol treat-ment (not shown), thus suggesting that MAb 3G2 rec-ognizes a conformational epitope, which is commonwith MAbs raised against native antigens. To maximizethe chance of presenting native conformation of theantigens to the MAb during Western blotting, weadopted a combination of the method of Birk andKoepsel (1987) and this of Dunn (1986) (see Materialsand methods).

In order to characterize the antigens with which MAb3G2 reacted, Western blot analysis of extracellularproteins from the conditioned medium of suspension-cultured Em and NEm was performed (Fig. 1b). MAb3G2 reacted strongly with a single 48 kDa polypeptidefrom the Em extracellular proteins (lane 1 in Fig. 1b).There was no signal when MAb 3G2 was probed withthe extracellular proteins from NEm (lane 2 in Fig. 1b).This, together with the negative result obtained with thepreimmune serum used as a control (Fig. 1c) demon-strates the high specificity of the monoclonal antibody.Screening of the extracellular proteins from microclus-ters of the other two embryogenic (E2 and E3) and non-embryogenic (NE2 and NE3) suspension cultures onimmunoblots showed that EP48 was found exclusively inthe embryogenic cell lines (data not shown).

Characterization of the antigen detected byMAb 3G2 was extended by showing that upon 2-D gel

Fig. 1 a-c Immunoblot analysis of MAb 3G2 reactivity withextracellular proteins from D. glomerata L. suspension cultures.Extracellular proteins from embryogenic (lane 1) and non-embryo-genic (lane 2) microcluster cells were separated by SDS-PAGE andeither silver-stained (a) or probed with MAb 3G2 after transfer toPVDF membrane (b). MAb 3G2 recognized a single protein(Mr 48,000) (arrow). Loading was at 8 lg per lane. The control withpreimmune serum (c) was negative. Molecular mass markers areindicated on the left in kilodaltons

814

electrophoresis and immunoblotting, EP48 migrated asa single spot with a pI 5.2 (Fig. 2a). Matching thecoordinates of the antibody-reactive spot to the sameposition on a silver-stained gel showed that the EP48was abundantly represented (Fig. 2b).

Periodate treatment of extracellular proteins immo-bilized on nitrocellulose had no effect on the binding ofMAb 3G2, while the Proteinase K treatment abolishedthe binding which suggests that the EP48 epitope maycontain a protein component (data not shown).

Expression of EP48 during somatic embryogenesis

The monoclonal antibody 3G2 was used to monitor theexpression of EP48 during various stages of somaticembryogenesis. For the purpose, extracellular proteinssecreted from different morphological structures of theE1 embryogenic cell line were separated by SDS-PAGEand immunoblotted with MAb 3G2 (Fig. 3). A transientexpression of EP48 during somatic embryogenesis wasobserved. The strength of the signal varied considerably.There was only a faint signal at the stage of transition of

single cells into microclusters (lane 1 in Fig. 3b). A verystrong signal was detected at the next stage when PEMsformed from microclusters (lane 2 in Fig. 3b) and then,upon embryo formation from PEMs and embryo mat-uration it decreased continuously (lanes 3 and 4 inFig. 3b)

Cellular localization of EP48

To test whether EP48 is a cell wall protein that becomesleached from cell walls by culture medium, we extractedliving suspension cells of various morphological struc-tures with a calcium chloride solution. This method al-lows for the isolation of ionically bound cell wallproteins from intact cells without contaminating cyto-plasmic proteins (O’Neill and Scott 1987). Immuno-blotting showed that EP48 was indeed present in the cellwall of embryogenic microclusters (lane 2 in Fig. 4b). Itwas not found in the intracellular soluble protein frac-tion (lane 3 in Fig. 4b). Intensive washing of the cellsprior to CaCl2 extraction resulted in a loss of EP48 from

