A monoclonal antibody (JIM4) has been derived that recognizes a series of glycoproteins associated with the plasma membrane of a suspension-cultured carrot cell line and also an arabinogalactan proteoglycan secreted by the cultured cells. Immunocytochemistry indicated that the plasma membrane antigen(s) recognized by JIM4 are specific to certain cells of Daucus carota L. seedlings. In the root apex JIM4 labelled two segments of the stele. These were centred upon the poles of the protoxylem. An axis of unlabelled cells connected the two phloem regions. Two sections of the pericycle with characteristic oblique longitudinal divisions were particularly reactive with JIM4. This pattern of reactive cells, reflecting cell position rather than a specific future cell type, would appear to be a unique observation in plants. The association of JIM4 antigens with these vascular tissues is maintained through the transition from root to the shoot tissue of the cotyledons and the mature plant. Examination of JIM4 labelling upon ultrathin frozen sections of the carrot seedling root apical meristem indicated that the expression of the antigen is a very early event in root development. Cells express the surface epitope, within one or two cells of the dome of apical initials, before the pattern of future vascular tissue can be discerned and well before its actual differentiation.

     
  • AGPs

    arabinogalactan proteins

  •  
  • DAPI

    4’,6-diamidino-2-phenyl-indole

  •  
  • ELISA

    enzyme-linked immunosorbent assay

  •  
  • McAb

    monoclonal antibody

  •  
  • PBS

    phosphate-buffered saline

  •  
  • SDS–PAGE

    sodium dodecylsulphate-polyacrylamide gel electrophoresis

  •  
  • TBS

    Tris-buffered saline

All of the cells in a higher plant derive ultimately from the dividing cell populations in the shoot and root apical meristems. These meristems, in giving rise to the organs of the plant, produce cell lineages that develop into the characteristically patterned vasculature of these structures. Despite our knowledge of the physical structure of many apical meristems and some of the physical determinants of their anatomy, the molecular processes that determine plant morphology and cell differentiation remain unknown.

As dividing cells in the meristematic region of a root move back from the apex, as files of related cells, two patterns are established that reflect the mature root morphology. A concentric pattern underlies the classical division into three tissue systems - epidermis, cortex and stele. Within the stele there is a species-specific radial pattern of vascular development into phloem and xylem. In the case of carrot (Daucus carota L.), the root shows a bilaterally symmetrical diarch structure. The establishment of such tissue patterns and subsequent differentiation requires the acquisition of some form of identity by cells or cell lineages, although the chemical nature of any such identity is unknown. To understand how these patterns of xylem and phloem are established, we need molecular markers of early pattern formation, or cell identity, at a time well before the cells finally differentiate into recognizable cell types.

Arabinogalactan proteins (AGPs) form a very large and diverse group of macromolecules in plants and can be subdivided most readily into extracellular proteoglycans and membrane-associated glycoproteins (Fincher et al. 1983; Pennell et al. 1989). No clear function for any of these molecules has emerged.

AGPs are antigenic and capable of generating monoclonal antibodies (McAb) with reactivities inhibitable by L-arabinose, D-galactose and/or associated disaccharides (Anderson et al. 1984). Such antibodies have been generated in response to complex membranous plant immunogens (Meyer et al. 1987; Bradley et al. 1988; Brewin et al. 1988). In one case an antibody has been shown to be specific for AGPs and not to crossreact with other cell surface arabinosylated and hy droxyproline-containing glycoproteins such as the extensins and the lectins of the Solanaceae (Pennell et al. 1989).

In this report we describe the specific, restricted and novel distribution of a set of plasma membrane antigens in seedlings of Daucus carota L. The McAb JIM4 recognized an epitope common to a set of glycoproteins of the plasma membrane of a carrot cell line and an arabinogalactan protein secreted by the cell line. We demonstrate that in the carrot seedling the most abundant expression of the membrane-associated epitope occurs in the vascular tissues. Examination of the root apical meristem of carrot seedlings indicated that the expression of the epitope is a very early event in the continuing development of the root meristem and seems to reflect early stages in the formation of the vascular pattern.

Plant materials and cell culture

Carrot (Daucus carota L. cv. Early Nantes) seeds were germinated on moist tissue paper and grown for 5 days in the dark at 22°C before use. Mature carrot plants were collected locally. A suspension culture of carrot cells (Lloyd et al. 1979) was maintained in Murashige and Skoog medium supplemented with 1 mgl-1 2,4-dichlorophenoxy-acetic acid and 25 gl-1 sucrose. Cells were subcultured every 7 days and used for experimental purposes 5 or 6 days after subculture.

Immunization and production of hybridoma

The hybridoma secreting the monoclonal antibody JIM4 was developed from a series of immunizations of rats with intact protoplasts prepared from the carrot suspension cell culture. The protoplasts (1×106 cells in 300 μ1 phosphate-buffered saline, PBS) were injected intraperitoneally into a male LOU/c rat (5 weeks old) on days 0, 22, 56 and 103. Spleen lymphocytes were collected and fused with the IR983F myeloma cell line (Bazin, 1982) on day 106. Procedures used for the fusion, HAT selection and maintenance of hybridomas were essentially as described by Galfre & Milstein (1981).

