Planta

Planta (1988) 176: 472481

~ Springer-Verlag 1988

Monoclonal antibodies identify common and differentiation-specific antigens on the plasma membrane of guard cells of Vicia faba L. G6ran Key and Elmar W. Weiler Plant Physiology, University of Osnabriick, Postfach 4469, D-4500 Osnabrtick, Federal Republic of Germany

Abstract. Monoclonal antibodies were raised against the plasma membrane of Vicia faba L. guard cells by immunizing either with total membranes from purified guard-cell protoplasts or with sealed, predominantly right-side-out plasma-membrane vesicles prepared from abaxial epidermes of V.faba by aqueous two-phase partitioning. Hybridoma screening was performed by enzyme-linked immunosorbent assay using polystyrene-adsorbed plasma-membrane vesicles as solid phase and by indirect immunofluorescence analysis using unfixed, immobilized protoplasts in a microvolume Terasaki assay. A range of monoclonal antibodies was characterized and is reported here. One monoclonal antibody, G26-6-B2, is guard-cell-specific and does not react with mesophyll-cell protoplasts of the same species. It binds to a periodate-resistant but trypsin-labile epitope, probably a differentiation-specific plasma-membrane protein. Key words: Epitope (characterization) - Guard cell (protoplasts) - Immunofluorescence (plant protoplasts) - Monoclonal antibody (plasma membrane, guard cell) - Plasma .membrane (antigens) - Vicia (plasma membrane antigens)

Introduction

Monoclonal antibodies (MAB) are widely used to study cell membrane differentiation and cell typing in animal or human systems (e.g. Feizi 1985). They can also be generated against plasma-membrane antigens of higher plants, (e.g. Norman et al. 1986; Abbreviations: ELISA=enzyme-linked immunosorbent assay;

FITC=fluorescein isothiocyanate; G C P = g u a r d cell protoplast(s); Ig = immunoglobulin; MAB = monoclonal antibody; MCP = mesophyll-cell protoplast(s); SDS-PAGE = sodium dodecyl sulfate-polyacrylamide gel electrophoresis

Lyner et al. 1987), algae (Homan et al. 1987) and slime molds (Bozzaro and Merkl 1985). Most of these antibodies are directed against sugar determinants (Anderson et al. 1984; Bozzaro and Merkl 1985; Homan et al. 1987). In addition to characterizing specific membrane components, MABs against common epitopes of plant plasma membranes were recently used to identify somatic hybrids at the stage of protoplast fusion and early protoplast culture (Fitter et al. 1987). Very little information is available about differentiation-specific plasma-membrane components in higher plants. The identification and characterization of such antigens would be of considerable theoretical and practical value. We therefore decided to test whether or not the hybridoma technique would be suitable for identifying cell-type-specific antigens at the plasma membrane of higher plants. We selected mesophyll and guard cells because these can be prepared in Viciafaba L. in large amounts and with a high degree of purity. The selection of cell-surface-recognizing MABs is still problematic for higher-plant cells due to the fragility of the isolated protoplasts. Therefore, suitable, large-capacity screening techniques to identify MAB binding to the surface of intact, living protoplasts were devised and are reported here. The technique allows the evaluation of several different cell types simultaneously. Materials and methods Chemicals and reagents. Trypsin, trypsin inhibitor, 5-bromo-4chloro-3-indolyl phosphate and the following sugars were obtained from Sigma, Dreieich, F R G (Gal = galactose; Glc = glucose; Man =mannose; Me = methyl): N-acetyl-D-galactosamine; N-acetyl-o-glucosamine (GlcNAc); L-(+)-arabinose; ~o-(+ )-fucose; fl-D-Gal-(1-3)-D-GlcNAc; fl-D-Gal-(1-4)-DGlcNAc; fl-D-Gal-(1-6)-D-GIcNAc; fl-D-GlcNAc-(1-6)-~-oMan-l-OMe; D-(+)-mannose; ~-D-Man-(1-2)-~-D-Man- 1-

G. Key and E.W. Weiler: Monoclonal antibodies against guard-cell plasma membranes OMe; ~t-D-Man-(1-3)-:t-D-Man-l-OMe; ~X-D-Man-(I-6)-~-DMan-l-OMe; C~-D-(+)-melibiose; ~-methyl-D-mannoside; S-Lrhamnose. Methyl-c~-D-glucopyranoside was a gift from Professor Lengeler, Osnabriick, FRG and D-(+)-cellobiose, D-(+)galactose, D-glucosamine, L-sorbose and D-xylose were purchased from Serva, Heidelberg, FRG. Complete and incomplete Freund's adjuvant were purchased from Sigma. Terasaki and microtiter plates were obtained from Nunc, Roskilde, Denmark. Western-blots were done with the Protoblot kit, Atlanta, Heidelberg, FRG. All reagents used were of the highest grade available.

