Planta (1989)178:315 324
P l a n t a 9 Springer-Verlag 1989
Immunolocalization of avenin and globulin storage proteins in developing endosperm of Avena sativa L. Craig R. Lending, Ruth S. Chesnut, Katy L. Shaw, and Brian A. Larkins* Department of Botany and Plant Pathology, Purdue University, West Lafayette, IN 47907, USA
Abstract. The seed storage proteins of oats (Arena sativa L.) are synthesized and assembled into vacuolar protein bodies in developing endosperm tissue. We used double-label immunolocatization to study the distribution of these proteins within protein bodies of the starchy endosperm. When sections of developing oat endosperm sampled 8 d after anthesis were stained with uranyl acetate and lead citrate, the vacuolar protein bodies consisted of light-staining regions which were usually surrounded by a darker-staining matrix. Immunogold staining of this tissue demonstrated a distinct segregation of proteins within protein bodies; globulins were localized in the dark-staining regions and prolamines were localized in the light-staining regions. We observed two additional components of vacuolar protein bodies: a membranous component which was often appressed to the outside of the globulin, and a granular, dark-staining region which resembled tightly clustered ribosomes. Neither antibody immunostained the membranous component, but the granular region was lightly labelled with the anti-globulin antibody. Anti-globulin immunostaining was also observed adjacent to cell walls and appeared to be associated with plasmodesmata. Immunostaining for both antigens was also observed within the rough endoplasmic reticulum. Based on the immunostaining patterns, the prolamine proteins appeared to aggregate within the rough endoplasmic reticulum while most of the globulin appeared to aggregate in the vacuole. * To whom correspondence should be addressed Present address: Department of Plant Sciences, University of
Arizona, Tucson, AZ 85721, USA Abbreviations: D A A = d a y s after anthesis; IgG=immunoglobulin G; Mr = apparent molecular mass; RER = rough endoplasmic reticulum; SDS-PAGE=sodium dodecyl sulfate polyacrylamide gel electrophoresis
Key words: Arena (seed storage proteins) - Avenin - Endoplasmic reticulum - Globulin - Protein body Storage protein
Introduction
Developing angiosperm seeds store nitrogen in the form of storage proteins which accumulate in the embryo, the endosperm, or both. In many dicotyledonous plants such as legumes, the principal storage proteins are globulins. These proteins are synthesized predominantly in the cotyledons of the developing embryo (Derbyshire et al. 1976). Globulins are saline-soluble storage proteins that are distinguished on the basis of sedimentation coefficients as 7S or 11S. Both types of globulins are composed of multiple subunits that are synthesized on rough endoplasmic reticulum (RER). Immunolocalization data (Craig and Miller 1984; Greenwood and Chrispeels 1985) support the hypothesis that the products are then transported to vacuoles by way of the Golgi system, where they form insoluble aggregates which have classically been called protein bodies. In contrast to dicotyledons, most cereals have alcohol-soluble storage proteins, called prolamines, which are synthesized in endosperm tissue. Prolamines are synthesized on RER but vary with regard to transport and association into protein bodies. In maize and sorghum (Larkins and Hurkman 1978; Taylor et al. 1984), the prolamine proteins remain within the RER where they aggregate into protein bodies. Wheat and barley prolamines are transported from the RER, presumably by way of the Golgi-system (Kim et al. 1988), and form protein bodies within vacuoles. Protein bodies in
316
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
wheat have also been observed within membranes of RER, but whether or not they remain there is unknown (Campbell et al. 1981; Parker and Hawes 1982). Oat and rice seeds are unusual among cereals in that their endosperms contain relatively small amounts of prolamine proteins. The predominant storage proteins in these cereals are structurally similar to the 11S globulins of dicotyledons; however, these proteins are much less soluble in saline solutions (Brinegar and Peterson 1982; Luthe 1983). Protein bodies that form in rice endosperm have distinctive morphologies; some protein bodies are found within smooth membranes of vacuoles, whereas others form directly within membranes of the RER. Recent histochemical studies have shown that the globulin-like (glutelin) protein of rice (Takaiwa et al. 1986) is found within the vacuole whereas the prolamines aggregate into protein bodies within RER (Krishnan et al. 1986). Previous electron-microscopic studies of developing starchy endosperm of oat have shown that protein bodies are located primarily within vacuolar membranes (Saigo et al. 1983). These protein bodies generally have two regions with different staining properties when prepared for conventional electron microscopy and post-stained with uranyl acetate and lead citrate. The more predominant dark-staining regions have small locules of lightstaining material within their boundaries. Bechtel and Pomeranz (1981) used digestion with various proteases to demonstrate that these two components were different. The light-staining inclusions were assumed by Saigo et al. (1983) to be lipid bodies, although these authors speculated that the inclusions might instead be the prolamine component. Burgess and Miflin (1985) isolated oat endosperm protein bodies by sucrose density gradient ultracentrifugation and determined their polypeptide composition. Two overlapping populations of protein bodies corresponding to globulins and prolamines were observed, although the difference in their densities was not sufficient to allow their resolution. These protein bodies were not examined microscopically to determine their ultrastructural characteristics. To determine the organization and distribution of both the prolamine (avenin) and globulin proteins, we have used double-label immunogold staining with antibodies directed against these proteins to examine their distribution within developing oat starchy endosperm. The globulins occur in the dark-staining regions of the protein bodies, whereas the avenins are found in the light-staining inclusions.