The JIM4 McAb was selected as a plasma membrane reactive antibody by its ability to bind to a preparation of membranes from the carrot cell line and indirect immunofluorescence on carrot protoplasts prepared from the cultured cells. The JIM4 hybridoma secretes immunoglobulins of class IgM.

Preparation of protoplasts

The procedures used for the preparation of protoplasts from the suspension-cultured cells and their use for obtaining indirect immunofluorescence with plasma membrane-reactive antibodies are described elsewhere (Pennell et al. 1989).

Preparation of microsomal membrane fractions

Protoplasts, prepared from the carrot cell line or the intact carrot seedlings, were homogenized in 4 mlg-1 fresh weight of 50 mm-Tris(hydroxymethyl)aminomethane-HCl (TRIS) buffer pH 7·5 containing 0·25 m-sucrose, 3mm-disodium ethylenediaminetetra-acetate (EDTA), 2·5mm-dithiothreitol (DTT) and 1 mm-phenylmethylsulphonyl fluoride (PMSF). Homogenization was performed with a glass hand-held homogenizer. A pellet collected by centrifugation at 5000g for 10 min was discarded and the subsequent pellet at 100000 g (1 h) was washed in the homogenization buffer and resuspended in 5 mm-potassium phosphate buffer pH 7·8 (containing the additions of the homogenization buffer other than the EDTA) and retained as the membrane fraction and stored at −20°C. All procedures were performed at 4°C. Protein was determined according to the method of Lowry et al. (1951).

Preparation of a fraction containing the arabinogalactan protein of the carrot culture medium

The culture medium of the carrot cell line, conditioned by 5 days growth subsequent to subculture was separated from the cells by centrifugation and filtered through Whatman No. 1 paper and brought to 90 % (v/v) acetone at 4 °C. After 30 min the precipitate was collected by centrifugation, extracted with water and the soluble components lyophilized.

Electrophoresis and immunoblotting

Sodium dodecylsulphate-polyacrylamide gel electrophoresis (SDS–PAGE) was performed using 10% or 8% (w/v) acrylamide slab gels according to the method of Laemmli (1970). Gels were blotted on to nitrocellulose by means of a semi-dry electroblotting system (Sartorius, Goettingen, FRG). Nitrocellulose sheets were blocked with 5 % (v/v) calf serum in PBS for at least 1 h before incubation with a 100-fold dilution of McAb culture supernatant in the same buffer overnight at 4°C. Extensive washing in PBS containing 0·05 % Tween 20 was performed before and after incubation with a 2000-fold dilution of rabbit anti-rat Ig linked to alkaline phosphatase (Sigma) in the PBS containing calf serum for 2 h. The enzyme substrate was developed according to the manufacturer’s method.

Enzyme-linked immunosorbent assays

ELISAs were performed in microtitre plates coated with the membrane preparation at 50 mg I-1 protein (18 h at 4°C) and blocked for at least 1 h with 5 % (v/v) calf serum in PBS. JIM4 binding was detected by means of a second antibody (rabbit anti-rat Ig linked to horseradish peroxidase, ICN Biomedicals) developed by conventional reactions. The dilution of JIM4 giving 90 % of maximal binding was used for assessment of hapten and glycoprotein inhibition. In certain cases, the immobilized antigens were treated prior to antibody incubations with Pronase E (Sigma) at 1 g1-1 in 50 mm-TRIS-HCl buffer pH 7·5 or 25mm-sodium metaperiodate in 50 IDm-sodium acetate buffer, pH 4·3 for Ih in the dark. After such treatments the plates were washed extensively in water and re-blocked prior to antibody incubations and assessment of antigen degradation.

Immunogold electron microscopy

The fixation and subsequent treatment of suspension-cultured carrot cells and carrot seedlings for the analysis of JIM4 binding by means of immunogold electron microscopy is described elsewhere (Pennell et al. 1989).

Tissue fixation and microtomy

Root apices of 5-day-old carrot seedlings were excised and immersed in freshly prepared 4% (w/v) formaldehyde in fixation buffer (50mm-piperazine-N,N’-bis[2-ethanesulfonic acid] {Pipes}, 5mm-MgSO4 and 5mm-ethylene glycol bis[β-aminoethylether]N,N,N’, N’-tetra acetic acid {EGTA}, pH 6·9) for 2h, washed in fixation buffer and infused for at least 72 h with 1·5 m-sucrose and 0·5% formaldehyde in the same buffer. Before sectioning, root apices were trimmed to approx. 1mm, transferred to freezing stubs and plunged into liquid ethane. The stub with frozen tissue was fitted to an

Ultracut EFC 4D cryoultramicrotome (Reichert-Jung, UK, Slough) and the tissue sectioned at −110°C at a thickness of 0·5 μm. Sections were collected on small drops of 2m-sucrose in water and settled on to multiwell slides. Before use the slides were extensively washed with Tris-buffered saline (TBS). Plant tissues other than root apices were similarly immersed in 4% formaldehyde in fixation buffer for at least 2h before embedding with OCT compound (Miles Scientific, Illinois, USA) and frozen at −20°C. Sections (5−10 μm) were made at -20°C using a Bright 5030 cryomicrotome (Cambridge, UK) and collected on multiwell slides coated with poly-L-lysine and allowed to dry. Before use the slides were extensively washed with water to remove the OCT compound.