Fig. 2 a, b Immunodetection ofEP48 in the medium of D.glomerata L. E1 embryogenicsuspension-culturedmicroclusters after two-dimensional polyacrylamidegel electrophoresis. Theextracellular proteins wereprobed with MAb 3G2 aftertransfer to PVDF membrane(a). A duplicate gel was silver-stained (b). The position ofEP48 (48/5.2) is indicated byarrow

Fig. 3 a, b Immunochemical detection of EP48 during somaticembryogenesis of D. glomerata L. E1 embryogenic suspensionculture. Extracellular proteins from the medium of single cells (lane1), microclusters (lane 2), PEMs (lane 3) and globular embryos(lane 4) were separated by SDS-PAGE and silver-stained (a) orprobed with MAb 3G2 after transfer to PVDF membrane (b).Equal amounts of protein were applied to each lane (10 lg/ lane).Protein mass markers are shown on the left in kilodaltons

Fig. 4 a, b Cellular localization of EP48 in D. glomerata L.embryogenic microclusters. Extracellular proteins from culturemedium (lane 1), ionically-bound cell wall proteins (lane 2) andsoluble intracellular proteins (lane 3) were separated on SDS-PAGE and stained with silver (a) or probed with MAb 3G2 afterblotting on PVDF membrane (b). Equal amounts of protein wereapplied to each lane (12 lg/ lane). Molecular mass markers areindicated to the left

815

the cell wall, which indicates that the protein is weaklybound to the cell wall. This might explain its abundancein the culture medium (lane 1 in Fig 4b). Immunoblotswith cell wall proteins isolated by the same method fromPEMs showed identical results, while a very faint im-munosignal in the cell wall of globular embryos wasobserved (data not shown).

Immunocytochemical localization of EP48in embryogenic suspension culture

The distribution of the 3G2 epitope in distinct mor-phological structures during somatic embryogenesis wasexamined by indirect immersion immunofluorescence(Fig. 5). D. glomerata L. embryogenic suspension cul-tures contain single cells with different morphologiessuch as small isodiametric, elongating and elongatedcells (Conger et al. 1989). Some of the suspension cellsshowed intense fluorescence after labeling with MAb3G2 (Fig. 5a). The majority of the signal-positive cellswere small, isodiametric, cytoplasm-rich cells, thoughsome elongated, banana-shaped cells were labeled, too.There were, however, many cells that remained unla-beled (Fig. 5a) as in control samples (Fig. 5e) which arein agreement with the very low level of EP48 secretionby single cells (lane 1 in Fig. 3b). As development of thesingle cells proceeded (7 days in SH-30 medium), somenewly formed microclusters showed distinct fluorescenceafter MAb 3G2 binding (Fig. 5b). All the cells in thesemicroclusters were strongly fluorescent as they divided.Immunofluorescence was located at sites of cell–cellcontact but could also be found on cell surface regionsthat were not in direct contact with neighboring cells. Inaddition, many clusters were present that were com-pletely devoid of EP48-expressing cells (not shown).PEMs, which formed from microclusters in the course ofsomatic embryogenesis, had 3G2- positive material inthe majority of their cell walls (Fig. 5c). The distributionof EP48 was uneven though, it was less intense or evenabsent from the regions of the surface of PEMs wherecells had no neighbors. Only a few newly formed glob-ular embryos reacted weakly with the antibody (notshown). MAb 3G2 did not stain single cells and mi-croclusters from the NE1 non-embryogenic suspensionculture (Fig. 5d), which is in agreement with the totalabsence of EP48 from the medium and the cell walls ofthese cells (lane 2 in Fig. 1b).