Materials and methods Chemicals and plant material Bovine serum albumin (BSA), Tris (2-amino-2-[hydroxymethyl]-l,3-propanediol), Tween 20 (polyoxyethylenesorbitan monolaurate), Coomassie Brilliant Blue R, chloroauric acid, goat anti-mouse IgG (immunoglobulin G)-horseradish peroxidase, and goat anti-rabbit IgG were from Sigma Chemical Co., St. Louis, Mo., USA. Goat anti-rabbit IgG-horseradish peroxidase was from Bio-Rad, Richmond, Cal., USA, goat antimouse IgG was from Boehringer Mannheim Biochemicals, Indianapolis, Ind., USA, and LR White resin was from Structure Probe, West Chester, Penn., USA. All other chemicals were reagent grade. Oats (Arena sativa L. cv. Noble, a gift of Dr. Gregory Shaner, Department of Botany and Plant Pathology, Purdue University) were cultivated in a greenhouse with supplemental lighting for 16 h per day. The daytime photon lluence rate was 500 g m o l . m - 2 . s -1 PPFD (photosynthetic photon flux density), and the daytime and nighttime temperatures were maintained at 20 25 ~ C and 18-23 ~ C, respectively. The day of anthesis was noted for individual primary florets, and developing seeds were harvested 8 d after anthesis (DAA). The seeds were harvested and immediately fixed for electron microscopy. Fixation and embedment of seeds Oat seeds (caryopses) were prepared for electron microscopy and immunoeytochemistry by fixation in a solution that contained 1% (v/v) glutaraldehyde and 4% (w/v) freshly prepared formaldehyde in PB (phosphate buffer: 50 m M KHzPO4/ Na2HPO4, pH 7.0). The seeds were sliced into 1-mm-thick sections after they were immersed in the fixative; all steps prior to infiltration with plastic were performed on ice or at 4 ~ C. After overnight fixation, the tissue slices were washed three times (5 min each) with PB and post-fixed for 2 h in 2% (w/v) osmium tetroxide in PB. Samples were washed three times with PB and two times with distilled water (5 min each), and were then dehydrated in a graded ethanol series: 15 min each in 10%, 30%, 50%, 70%, 90%, 100% ethanol. After an additional 15 min in fresh 100% ethanol, the samples were infiltrated with 50% LR White resin in ethanol for 15 rain and then placed in 100% LR White resin for 15 min. Fresh resin was added, and the samples were infiltrated overnight; they were polymerized in fresh resin in open molds at 60 ~ C inside a sealed oven purged with nitrogen. After 24 h, samples were removed from the molds, and thin sections were cut with an Ultracut E ultramicrotome (Reichert-Jung, Vienna, Austria). Sections were collected on copper grids coated with formvar and carbon (Hayat 1986) for immunocytochemical staining. Some sections were not immunostained, but were stained with uranyl acetate and lead citrate for examination. Grids were examined with a Philips (Mahwah, N.J., USA) EM-400 electron microscope. Extraction of oat proteins Total seed protein. Oat seeds were ground to a fine flour and stirred in ice-cold acetone (3 ml/g) for 1-2 h at 4 ~ C to remove lipids. Total sodium dodecyl sulfate (SDS)-soluble oat seed protein was obtained by mixing 10 mg of the defatted oat flour in 1 ml of SDS sample buffer (62.5 mM Tris, pit 6.8, 2.3% SDS, 10% glycerol, 5% 2-mercaptoethanol). Insoluble material was pelleted by centrifugation at 14000-g for 1 rain, and the supernatant was used for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