Immunocytochemistry

The sections were treated with a 5-fold dilution of hybridoma culture supernatant into 5 % (v/v) calf serum in TBS for up to 12 h at 4 °C. The sections were rinsed in TBS before treatment with goat anti-rat Ig linked to fluorescein isothiocyanate, (ICN Biomedicals) diluted 100-fold into the TBS with calf serum, for at least 2h. The final, extensive washing of the slides in TBS also included a 30s incubation with 4’,6-diamidino-2-phenyl-indole (DAPI) at Imgl-1 in TBS. Final mounting was with Citifluor anti-fade mountant (Citifluor, London, UK) and the sections were observed on a Zeiss Photomicroscope III equipped with epifluorescence irradiation.

Characterization of JIM4 antigens of the plasma membrane of suspension-cultured carrot cells

The McAb JIM4 bound to the surface of protoplasts prepared from the carrot cell line as revealed by indirect immunofluorescence (Fig. 1A). The specific localization of the JIM4 epitope to the plasma membrane of these cells was confirmed by means of immunogold electron microscopy performed on ultrathin sections of such cells retaining an intact cell wall (Fig. 1B).

Fig. 1.

(A) Immunofluorescence generated by JIM4 binding to the surface of intact protoplasts prepared from the carrot cell line. Note the agglutination of the protoplasts. Bar, 10 μm. (B) Immunogold localization of JIM4 antigens to the plasma membrane (pm) of carrot suspension cell, w, cell wall; c, cytoplasm. Bar, 0·1 μm. (C) Immunogold electron microscopy confirms JIM4 binding to the plasma membrane of cells from a carrot seedling root, w, cell wall; c, cytoplasm. Bar, 0·1 μm.

Fig. 1.

(A) Immunofluorescence generated by JIM4 binding to the surface of intact protoplasts prepared from the carrot cell line. Note the agglutination of the protoplasts. Bar, 10 μm. (B) Immunogold localization of JIM4 antigens to the plasma membrane (pm) of carrot suspension cell, w, cell wall; c, cytoplasm. Bar, 0·1 μm. (C) Immunogold electron microscopy confirms JIM4 binding to the plasma membrane of cells from a carrot seedling root, w, cell wall; c, cytoplasm. Bar, 0·1 μm.

Immunoblotting of electrophoretically separated membranes indicated that JIM4 recognized a series of discrete bands in the range of Mr 20000 to 60000 (Fig. 2). These corresponded with the binding pattern of an anti-AGP McAb MAC 207 (Pennell et al. 1989). In addition, JIM4 also reacted with an arabinogalactan protein (Pennell et al. 1989), derived from the culture medium of the carrot cells, which was electrophoretically resolved as a smear of Mr 70000 to 100 000 (Fig. 2).

Fig. 2.

Immunoblotting of JIM4 (lane a) and MAC 207 (lane b) to the membrane preparation of carrot protoplasts separated by SDS–PAGE and transferred to nitrocellulose. Loading was at 50 μg protein per lane. JIM4 also reacted with the arabinogalactan protein (Mr 70000 to 100000) in an acetone precipitate of the conditioned medium of the carrot cells (lane c, 10μg protein loading). Positions of protein markers (Mr×10−3) and the dye front (arrowhead) are indicated.

Fig. 2.

Immunoblotting of JIM4 (lane a) and MAC 207 (lane b) to the membrane preparation of carrot protoplasts separated by SDS–PAGE and transferred to nitrocellulose. Loading was at 50 μg protein per lane. JIM4 also reacted with the arabinogalactan protein (Mr 70000 to 100000) in an acetone precipitate of the conditioned medium of the carrot cells (lane c, 10μg protein loading). Positions of protein markers (Mr×10−3) and the dye front (arrowhead) are indicated.

Further analysis of the JIM4 antigen and epitope was by means of an enzyme-linked immunosorbent assay (ELISA) of its binding to membranes prepared from protoplasts of the carrot cell line. Treatment of the immobilized membranes with either periodate or a protease reduced JIM4 binding by 76 % and 68 %, respectively, indicating that the antigen contained both carbohydrate and protein components. The binding of JIM4 to the membranes was inhibited by 50 % in the presence of the exuded AGP of Acacia Senegal at 60 mg1-1, but not by the Solanum tuberosum lectin at up to 2 g1-1. These characteristics are essentially similar to those of MAC 207 (Pennell et al. 1989) but differ in our observation that JIM4 displayed no hapten inhibition by any tested mono- or disaccharides, (see Anderson et al. 1984 and Pennell et al. 1989).

Distribution of JIM4 antigens in root apex

Indirect immunofluorescence, utilizing JIM4, on sections prepared from the root apices of carrot seedlings, revealed a specific and restricted distribution of its antigens. This is in stark contrast to MAC 207 which recognizes the plasma membrane of all cells in the carrot root (Pennell et al. 1989, and see Fig. 3B). Immunogold electron microscopy confirmed the plasma membrane location of the JIM4 antigens in the carrot seedling (Fig. 1C).

Fig. 3.