Discussion

By screening of a large panel of hybridoma clones, wehave identified a highly specific monoclonal antibodyMAb 3G2 which reacts with a protein of molecular mass48 kDa and pI 5.2 (designated EP48), found in the cellwall and the culture medium of microcluster cells fromD. glomerata L. embryogenic suspension cultures. It iswell known that the initial stage of transition of somatic

cells into embryogenic ones is crucial for somaticembryogenesis. It is the time when drastic changes in cellmorphology and function occur (Mordhorst et al. 1997;Feher et al. 2003). Marker proteins for those very earlystages of development are of particular interest andwould help to better define the key mechanisms of so-matic embryo development. The appearance of EP48 ata very early stage of development of only embryogenicsuspension cultures and its absence from the mediumand the cell wall of a similar morphological stage of thenon-embryogenic cell lines derived from the samegenotype and cultured under the same conditions pointsto the embryo-specific nature of this protein. This opens

Fig. 5 a-e Indirect immunofluorescent localization of EP48 onintact D. glomerata L. suspension cells during somatic embryogen-esis. MAb 3G2 labeled the cell wall of small, isodiametric singlecells as well as elongated, banana-shaped single cells whereas othersingle cells (shown by arrows) remained unstained (a). The singlecells and microclusters from the non-embryogenic suspensionculture were nonreactive, too (d). The fluorescence due to theantibody binding is most intense at the regions of cell adhesion ofmicrocluster cells (b) and PEMs (c) (large arrowheads) while regionsof cell wall without neighbors are unlabeled in PEMs (smallarrowheads). Control cells (e) were labeled with secondary antibodyonly. f-h Microclusters (f) and PEMs (g) from E1 embryogenicsuspension culture and single cells and microclusters from NE1

non-embryogenic suspension culture (h) are visualized under visiblelight. Bars=30 lm (a–c, e–g), 60 lm (d, h)

816

up the possibility to use it as a marker for embryogenicpotential of the suspension cultures. Interestingly, EP48was found in the embryogenic cell line E2 too, in whichmature embryos fail to regenerate. The presence of EP48in this line suggests that this protein is related toembryogenic competence (ability to form embryos) atearly stages of somatic embryogenesis and not toregeneration potential. This is in support of Stirn et al.(1995) who proposed that the formation of somaticembryos and the subsequent regeneration are twoindependent processes under a different control. Thepresence of EP48 in rapidly dividing cells may suggestthat the EP48 might be correlated to cell proliferation.However, its decreased quantity and almost total ab-sence from the medium of actively dividing PEMs andembryos, respectively, as well as its absence from pro-liferating cultures of the non-embryogenic cell lines aregood arguments to assume another role of this proteinthan its involvement in cell proliferation.

The molecular mass of EP48 falls in the range of 40–55 kDa proteins from other plant species related toembryo development. Several marker proteins forembryogenic potential were identified in suspensioncultures of D. carota (Kiyosue et al. 1993); Pisum sati-vum L. (Altherr et al. 1993); Hordeum vulgare (Stirn etal. 1995) and Triticum aestivum (Fellers et al. 1997). It ishighly possible that the synthesis of 40–55 kDa poly-peptides represents a conserved set of events that areessential for early plant embryo development.

The growth medium of plant cell cultures may beregarded as a large extension of the intercellular space;soluble secreted molecules that inhabit the apoplast inplanta will accumulate in the medium when cells aregrown in suspension. Immunochemical evidence indi-cates that orchard grass cell culture medium contains asubstantial quantity of EP48, which could be explainedby its cell wall localization. Our results show that EP48is present in the cell wall as an ionically bound proteinwhose weak binding could explain its abundance in themedium. The lack of correlation between the strength ofthe immunosignal in the medium and the cell walls of thesingle cells is due to the small number of single cells thatare labeled by MAb 3G2 in the first stage of somaticembryogenesis.