Plant material. Viciafaba L. cv. Osnabrficker Markt was grown in a greenhouse as described by Blum et al. (1988). Zea mays L. cv. Anjou 210, Pisum sativum L. cv. Kleine Rheinlfinderin and Nicotiana tabacum L. cv. Virginia were grown in the greenhouse under the same conditions. Arabidopsis thaliana L. was grown in short days (8 h photoperiod) in a growth chamber at 24~ (day) and 20~ (night) and 75% relative humidity, and Commelina communis L. was grown in a 16-h photoperiod, 25 ~ C (day), 20 ~ C (night) and 80% relative humidity. Protoplast preparation. Guard-cell protoplasts (GCP) and mesophyll-cell protoplasts (MCP) were prepared from the leaves of three-week-old V.faba, including a final purification on density gradients as described by Feyerabend and Weiler (1988). The GCP preparations contained a minimum of 98% of the desired cell type. The major contaminations were mini-protoplasts derived from mesophyll cells (typically < 1%) and occasional chloroplast aggregates from lysed mesophyll cells. Epidermal-cell protoplasts were also occasionally observed ( < 1%). The MCP preparations were composed almost exclusively ( > 98%) of mesophyll cells with some aggregated chloroplasts from lysed cells. In some preparations, up to 5-10% of mesophyll vacuoles were present. Such preparations were also used for the immunofluorescence assays because they allowed MAB binding to tonoplast membranes to be checked. Mesophyll-cell protoplasts from the other plant species were prepared similarly, but using the following enzyme solutions: Nicotiana and Arabidopsis: 2% Onozuka R-10 cellulase, 0.02% Pectolyase Y23, 1% bovine serum albumin (BSA); Zea and Commelina: 1.5% Onozuka R-10, 0.025% Pectolyase Y-23, 1% BSA; Pisum: 5% Cellulysin, 1% Onozuka R-10, 0.02% Pectolyase Y-23, 1% BSA.

Preparation of plasmalemma vesicles. Plasmalemma vesicles from the leaves of three-week-old V.faba plants were prepared from epidermal peels by aqueous two-phase partitioning (Larsson 1985) as described in Blum et al. (1988), but using a polymer concentration of 6.4% and 3 mM KC1. The purity of the preparations was the same as that given in Blum et al. (1988). Immunization and cell fusion. The following cell fractions were used as immnnogens: total pelletable (80000.g) membranes of V.faba GCP (30 lag of protein per injection per animal) and plasmalemma vesicles of V.faba epidermis (10 lag protein per injection per animal). The "long-term" immunization schedule was as in Weiler (1986); "short-term" immunization was as follows: day 0 : 1 0 lag of protein, intraperitoneally (i.p.), in the presence of complete Freund's adjuvant (CFA); day 7 : 1 0 lag of protein, i.p., in incomplete Freund's adjuvant (ICFA); day 14:10 gg of protein, i.p., in the absence of adjuvant; day 17: fusion. All protein determinations were performed with the Bradford assay (assay kit, Biorad, Mtinchen, FRG). The fusion protocol, selection of myelomas in media containing hypoxanthin, aminopterin and thymidine, as well as the culture of hybridomas and MAB production in vitro and

473

in vivo are described elsewhere (Mertens et al. 1983; Eberle et al. 1986).