(A) Immunofluorescence generated by JIM4 on a transverse section of a carrot root, 50−100 μm from the apical initials is restricted to two segments of the vascular cylinder and is most abundant in the cells of the pericycle (p). At the centre of the cylinder 3 xylem vessel mother cells are weakly labelled (The centre cell is indicated with x). The adjacent regions that will develop into the phloem (ph) are unlabelled, c, cortex. The arrow and arrowhead refer to the sections shown in Fig. 4. Bar, 50 μm. (B) Immunofluorescence generated by MAC 207 on a comparable section to A reveals reactivity with all cells. Bar, 50 μm. (C) Anatomical diagram of transverse section through the carrot root apex to show position of JTM4-reactive cells (fine shading) in relation to future vessel elements. Inner emphasized line is the boundary between future pericycle and endodermis. Outer emphasized line is the boundary between root epidermis and ensheathing root cap. Adapted from Esau (1940).

Fig. 3.

(A) Immunofluorescence generated by JIM4 on a transverse section of a carrot root, 50−100 μm from the apical initials is restricted to two segments of the vascular cylinder and is most abundant in the cells of the pericycle (p). At the centre of the cylinder 3 xylem vessel mother cells are weakly labelled (The centre cell is indicated with x). The adjacent regions that will develop into the phloem (ph) are unlabelled, c, cortex. The arrow and arrowhead refer to the sections shown in Fig. 4. Bar, 50 μm. (B) Immunofluorescence generated by MAC 207 on a comparable section to A reveals reactivity with all cells. Bar, 50 μm. (C) Anatomical diagram of transverse section through the carrot root apex to show position of JTM4-reactive cells (fine shading) in relation to future vessel elements. Inner emphasized line is the boundary between future pericycle and endodermis. Outer emphasized line is the boundary between root epidermis and ensheathing root cap. Adapted from Esau (1940).

The place and plane of sectioning that provided the most informative and clearly resolved specificity of the restricted distribution of the JIM4 antigens was a transverse section through the root apex as shown in Fig. 3A. At this region, 50-100 gm from the most apical meristematic cells, the predominant feature was the labelling of the plasma membranes of cells in two segments of the stele. The segments are centred upon the protoxylem poles of the diarch xylem. The larger cells of the future xylem axis, weakly labelled in the centre of the stele (Fig. 3A), are the vessel mother cells of the future xylem plate (Esau, 1940). The sites of the protophloem lie perpendicular to this region and are essentially unlabelled with JIM4. The most intense labelling with JIM4 occurred in two arcs of pericycle cells with oblique longitudinal divisions whereas the adjacent cells of the endodermis are labelled only weakly (Fig. 3A). Esau (1940) has described the occurrence of such divisions of the pericycle cells, but does not appear to have noted their restriction to the two regions that can be clearly seen in Fig. 3A. It is of interest that the MAC 207 epitope appeared more abundant in these regions although also occurring on the plasma membrane of every cell of the root apex (Fig. 3B). The distribution of JIM4-reactive cells in relation to the tissues of the carrot root apex is shown diagrammatically in Fig. 3C which is adapted from Esau (1940).

Serial longitudinal sectioning through the root meristem of a 5-day-old carrot seedling was performed to determine the expression of the JIM4 antigens in relation to the developing meristem. Immunofluorescent micrographs of two non-median and a median section through the most apical region of the root apex are shown in Fig. 4. The photographs are reproduced to allow observation of labelled cells in relation to nonlabelled cells. The plane of sectioning in relation to the JIM4 pattern of labelling of the transverse section is indicated in Fig. 3A. The specific labelling of certain cell lineages was observed to be a very early event of development, being observed to within one or two cells, 20 μm, of the most apical cells. The cells labelled in Fig. 4C are of the developing pericycle. In no cases were cells of the root cap observed to be labelled.

Fig. 4.

JLM4-generated immunofluorescence restricted to cells of the stele in serial longitudinal sections of a carrot apex. The plane and direction of sectioning in relation to Fig. 3A is shown in that figure with the arrowhead. (A) A non-median section in which all the cells of the vascular cylinder appear to be labelled as well as isolated groups of distal cells (arrowheads) and certain epidermal cells (e). (B) A further non-median section closer to the centre of the root, r, root cap. (C) A median section (position indicated by the arrow in Fig. 3A) indicating JIM4 reactivity with two lineages of pericycle cells to within 20 μm of the apical initials (a). (D) DAPI staining of DNA in the section shown in C. Bar, 100 μm.

Fig. 4.

JLM4-generated immunofluorescence restricted to cells of the stele in serial longitudinal sections of a carrot apex. The plane and direction of sectioning in relation to Fig. 3A is shown in that figure with the arrowhead. (A) A non-median section in which all the cells of the vascular cylinder appear to be labelled as well as isolated groups of distal cells (arrowheads) and certain epidermal cells (e). (B) A further non-median section closer to the centre of the root, r, root cap. (C) A median section (position indicated by the arrow in Fig. 3A) indicating JIM4 reactivity with two lineages of pericycle cells to within 20 μm of the apical initials (a). (D) DAPI staining of DNA in the section shown in C. Bar, 100 μm.