In both cell walls and medium, we observed a tran-sient pattern of EP48 expression during early embryodevelopment up to the globular stage. EP48 appearedfirst at the surface of a small subpopulation of singlecells in culture independently of their morphology. Somemicrocluster cells and PEMs expressed EP48 to a sig-nificant level then, during embryo formation, itsexpression was completely abolished. The absence ofEP48 from the cell wall of leaf explants used to initiatethe suspension cultures (not shown) indicates that itssynthesis is developmentally regulated. The pattern ofMAb 3G2 binding differed in microcluster cells andPEMs. The antibody labeled the entire surface of mi-crocluster cells as well as the region of cell-to-celladhesion while in PEMs regions of cell wall without

neighbors remained unstained. This differential labelingis difficult to explain at present. It is well known that cellwalls differ among different cell types (Willats et al.1999b) and even among the wall surrounding a singlecell (Samaj et al. 1999). It is possible that during devel-opment of PEMs, there occurs a local change in the cellwall of some cells leading to the loss of MAb 3G2 epi-tope. Interestingly, MAb 3G2 binding in PEMs is simi-lar to that of a phage display monoclonal antibodyPAM1 with specificity for de-esterified blocks of pectichomogalacturonan (HG) (Willats et al. 1999a). HGblocks cross-linked with calcium play an important rolein cell adhesion. PAM1 labeling was restricted to regionsof direct cell-to-cell contact at the surface of cell clumps.

Cell tracking (Toonen et al. 1996) or immuno-mag-netic sorting (McCabe et al. 1997) has been applied todetermine the developmental fate of carrot single cellslabeled with JIM8. Cell tracking also showed that SERK(somatic embryo receptor-like kinase) in single cellscould be indeed correlated with subsequent formation ofsomatic embryos in carrot (Schmidt et al. 1997), D.glomerata L. (Somleva et al. 2000) and Arabidopsis(Hecht et al. 2001). Whether the single cells, microclus-ters and PEMs in D. glomerata L. embryogenic sus-pension cultures labeled with MAb 3G2 are competentfor embryogenesis remains to be elucidated but certainlyit could be used as a marker for embryogenic potentialof the entire suspension culture.

The monoclonal antibodies prepared against differentcomponents of the plant cell wall and extracellular pro-teins from the culturemediumare usefulmolecular probesto study the complex organization and dynamics ofinteraction between single components of the cell wall as apart of the plant extracellular matrix (Knox 1997, 1999;Smallwood et al. 1995, 1996). It is assumed that theextracellular proteins are indispensable for differentiationand morphogenesis taking part in signal transduction,cell–cell recognition, cell expansion and adhesion.

In the present study, with the help of a new mono-clonal antibody, we have identified based on its locali-zation and pattern of accumulation a cell wall proteinEP48 as a marker for embryogenic potential of D.glomerata L. suspension cultures. Research is currentlyunderway to determine the structure of EP48 and toelucidate its possible role for somatic embryo develop-ment.

Acknowledgements This work was supported by grant 310496 fromthe Research Fund of Sofia University’’St. Kl. Ohridski’’.

References

Altherr S, Stirn S, Jacobsen H-J (1993) Immunobiochemicalanalysis of a nuclear protein marker for regeneration potentialin higher plants. J Plant Physiol 141:415–422

Birk H-W, Koepsell H (1987) Reaction of monoclonal antibodieswith plasma membrane proteins after binding on nitrocellulose:renaturation of antigenic sites and reduction of non-specificantibody binding. Anal Biochem 164:12–22

817

Blum H, Beier B, Gross H (1987) Improved silver staining of plantproteins, RNA and DNA in polyacrylamide gels. Electropho-resis 8:93–99

Bradford M (1976) A rapid and sensitive method for the quanti-tation of microgram quantities of proteins utilizing the principleof protein-dye binding. Anal Biochem 72:248–254

Conger BV, Hovanesian JC, Trigiano R, Gray D (1989) Somaticembryo ontogeny in suspension cultures of orchard grass. CropSci 29:448–452

De Jong AJ, Cordewener J, LoSchiavo F, Terzi M, Van-dekerckhove J, Van Kammen A, De Vries SC (1992) A carrotsomatic embryo is rescued by chitinase. Plant Cell 4:425–433

De Vries SC, Booij H, Janssens R, Vogels R, Saris L, LoSchiavo F,Terzi M, Van Kammen A (1988) Carrot somatic embryogenesisdepends on the phytohormone-controlled presence of correctlyglycosylated extracellular proteins. Genes Dev 2:462–476