hnmobilized-protoplast immunofluorescence assay. Terasaki plates were coated for 15 min at 22 ~ C with 1.5.103 GCP and/or 1.5.103 MCP in 10 lal of protoplast-isolation medium lacking the hydrolytic enzymes (0.5 M mannitol, 1 mM CaCI2, 10 mM ascorbate in 10 mM KPi, pH 6.5). After this time, the protoplasts had adsorbed to the polystyrene. Plates were carefully aspirated using a pasteur pipet with constricted tip and mild suction. The plates were washed once with washing buffer (phosphate-buffered saline, PBS): 8 m M K2HPO4, 2 m M NaH2PO4, 150 mM NaCI2, pH 7.4, containing 0.2 M mannitol, 0.1% gelatin, pH 7.4, 500 mosm). The washing solution was aspirated and 10 lal of hybridoma culture supernatants, or MAB preparations, previously adjusted to 500 mosm by the addition of 1/3 vol. of 1 M mannitol, 0.1% gelatin, in PBS, was added and incubated for 60 min at 22 ~ C. After aspiration, the plates were washed three times with washing buffer and then incubated for 1 h in the dark at 22 ~ C with fluorescein isothiocyanate (FITC)-labeled sheep-anti-mouse immunoglobulin (Sigma, No. F-2883), diluted 1 : 50 in washing buffer. The plates were then washed three times as above, and inspected in an inverted position under oil in a Zeiss (Oberkochen, FRG) Standard epifluorescence microscope (objective 461706-9904, 40:1, aperture 0.85 oil). Light source: HBO 50W high-pressure arc lamp, in combination with excitation filter (Zeiss No. BP 485) beam splitter FT 510 and band-pass filter BP 515-565. This combination excludes chlorophyll fluorescence. Photographic recordings were made on Ilford HP5, 400 ASA film developed in Kodak HC 110 developer. Exposure times were 1 rain for fluorescence images and were set automatic for brightfield pictures. Enzyme-linked immunoassay (ELISA). Microtitration plates (flat-bottom type) were coated at 4 ~ overnight or for 2 h at 37~ with 50 lal plasmalemma vesicles (0.5 lag protein per well) in PBS, pH 7.4. Plates were then decanted, washed once with PBS and incubated for 15 min at 37 ~ C with 0.1 ml per well of PBS containing 0.2% gelatin (blocking of residual protein-adsorption sites). After washing with PBS, plates were flicked dry and 50 lal hybridoma culture supernatants or appropriate MAB-containing fractions were added. Incubation took place for I h at 37 ~ C, then the plates were washed three times with PBS and flicked dry. Bound MAB was detected by incubating the plates for 1 h at 37 ~ C with 50 lal of horseradishperoxidase-labeled goat anti-mouse immunoglobulin (Biorad, Cat. No. 170-6516), diluted 1:7000 in PBS with 0.2% gelatin. After washing (3 • 50 lal of tetramethylbenzidine (62.5 lag-ml 1 in 0.1 M acetate-citrate buffer, pH 6.0 containing 2.4 x 10 3% H202 ) were added to determine peroxidase activity. After 10-20 min at 22 ~ C, the reaction was stopped with 2 N H2SO4 and the absorbance was read in a Dynatech (Denkendorf, FRG), MR600 ELISA reader at 450 nm. To detect periodate-sensitive epitopes, plates were incubated, after the gelatin blocking step, with 40 mM NaIO4 in 50 mM sodium acetate, pH 4.5 (15 min at 22 ~ C). Controls were incubated in acetate buffer only. After washing (3 • PBS) the wells were filled with 50 121sodium borohydride (60 mM in PBS, pH 7.4), incubated for 30 rain at 22~ C and washed again (3 • PBS). The assay was then continued as above. To determine protease-labile epitopes, wells were, after the incubation with plasmalemma vesicles and one PBS wash, incubated with 50 lal trypsin (50 lag. ml-1 in 50 mM 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid (Hepes)-KOH buffer, pH 7.4). Controls were incubated with 50 lal trypsin (50 lag. ml-1) plus trypsin inhibitor (250 gg-ml 1) in the same buffer.

474

G. Key and E.W. Weiler: Monoclonal antibodies against guard-cell plasma membranes

Incubation times were 1 h at 37 ~ C, followed by five PBS-wash cycles. Then the assay was continued as above. To determine the ability of simple sugars to compete with the membrane antigens for antibody binding, the antibodycontaining hybridoma media were suitably diluted in PBS (containing 0.2% gelatin). An aliquot of these fractions was mixed with an equal volume of sugar (2 mM in 50 mM KPI buffer, pH 7.4) and pre-incubated for 1 h at 22 ~ C (final sugar concentration 1 mM). Then, 50 lal of the pre-incubated antibody was delivered into plasma-membrane-coated wells of microtitration plates and the ELISA continued as described above.