Distribution of JIM4 antigens in carrot plants

The pattern of labelling observed in the root apex was reflected in other regions of the seedling and mature carrot plant. Ultramicrotomy was not suitable for other tissues, and immunofluorescent labelling could not be so readily resolved on thicker sections. However, the two segments of labelling, each centred upon an end of the xylem axis, can be clearly seen in the stele in the hypocotyl (Fig. 5A). An equivalent section, with JIM4 omitted from the labelling procedure, is shown in Fig. 5B. The association of JIM4 antigens with vascular tissue was observed through the region of the transition of the vascular pattern from the root to that of the shoot (occurring in the upper hypocotyl), and its association with the several separate cotyledonary traces are shown in Fig. 5C. In Daucus the primary vascular tissue of the hypocotyl diverges entirely into the cotyledons and is thus not continuous with that of the epicotyl (Havis, 1939; Esau, 1940). This can be seen in a median longitudinal section through the cotyledons at the cotyledonary node and shoot apex (Fig. 5C). No labelling of cells in the shoot apex at this stage can be seen. The position of the apical meristem can be seen more clearly in the DAPI-stained image of the same section (Fig. 5D). Subsequently, of course, the vascular tissues of the root and shoot are continuous. JIM4 was found to be expressed in the shoot tissues of a mature plant. Contrasting immunofluorescent micrographs of sections of a petiole from a mature carrot labelled with JIM4 and MAC 207 are shown in Fig. 5E and 5F. JIM4 can be seen to be reactive with the xylem and associated tissues of the vascular bundles and not the surrounding parenchyma cells. In contrast MAC 207 reacted with all the tissues of the petiole and the phloem regions of the vascular bundles and the collenchyma bundles were particularly strongly labelled.

Fig. 5.

(A) Immunofluorescence generated by JIM4 on a transverse section of the hypocotyl of a 5-day-old carrot seedling is restricted to two segments of the vascular tissue and also the epidermis (e). c, cortex; vc, vascular cylinder. Bar, 100μm. (B) A comparable section to A in which the JIM4 incubation was omitted. Autofluorescence indicates the diarch distribution of the mature xylem vessels. Bar, 100 μm. (C) JIM4 immunofluorescence on a longitudinal and median section through the cotyledons at the cotyledonary node indicates reactivity remains with the vascular tissue as it diverges into the cotyledons (cd). A region of the adaxial cotyledon epidermis adjacent to the shoot apex (a) was also reactive (arrowheads). Bar, 100 μm. (D) DAPI staining of the section in 4C. (E) JIM4 immunofluorescence on a transverse section through a region of a petiole from a mature carrot plant is predominantly associated with the xylem tissues (x) of the vascular bundles, the epidermis and the outer parenchyma cells but not the collenchyma (cl), ad, adaxial epidermis. Bar, 500 μm. (F) A comparable section displaying MAC 207 immunofluorescence. All tissues are reactive including the phloem and the collenchyma bundles. Bar, 500 μm.

Fig. 5.

(A) Immunofluorescence generated by JIM4 on a transverse section of the hypocotyl of a 5-day-old carrot seedling is restricted to two segments of the vascular tissue and also the epidermis (e). c, cortex; vc, vascular cylinder. Bar, 100μm. (B) A comparable section to A in which the JIM4 incubation was omitted. Autofluorescence indicates the diarch distribution of the mature xylem vessels. Bar, 100 μm. (C) JIM4 immunofluorescence on a longitudinal and median section through the cotyledons at the cotyledonary node indicates reactivity remains with the vascular tissue as it diverges into the cotyledons (cd). A region of the adaxial cotyledon epidermis adjacent to the shoot apex (a) was also reactive (arrowheads). Bar, 100 μm. (D) DAPI staining of the section in 4C. (E) JIM4 immunofluorescence on a transverse section through a region of a petiole from a mature carrot plant is predominantly associated with the xylem tissues (x) of the vascular bundles, the epidermis and the outer parenchyma cells but not the collenchyma (cl), ad, adaxial epidermis. Bar, 500 μm. (F) A comparable section displaying MAC 207 immunofluorescence. All tissues are reactive including the phloem and the collenchyma bundles. Bar, 500 μm.

However, in these tissues, certain cells of the epidermal layer were also labelled with JIM4. The pattern of labelling could not always be clearly defined and in certain instances isolated cells or discrete groups of cells were observed to express the antigen (Fig. 4A, 5A,C). At the cotyledonary node, no continuous labelling of the epidermal tissues of the hypocotyl occurred. A very reactive region occurred on the adaxial epidermal tissues for the 100 μm adjacent to the shoot apex (Fig. 5C). Serial sectioning revealed that the reactive tissue was continuous through the node, around the apex, connecting the two regions labelled in Fig. 5C (data not shown). In the petiole the outer region of parenchyma cells was also labelled in addition to the epidermal and vascular tissue (Fig. 5E). Labelling of isolated groups of cells was observed in the root apex and the abrupt labelling of the epidermal cells was observed in certain instances but in all cases at least 100 μm from the apical initials (Fig. 4A).

JIM4 appeared to be relatively species specific in that it did not react with sections of Pisum or Allium roots or Nicotiana petioles. It did, however, cross-react with Petroselinum crispum. (Not shown).

Electrophoretic analysis of JIM4 reactive membrane antigens in intact seedlings

Immunoblotting of membranes prepared from 5-day-old canot seedlings indicated a wide range of bands and smears reactive with MAC 207 (Fig. 6). A smear of Mr 50000 to 150000 was most intensely labelled, but reactivity extended from the dye front to Mr of 200000. JIM4 reactivity, with equivalent loadings and incubations, appeared to differ only quantitatively from that of MAC 207 with a much reduced signal (Fig. 6). No obvious qualitative differences in the binding of these two McAb to the electrophoretically separated membranes were detectable.