Domon J-M, Neutelings G, Roger D, David A, David H (2000) Abasic chitinase-like protein secreted by embryogenic tissues ofPinus caribaea acts on arabinogalactan proteins extracted fromthe same cell line. J Plant Physiol 156:33–39

Dunn S (1986) Effects of the modification of transfer buffer com-position and the renaturation of proteins in gels on the recog-nition of proteins on Western blots by monoclonal antibodies.Anal Biochem 157:144–153

Egertsdotter U, von Arnold S (1995) Importance of arabinoga-lactan proteins for the development of somatic embryos ofNorway spruce (Picea abies). Physiol Plant 93:334–345

Feher A, Pasternak T, Dudits D (2003) Transition of somatic plantcells to an embryogenic state. Plant Cell Tissue Organ Cult74:201–228

Fellers JP, Guenzi AC, Porter DR (1997) Marker proteins associ-ated with somatic embryogenesis of wheat callus cultures. JPlant Physiol 151:201–208

Gaff DF, Okong’o-Ogola O (1971) The use of non-permeatingpigments for testing the survival of cells. J Exp Bot 22:756–758

Gavish H, Vard A, Fluhr R (1992) Suppression of somaticembryogenesis in Citrus cell cultures by extracellular proteins.Planta 186:511–517

Hecht V, Vielle-Calzada J-P, Hartog MV, Schmidt EDL, BoutilierK, Grossniklaus U, De Vries SC (2001) The Arabidopsis So-matic Embryogenesis Receptor Kinase 1 gene is expressed indeveloping ovules and embryos and enhances embryogeniccompetence in culture. Plant Physiol 127:803–816

Kiyosue T, Yamaguchishinozaki K, Shinozaki K, Kamada H,Harada H (1993) cDNA cloning of ECP40, an embryogenic-cellprotein in carrot and its expression during somatic and zygoticembryogenesis. Plant Mol Biol 21:1053–1068

Knox JP (1997) The use of antibodies to study the architecture anddevelopmental regulation of plant cell walls. Int Rev Cytol171:79–120

Knox JP (1999) Intriguing, complex and everywhere: getting togrips with arabinogalactan-proteins. Trends Plant Sci 4:123–125

Kreuger M, Van Holst G-J (1993) Arabinogalactan proteins areessential in somatic embryogenesis of Daucus carota L. Planta189:243–248

Liddell J, Cryer A (1991) A practical guide to monoclonal anti-bodies. Wiley Chichester

Maes O, Coutos-Thevenot P, Jouenne T, Boulay M, Guern J(1997) Influence of extracellular proteins, proteases and prote-ase inhibitors on grapevine somatic embryogenesis. Plant CellTissue Organ Cult 50:97–105

McCabe PF, Valentine TA, Forsberg LS, Pennell RI (1997) Solublesignals from cells identified at the cell wall establish a devel-opmental pathway in carrot. Plant Cell 9:2225–2241

Mordhorst AP, Toonen MAJ, De Vries SC (1997) Plant embryo-genesis. Crit Rev Plant Sci 16:535–576

Nielsen KA, Hansen IB (1992) Appearance of extracellular pro-teins associated with somatic embryogenesis in suspensioncultures of barley (Hordeum vulgare L.). J Plant Physiol139:489–497

Okadjima T, Tanabe T, Yasuda T (1993) Nonurea sodium dodecylsulfate-polyacrylamide gel electrophoresis with high-molaritybuffers for the separation of protein and peptides. Anal Bio-chem 211:293–300

O’Neill R, Scott T (1987) Rapid effects of IAA on cell surfaceproteins from intact suspension culture cells. Plant Physiol84:443–446

Pennell RI, Janniche L, Scofield G, Booij H, De Vries SC, RobertsK (1992) Identification of a transitional cell state in the devel-opmental pathway to carrot somatic embryogenesis. J Cell Biol119:1371–1380