Gel electrophoresis and Western blot analysis. Plasmalemma proteins were precipitated with methanol as follows: The membrane vesicles (1 mg protein), suspended in I ml PBS were mixed with a threefold excess of ice-cold methanol, incubated for 5 min at 0 ~ and centrifuged for 3 min at 10000.g. The pellet was air-dried and resuspended in sample buffer (Laemmli 1970) containing 5% sodium dodecyl sulfate (SDS), boiled for 3 min and then subjected to SDS-polyacrylamide gel electrophoresis (PAGE) using the system of Laemmli (1970) and 10% separation gels (85.85 mm, 1.0 mm thick). Protein was stained with Coomassie brilliant blue, or the gel was electroblotted onto nitrocellulose membranes (Towbin etal. 1979) using 20 mM 2-amino-2-(hydroxymethyl)-l,3-propanediol (Tris), pH 8.3, 150 mM glycine, 0.1% SDS and 20% methanol as blotting-buffer. Transfer of proteins was performed for 12 h at 4 ~ C (0.7 mA.cm-2). After incubation for 1 h in 1% BSA (22 ~ C), the nitrocellulose strips were incubated for 1 h at 22 ~ C with MAB suitably diluted in Tris buffered saline (TBS) containing 0.05% Tween 20 and 1% BSA. Strips were then washed three times in 0.05% Tween 20 in TBS. The bound MAB were visualized by alkaline-phosphatase-labeled goat anti-mouse immunoglobulin using 5-bromo-4-chloro-3-indolyl phosphate as a substrate (Protoblot-protocol).

Lipid andprotein separation. Membrane vesicles were separated into a lipid- and protein-rich fraction according to Stein and Smith (1982). Briefly, membrane vesicles corresponding to 1 mg of protein, suspended in 1 ml PBS, were mixed with 19 ml chloroform:methanol, 2:1 (v/v) and vortexed. Then, 4 m l 0.9% NaC1 was added and the phases mixed again. Phase separation required mild centrifugation (5 min, 1000.g). The protein precipitate was recovered from the interphase and diluted with an excess of ice-cold methanol. After centrifugation for 3 min at 10000.g, the pellet was washed once with CHCI3: methanol, 2:1 (v/v) and recentrifuged. The residual, lipid-containing phases were all pooled and the organic solvents evaporated in vacuo. The residual aqueous fractions were adjusted to 5 ml with PBS (lipid emulsion). Aliquots (50 I~1)of the lipid emulsion in PBS and of the protein precipitate (resuspended in 5 ml PBS as a fine suspension) were then used to coat microtitration plates (overnight at 4 ~ C). These solid-phases were used for ELISA which was performed as described above.

Results

Immune response. Balb/c mice were either immunized with total GCP membranes or with purified plasma membranes from epidermal preparations. Guard-cell protoplasts represent a highly enriched population of specifically differentiated cells which can be discriminated easily from protoplasts or other leaf cell types such as epidermal or mesophyll

1.0 E 0.8 O

0.6 O 0.4 0.2 0~ ,

I

II

Ill

I

I

IV

JL V

X63

VII

VI

VIII

PAl Plate number

Fig. 1. Primary screening of fusion G26 by ELISA using immobilized plasmalemma vesicles prepared from abaxial epidermes of Viciafaba leaves by aqueous two-phase partitioning. Immunogen: epidermal plasmalemma vesicles. X63=myeloma line X63/Ag 8.653; PAl = myeloma line PAl which served as fusion partner

cells. The epidermal preparation, although it contains membranes from different cell types, is highly enriched for plasma membrane and its properties have been described in detail elsewhere (Blum et al. 1988). Both antigens were administered intraperitoneally in low doses (10 ~tg of plasmalemma vesicles, or 30 lag of GCP membranes, corresponding t o 106 cells of protein per injection) and were found highly immunogenic in both short-term and longterm immunization experiments (extending over 17 d (short-term) or up to 84 d (long-term), see Materials and methods). From altogether three fusions using spleen cells from animals immunized with total GCP membranes, 40 positive hybridomas were initially identified by ELISA. From seven fusions (animals immunized with epidermal plasma membranes), a total of 120 hybridomas (representing 76% of wells exhibiting growth) were initially selected by ELISA. A typical fusion screened by ELISA is shown in Fig. 1. Because the main objective of the present work was to identify MAB against cell-type-specific cellsurface antigens, all ELISA-positive MAB selected initially were subjected to indirect immunofluorescence and only MAB positive in this assay were characterized further. Only a selection of these, representative of the types of responses obtained, are reported here. The general characteristics of these MAB are shown in Table 1. All MAB originating from fusions in which B-lymphocytes from mice immunized according to the long-term schedule were used belonged to the immunoglobulin-G

G. Key and E.W. Weiler: Monoclonal antibodies against guard-cell plasma membranes

475

Table 1. General characteristics of representative MAB against plant membrane antigens Hybridoma code