Fig. 6.

Immunoblotting of MAC 207 (lane a) and JIM4 (lane b) against a membrane preparation from 5-day-old carrot seedlings separated by SDS–PAGE. Loading (50 μg protein per lane) and antibody incubations were equivalent. Reactivity occurred throughout the indicated Mr range and the difference in the antibodies was predominantly quantitative. Positions of protein markers (Mrx10−3) and dye front (arrowhead) are indicated.

Fig. 6.

Immunoblotting of MAC 207 (lane a) and JIM4 (lane b) against a membrane preparation from 5-day-old carrot seedlings separated by SDS–PAGE. Loading (50 μg protein per lane) and antibody incubations were equivalent. Reactivity occurred throughout the indicated Mr range and the difference in the antibodies was predominantly quantitative. Positions of protein markers (Mrx10−3) and dye front (arrowhead) are indicated.

JIM4 labels a novel distribution of plasma membrane antigens

The plasma membrane location of the JIM4 epitope is clearly seen in Figs 1 and 3A. This epitope was restricted to specific regions of the carrot seedling. The most striking and easily discernible distribution of this antigen is as a cell surface determinant on a series of cells associated with, but not unique to, the xylem and pericycle tissue in the root.

Certain cells of the pericycle, distinctive due to a series of oblique longitudinal divisions, first discernible at 40 μm from the most apical cells (Esau, 1940), are particularly reactive with JIM4. Serial sectioning of such a primary root meristem revealed that labelling of the pericycle cells occurred to within one or two cells of the apical initials. It is at this distance from the apex that the first cells of the stele, the pericycle cells, become individualized and provide the first indication of pattern arising from the meristem (Esau, 1940). It is at 30 μm (3 to 4 cells) from the apex that vessel mother cells of the future xylem plate become enlarged. These cells can be seen, though only weakly labelled, in Fig. 3A. However, the first vascular elements to reach maturity are the two protophloem sieve tubes at approximately 300 μm from the apex (Esau, 1940). It is thus clear that the specific expression of JIM4 antigens by certain cells of the stele occurred in a region of the root apical meristem well before differentiation of the vessel elements and appeared to occur in conjunction with the beginnings of pattern formation. The bilateral symmetry of JIM4 labelling within the stele reflects the position and orientation of the future vascular pattern but is not correlated directly with a certain future cell type or tissue such as the xylem, phloem or pericycle cells. These distinct groups of reactive cells appear to be related to the inheritance of JIM4 antigen expression within certain cell lineages.

An esterase activity has been shown to be an early marker of meristematic cells that will form the tissues of the stele in roots of Pisum sativum, but the precise pattern of activity in relation to future cell types is unclear (Gahan, 1981; Rana & Gahan, 1982). The reactivity of McAb JIM4 with the cell surface would seem to be unique as a marker of cell position in the pattern-forming systems of plants. There are, however, parallels with animal systems. Antigens have been detected that are markers for cell position and not cell type in retina (Trisler et al. 1981) and in the early development of chick limb buds (Ohsugi & Ide, 1986), although in these cases position is reflected in terms of a linear gradient of antigen rather than a symmetrical distribution.

JIM4 reactivity with epidermal tissues was not consistent and is not so readily interpreted. In certain instances, labelling of epidermal cells was observed to begin abruptly behind the root apex. A very active region of expression was located on the adaxial epidermal tissues of the cotyledonary node around the shoot apex but not of other epidermal tissues in this region. In the petiole of a mature plant outer cortical cells carried the JIM4 epitope in addition to the epidermal cells.

JIM4 recognizes an epitope of AGPs

The McAb JIM4 recognized a series of glycoproteins of the plasma membrane of a suspension-cultured carrot cell line and also an extracellular AGP derived from the culture medium of the cell line. These are exactly the same series of antigens recognized by the L-arabinose inhibitable anti-AGP McAb MAC 207 (Pennell et al. 1989).

By contrast, in the carrot seedling, the JIM4 epitope was restricted to a discrete series of cells as discussed above. MAC 207 recognized a determinant on all cells. Immunoblotting of membranes prepared from such seedlings revealed reaction of both antibodies with a more diverse array of membrane glycoproteins compared with the carrot cell line, although reactive components common to both preparations were observed. A predominantly quantitative and not qualitative difference in the binding pattern of these two antibodies with carrot seedling membranes was observed. In identical conditions, the JIM4-induced signal was considerably weaker than that of MAC 207. This observation suggests that though JIM4 and MAC 207 recognize distinct epitopes, (biotinylated JIM4 could not be inhibited from binding to membranes or the extracellular AGP by MAC 207; data not shown), these epitopes occur on an identical series of glycoproteins and also the extracellular AGP. Although both epitopes occur on a diverse array of glycoproteins, the evidence reported here indicates that the MAC 207 epitope can occur in the absence of the JIM4 epitope. The expression of the JIM4 reactivity could involve the addition or modification of existing sugar residues, or its absence could occur by masking of the epitope. The absence of the JIM4 epitope results in no change in the electrophoretic mobility of the antigens and presumably results from a subtle change in the glycan component of these glycoprotein antigens. It would thus appear that the plasma membranes of plant cells contain overlapping series of antigenic glycoproteins sharing distinct epitopes with each other as well as with extracellular arabinogalactan proteins.