Avenins. Avenins were extracted from defatted oat flour according to the procedure of Kim et al. (1978). Defatted oat flour was stirred in 10 volumes of 52% (v/v) ethanol for 1.5~4 h at room temperature. Insoluble material was removed by centrifugation at 10000-g (ray) for 20 min. Avenins were precipitated from the supernatant by the addition of two volumes of ice-cold 1.5% (w/v) NaC1, collected by centrifugation at 7000 .g for 30 rain, and freeze-dried. Avenin used for antibody production was redissolved in 52% ethanol and dialyzed against distilled water for 16 h at 4~ C, centrifuged at 4500.g for 30 rain, and the pellet was freeze-dried. Globulins. Defatted oat flour was stirred in 10 volumes of 50 mM Tris-HC1 and 1 M NaC1, pH 8.5, (Brinegar and Peterson 1982) for 3.5 h at room temperature to extract oat globulin. Insoluble material was removed by centrifugation at 9000.g for 10 rain. The supernatant was dialyzed against distilled water at 4~ C. Globulins were collected by centrifugation at 9000.g for 25 min, and freeze-dried.
317
A gel similar to those used for Western blots was stained with Coomassie Brilliant Blue R to visualize all proteins present in the extracts. For this gel, 30 gl of total SDS-soluble protein extract, 20 gg of o a t globulin, 40 gg of avenin, and 7 gg of BSA were subjected to SDS-PAGE in a 12.5% acrylamide gel.
Preparation of colloidal-gold conjugate Colloidal-gold conjugates for immunocytochemical stainings were prepared as described by Slot and Geuze (1985). Goat anti-rabbit IgG was complexed to colIoidal gold that had a mean diameter of 5 nm, and goat anti-mouse IgG was complexed to colloidal gold that had a mean diameter of 10 nm. The conjugates were separated from unbound antibodies by centrifugation in a 10%-30% linear glycerol gradient that contained 20 mM Tris-HC1 pH 8.2, 150 mM NaC1, 1% (w/v) BSA, and 20 mM sodium azide (Slot and Geuze 1981). The gold conjugates were stored at 4~ C and were diluted to a pale pink color immediately before use.
Immunoeytochemieal staining Preparation of antisera and IgG purification Rabbit anti-globulin. Antibodies to purified globulin were prepared as described by Walburg and Larkins (1983).
Mouse anti-avenin. Two 30-d-old Swiss mice (Charles River, Wilmington, Mass., USA) were injected intraperitoneally with 100 gg of avenin emulsified in Freund's Complete Adjuvant. The mice received a second injection of 125 gg of avenin emulsified in Ribi Adjuvant (Ribi Immunochem, Hamilton, Mont. USA) at 66 d after the initial injection, followed by a final injection of 330 gg of avenin emulsified in Freund's complete adjuvant at 135 d after the initial injection. Sarcoma 180 cells (No. ATCC CCL-8; American Type Culture Collection, Rockville, Md., USA) were injected intraperitoneally at 155 d after the initial injection (5.106 cells/mouse). A total of 85 ml of ascites fluid was collected over a period of several weeks, and this was stored at 80~ after cellular debris had been removed by centrifugation.