Ig-subclass

Ascites

Immunogen and immunization

Total number of positive clones

G10-4-C3 G26-5-B4 G26-6-B2 GI 6-1-B6 G23-7-D6

IgG3 [gM IgG2b [gG2b [gG1

+ + + + +

Whole GCP protein "long-term" immunization Epidermal plasmalemma "short-term" immunization Epidermal plasmalemma "short-term" immunization Epidermal plasmalemma "long-term" immunization Epidermal plasmalemma "long-term" immunization

12 44 44 11 2

1.0 1.2

E

~

0.6

O0.4

=El.0

f

0.8

/

0.6

0.2. 0

~:

1.4

o

0.8

/~

0.4 .

.

.

.

.

5

.

.

.

.

.

.

10

.

.

.

.

.

.

15

Time [hi Fig. 2. Time course of adsorption of plasmalemma vesicles (whole leaf of Vfaba) to polystyrene surfaces at (o e) 4~ C and (o -o) 37~ C. Data are averages of n= 10 replicas; S D < 5%. Adsorption was followed with saturating amounts of MAB G10-4-C3

(IgG) subclass whereas a significant proportion of MAB belonging to the IgM subclass were found when mice immunized only for short periods served as spleen donors. All MAB described here can be produced either in vitro, by fermentation of hybridomas, or in vivo, through ascites fluid.

Enzyme-linked immunosorbent assay. The rationale was to devise a simple, large-capacity screening assay allowing the rapid identification of positive hybridomas producing MAB against epitopes of membrane antigens in their native conformation. At the same time a set of routine assays was devised, allowing the discrimination of carbohydrate epitopes from protein epitopes. The ELISA using polystyrene-adsorbed plasmalemma vesicles appeared to be the method of choice for a first screening of fusions. Plasmamembrane vesicles prepared by aqueous two-phase partitioning adsorb strongly to polystyrene surfaces of microtitration plates at pH 7.4 (Fig. 2) and the adsorption is complete after 2 h with tempera-

0.2

/e

~,.~.~?~?.-..~-. 0.1 1

................................... 10 100 1000

Membrane protein [ng per well]

10000

Fig. 3. Standard curve for the determination of plasmalemma protein adsorbed to polystyrene surfaces by MAB G10-4-C3. Plasmalemma vesicles were prepared from whole V.faba leaf by aqueous two-phase partitioning. X-axis: amount of membrane protein added per well. The bars represent 4-SD of triplicates in a typical experiment

tures ranging from 4 to 37 ~ C. Routine coating was performed overnight at 4 ~ (pH 7.4). Optimum saturation of the polystyrene surface was obtained with 0.5 lag membrane protein per well (in 50 I-d, exposed surface area 82 mm2), but much lower amounts can be used, if the antigen preparation is limiting. The dose-response relationship between the amount of coating-protein and signal strength (enzyme activity immobilized onto the membrane vesicles) is strictly quantitative, and this permits us to quantitate as little as 10 ng of membrane protein by ELISA using MAB G10-4-C3 which produces exceptionally strong signals (Fig. 3). On continuous sucrose-density gradients loaded with plasma-membrane vesicles as described previously (Feyerabend and Weiler 1988), the immunoreactive fractions sedimented at an equilibrium density of 1.13 g. cm-3 as did the specific plasmalemma marker, vanadate-sensitive K +, Mg2+-ATPase (Fig. 4). These results strengthen the conclusion that MAB G10-4-C3 indeed labels the plasma membrane.

476

G. Key and E.W. Weiler: Monoclonal antibodies against guard-cell plasma membranes fluorescence intensities between specimens. It was f o u n d o p t i m u m to use Terasaki-plates as carriers which permit a low-volume screening and can be handled directly u n d e r an inverted or standard microscope at high magnification. Per well, 1500 cells (ratio M C P : G C P = 1:1) are sufficient (see Fig. 5). The optical properties o f the system are g o o d enough to allow even the discrimination o f different fluorescence patterns on the immunolabeled cell surfaces (see Fig. 5).

0.5

TO.4

/

~0.3"

6o 4o~

9; 0.2

0

2O =~ gO

02

Monoclonal antibodies identify common and differentiation-specific antigens on the plasma membrane of guard cells of Vicia faba L.

Monoclonal antibodies were raised against the plasma membrane of Vicia faba L. guard cells by immunizing either with total membranes from purified gua...
2MB Sizes 0 Downloads 0 Views