A recent report indicates the presence on the surface of Drosophila neurones of a specific carbohydrate epitope (also occurring on certain mannose-containing plant glycoproteins) that is carried on different proteins throughout the development of the neurone and is thus characteristic of neural cell surfaces rather than the epitope-bearing proteins, (Katz et al. 1988).

AGPs are widely distributed and organ-specific forms do occur. They have been demonstrated by electrophoresis in floral tissues of Gladiolus and Lilium (Gleeson & Clarke, 1980), by Yariv reagent crossed electrophoresis in Glycine root nodules and floral and vegetative tissues of Lycopersicon peruvianum (Cassab, 1986; van Holst & Clarke, 1986) and by serological activity and chemical composition in tissues of Raphanus sativus (Tsumuraya et al. 1988). AGPs of the stigma and style of Nicotiana alata have been separated by means of crossed electrophoresis, and those of the stigma were demonstrated to be developmentally regulated (Gell et al. 1986). In all the above cases, it would appear to be the soluble extracellular AGPs that are investigated, although the contribution of related membrane-associated glycoproteins is far from clear. In no case have the differing forms been localized.

Cell- and tissue-specific carbohydrate antigens, which may possibly be AGPs, have been reported in the floral tissues of Nicotiana tabacum (Evans et al. 1988). A glycoprotein, but not with AGP characteristics, has been isolated from the culture medium of carrot cells and has been immunologically observed to be restricted to dermal tissues of carrots, although the subcellular location is far from clear (Satoh & Fujii, 1988).

As discussed above, the possible function of cell surface determinants in such restricted lineages of cells may be concerned with cell and tissue identity in relation to subsequent pattern formation. Such a role in cell identity has been suggested for AGPs in which terminal substituents of the β-galactan and protein backbone may be involved in the expression of identity of tissues or cell type (Clarke et al. 1979; Fincher et al. 1983). In animal cells carbohydrate structures are common as developmentally regulated antigens and can exert important roles in cell identity and function (Feizi, 1985; Lefrancois et al. 1985; Katz et al. 1988).

Some form of cell-cell interaction and recognition may be important during the development of organized tissue and it is important not to forget that AGPs were originally characterized as β-lectins as demonstrated by their ability to bind artificial carbohydrate antigens (Yariv reagents). However, the extent and nature of any cell interactions involving cell surface lectin-like molecules during plant developmental processes are unknown.

It is of interest that clonal analysis indicates that cell fate, in terms of differentiation, may be independent of cell lineage in the early stages of shoot meristem development (Poethig, 1987). In the carrot root meris tem, the dependence of cell fate upon cell lineage is unknown. The expression of the JIM4 epitope, reflecting the future pattern of vascular element differentiation, in cells close to the apical dome, may indicate a role for these cell surface glycoproteins in the determination of the fate of cells.

This work was supported with a grant from the AFRC Cell Signalling & Recognition programme. We acknowledge the assistance of J. Cooke of the Food Research Institute, Norwich for the handling of rats and J. King and C. Cooper for the preparation and maintenance of the hybridoma. We thank Roger Pennell for the electron micrographs and useful criticisms and discussions.