Schenk RU, Hildebrandt AC (1972) Medium and techniques forinduction and growth of monocotyledonous and dicotyledon-ous plant cell cultures. Can J Bot 50:199–204

Schmidt EDL, Guzzo F, Toonen MAJ, De Vries SC (1997) Aleucine-rich repeat containing receptor-like kinase marks so-matic plant cells competent to form embryos. Development124:2049–2062

Smallwood M, Beven A, Donovan N, Neill SJ, Peart J, Roberts K,Knox JP (1994) Localization of cell wall proteins in relation tothe developmental anatomy of the carrot root apex. Plant J5:237–246

Smallwood M, Martin H, Knox JP (1995) An epitope of ricethreonine-and hydroxyproline-rich glycoprotein is common tocell wall and hydrophobic plasma-membrane glycoproteins.Planta 196:510–522

Smallwood M, Yates EA, Willats WGT, Martin H, Knox JP (1996)Immunological comparison of membrane-associated and se-creted arabinogalactan proteins in rice and carrot. Planta198:452–459

Somleva MN, Schmidt EDL, De Vries SC (2000) Embryogeniccells in Dactylis glomerata L. (Poaceae) explant identified bycell tracking and by SERK expression. Plant Cell Rep 19:718–726

Sterk P, Booij H, Schellekens GA, Van Kammen A, De Vries SC(1991) Cell-specific expression of the carrot EP2 lipid transferprotein gene. Plant Cell 3:907–921

Stirn S, Mordhorst AP, Fuchs S, Lorz H (1995) Molecular andbiochemical markers for embryogenic potential and regenera-tive capacity of barley (Hordeum vulgare L.) cell cultures. PlantSci 106:195–206

Tchorbadjieva M, Odjakova M (2001) An acidic esterase as abiochemical marker for somatic embryogenesis in orchard grass(Dactylis glomerata L.) suspension cultures. Plant Cell Rep20:28–33

Tchorbadjieva M, Pantchev I, Harizanova N (2004) Two-dimen-sional protein pattern analysis of extracellular proteins secretedby embryogenic and non-embryogenic suspension cultures ofDactylis glomerata L. Biotechnol Biotechnol Eq 18:20–27

Thoma S, Hecht U, Kippers A, Botella J, DeVries SC, Somerville C(1994) Tissue-specific expression of a gene encoding a cell wall-localized lipid transfer protein from Arabidopsis . Plant Physiol105:35–45

Toonen MAJ, Schmidt EDL, Hendriks T, Verhoeven HA, VanKammen A, De Vries SC (1996) Expression of the JIM8 cellwall epitope in carrot somatic embryogenesis. Planta 200:167–173

Toonen MAJ, Verhees JA, Schmidt EDL, Van Kammen A, DeVries SC (1997) AtLTP1 luciferase expression during somaticembryogenesis. Plant J 12:1213–1221

Samaj J, Baluska F, Bobak M, Volkman D (1999) Extracellularmatrix surface network of embryogenic units of friable maizecallus contains AGPs recognized by mAb JIM4. Plant Cell Rep18:369–374

Van Hengel A, Tadesse Z, Immerzeel P, Schols H, Van KammenA, De Vries SC (2001) N-acetylglucosamine and glucosamine-containing arabinogalactan proteins control somatic embryo-genesis. Plant Physiol 125:1880–1890

Willats WGT, Gilmartin PM, Mikkelsen JD, Knox JP (1999a) Cellwall antibodies without immunization: generation and use ofde-esterified homogalacturonan block-specific antibodies froma naıve phage display library. Plant J 18:57–65

818

Willats WGT, Steele-King CG, Marcus SE, Knox JP (1999b) Sidechains of pectic polysaccharides are regulated in relation to cellproliferation and cell differentiation. Plant J 20:619–628

Willats WGT, Steele-King CG, McCartney L, Orfila C, MarcusSE, Knox JP (2000) Making and using antibody probes to studyplant cell walls. Plant Physiol Biochem 38:27–36

819