Anderson
,
M. A.
,
Sandrin
,
M. S.
&
Clarke
,
A. E.
(
1984
).
A high proportion of hybndomas raised to a plant extract secrete antibody to arabinose or galactose
.
Plant Physiol.
75
,
1013
1016
.
Bazin
,
H.
(
1982
).
Production of rat monoclonal antibodies with the LOU rat non-secreting IR983F myeloma cell line
.
Prot. Biol. Fluids
29
,
615
618
.
Bradley
,
D. J.
,
Wood
,
E. A.
,
Larkins
,
A. P.
,
Galfre
,
G.
,
Butcher
,
G. W.
&
Brewin
,
N. J.
(
1988
).
Isolation of monoclonal antibodies reacting with peribacteroid membranes and other components of pea root nodules containing Rhizobium leguminosarum
.
Planta
173
,
149
160
.
Brewin
,
N. J.
,
Wood
,
E. A.
,
Bradley
,
D. J.
,
Harding
,
S. C.
,
Sindhu
,
S. S.
,
Perotto
,
S.
,
Kannenberg
,
E. L.
&
Vandenbosch
,
K. A.
(
1988
).
The use of monoclonal antibodies to study plant-microbe interactions in the pea nodule
. In
Nitrogen Fixation: Hundred years after,
(ed.
H.
Bothe
,
F. J.
De Bruijn
&
W. E.
Newton
). Gustav Fischer Stuttgart.
Cassab
,
G. I.
(
1986
).
Arabinogalactan proteins during the development of soybean root nodules
.
Planta
168
,
441
446
.
Clarke
,
A. E.
,
Anderson
,
R. L.
&
Stone
,
B. A.
(
1979
).
Form and function of arabinogalactans and arabinogalactan proteins
.
Phytochemistry
18
,
521
540
.
Esau
,
K.
(
1940
).
Developmental anatomy of the fleshy storage organ of Daucus carota
.
Hilgardia
13
,
175
226
.
Evans
,
P. T.
,
Hola Way
,
B. L.
&
Malmberg
,
R. L.
(
1988
).
Biochemical differentiation in the tobacco flower probed with monoclonal antibodies
.
Planta
175
,
259
269
.
Feizi
,
T.
(
1985
).
Demonstration by monoclonal antibodies that carbohydrate structures of glycoproteins and glycolipids are onco-developmental antigens
.
Nature, Land.
314
,
53
57
.
Fincher
,
G. B.
,
Stone
,
B. A.
&
Clarke
,
A. E.
(
1983
).
Arabinogalactan proteins: structure, biosynthesis and function
.
Ann. Rev. Plant Physiol.
34
,
47
70
.
Gahan
,
P. B.
(
1981
).
An early cytochemical marker of commitment to stelar differentiation in roots of dicotyledonous plants
.
Ann. Bot.
48
,
769
775
.
Galfre
,
G.
&
Milstein
,
C.
(
1981
).
Preparation of monoclonal antibodies
.
Methods Enzymol.
73
,
3
46
.
Gell
,
A. C.
,
Bacic
,
A.
&
Clarke
,
A. E.
(
1986
).
Arabinogalactan proteins of the female sexual tissue of Nicotiana alata. I. Changes during flower development and pollination
.
Plant Physiol.
82
,
885
889
.
Gleeson
,
P. A.
&
Clarke
,
A. E.
(
1980
).
Arabinogalactans of sexual and somatic tissues of Gladiolus and Lilium
.
Phytochemistry
19
,
1777
1782
.
Ha Vis
,
L.
(
1939
).
Anatomy of the hypocotyl and roots of Daucus carota
.
J. agrie. Research
58
,
557
564
.
Katz
,
F.
,
Moats
,
W.
&
Jan
,
Y. N.
(
1988
).
A carbohydrate epitope expressed uniquely on the cell surface of Drosophila neurons is altered in the mutant nac (neurally altered carbohydrate)
.
EMBO J.
7
,
3471
3477
.
Laemmli
,
U. K.
(
1970
).
Cleavage of structural proteins during the assembly of the head of bacteriophage T4
.
Nature, Land.
227
,
680
695
.
Lefrancois
,
L.
,
Puddington
,
L.
,
Machamer
,
C. E.
&
Bevan
,
M. J.
(
1985
).
Acquisition of cytotoxic T lymphocyte-specific carbohydrate differentiation antigens
.
J. exp. Med.
162
,
1275
1293
.
Lloyd
,
C. W.
,
Slabas
,
A. R.
,
Powell
,
A. J.
,
Macdonald
,
G.
&
Badley
,
R. A.
(
1979
).
Cytoplasmic microtubules of higher plant cells visualised with anti-tubulin antibodies
.
Nature, Lond.
279
,
239
241
.
Lowry
,
O. N.
,
Rosenbrough
,
A.
,
Farr
,
A.
&
Randall
,
R.
(
1951
).
Protein measurement with Folin phenol reagent
.
J. btol. Chem.
193
,
265
275
.
Meyer
,
D. J.
,
Afonso
,
C. L.
,
Harkins
,
K. R.
&
Galbraith
,
D. W.
(
1987
).
Characterization of plant plasma membrane antigens
. In
Plant Membranes: Structure, Function, Biogenesis,
(ed. C. Leaver &
H.
Sze
), pp.
123
140
.
New York
:
Alan R. Liss
.
Ohsugi
,
K.
&
Ide
,
H.
(
1986
).
Position specific binding of a monoclonal antibody in chick limb buds
.
Devi Biol.
117
,
676
679
.
Pennell
,
R. L
,
Knox
,
J. P.
,
Scofield
,
G. N.
,
Selvendran
,
R. R.
&
Roberts
,
K.
(
1989
).
A family of abundant plasma membrane-associated glycoproteins related to the arabinogalactan proteins is unique to flowering plants
.
J. Cell Biol, (in press)
.
Poethig
,
R. S.
(
1987
).
Clonal analysis of cell lineage patterns in plant development
.
Amer. J. Bot.
74
,
581
594
.
Rana
,
M. A.
&
Gahan
,
P. B.
(
1982
).
Determination of stelar elements in roots of Pisum sativum L
.
Ann. Bot.
50
,
757
762
.
Satoh
,
S.
&
Fujii
,
T.
(
1988
).
Purification of GP57, and auxin-regulated extracellular glycoproteins of carrots, and its immunocytochemical localization in dermal tissues
.
Planta
175
,
364
373
.
Trisler
,
G. D.
,
Schneider
,
M. D.
&
Nirenberg
,
M.
(
1981
).
A topographic gradient of molecules in retina can be used to identify neuron position
.
Proc. natn. Acad. Sci U.S.A.
78
,
2145
2149
.
Tsumuraya
,
Y.
,
Ogura
,
K.
,
Hashimoto
,
Y.
,
Mukoyama
,
H.
&
Yamamoto
,
S.
(
1988
).
Arabinogalactan proteins from primary and mature roots of radish (Raphanus sativus L
.)
Plant Phvsiol.
86
,
155
-
160
. ‘
Van Holst
,
G.-J.
&
Clarke
,
A. E.
(
1986
).
Organ specific arabinogalactan proteins of Lycopersicon peruvianum (Mill) demonstrated by crossed electrophoresis
.
Plant Physiol.
80
,
786
789
.