Gel electrophoresis and Western blots Proteins were separated by SDS-PAGE according to Laemmli (1970) in separating gels that contained 12.5% (w/v) polyacrylamide. Total SDS-soluble protein (150 nl of extract) and 0.2 gg each of oat globulin, avenin, and BSA were subjected to SDSPAGE. For Western-blot analysis, proteins were transferred from SDS gels to nitrocellulose filters (Sehleieher and Schuell, Keene, N.H., USA) with the glycine electrode buffer of Towbin et al. (1979) in a BioRad TransBlot apparatus. Electroblotting was performed at 250 mA for 2 h at 4~ C. After transfer, the filters were agitated for 5 min in TTBS (10 mM Tris-HC1, pH 7.4, 140 mM NaC1, 0.15% [v/v] Tween 20), and were then incubated for an additional 5 min in fresh TTBS. The filters were transferred to a solution of either avenin or globulin antiserum that had been diluted 1 : 1000 in TTBS, and were incubated overnight with agitation. After the filters were washed in three changes of TTBS over a period of 30 rain, protein bands reacting with avenin- or globulin-specific antibodies were identified by reaction with either goat anti-mouse IgG conjugated with horseradish peroxidase or goat anti-rabbit IgG conjugated with horseradish peroxidase (BioRad). Bound antibodies were visualized by reaction with 4-chloro-l-naphthol according to the manufacturer's specifications.
Thin sections were immunostained as described by Lending et al. (1988), with modifications to enable dual-labelling of the same section. The primary-antibody incubation was performed with a mixture of mouse anti-avenin serum and rabbit antiglobulin serum which were diluted 1 : 50 and 1 : 100, respectively. Both of the colloidal-gold conjugates were used simultaneously at a 1:10 dilution. Immunocytochemical controls were performed by replacing either of the two primary antibodies with a similar dilution of homologous preimmune serum.
Results Preparation and characterization o f antibodies. W e e x t r a c t e d a v e n i n f r o m o a t seeds a n d p u r i f i e d it f r o m c r o s s - c o n t a m i n a t i n g p r o t e i n s . W e u s e d this p r e p a r a t i o n to p r o d u c e m o u s e a n t i - a v e n i n a n t i b o d i e s ; t h e r a b b i t a n t i b o d i e s d i r e c t e d a g a i n s t the p u r i f i e d g l o b u l i n w e r e p r e v i o u s l y d e s c r i b e d (Walb u r g a n d L a r k i n s 1983). T h e a n t i b o d i e s r e a c t e d specifically w i t h the a n t i g e n to w h i c h t h e y were p r o d u c e d (Fig. 1). A n t i b o d i e s d i r e c t e d a g a i n s t the g l o b u l i n s r e a c t e d specifically w i t h the p o l y p e p t i d e s of Mr ( a p p a r e n t m o l e c u l a r mass) 22000 a n d 36000; no cross-reactivity was observed with the o t h e r p r o t e i n c o m p o n e n t s (Fig. 1B, l a n e s I a n d 2). A n t i b o d i e s a g a i n s t the a v e n i n f r a c t i o n w e r e specific for several p o l y p e p t i d e s o f M r 23 000 t o 29 000 (Fig. 1 C, l a n e s 1 a n d 3).
Endosperm
ultrastructure. The subaleurone ( s t a r c h y e n d o s p e r m ) cells o f o a t seeds 8 D A A c o n t a i n e d n u m e r o u s p r o t e i n b o d i e s w h i c h were l o c a t ed w i t h i n v a c u o l e s (Fig. 2). T h e s e cells also c o n tained m i t o c h o n d r i a , large starch granules, a n d abundant RER. P r o t e i n b o d i e s were f r e q u e n t l y f o u n d to c o n t a i n f o u r d i s t i n c t c o m p o n e n t s , a l t h o u g h the p r o p o r t i o n s o f these w e r e v a r i a b l e . T h e m a j o r c o m p o -
318
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
Fig. 1A-C. Immunospecificity of avenin and globulin antisera. A Total SDS-soluble protein extract (30 gl) from defatted oat flour (lane 1), or 20 !-tg of oat globulin (lane 2), or 40 I-tg of avenin (lane 3) or 7 gg of BSA (lane 4) was separated by SDS-PAGE and stained with Coomassie Blue. B, C Total SDS-soluble protein extract (150 nl) from defatted oat flour (lane I), or 0.2 gg of oat globulin (lane 2), or 0.2 gg of avenin (lane 3) or 0.2 gg of BSA (lane 4) was separated by SDS-PAGE and transferred to nitrocellulose for immunodetection. All proteins except BSA are from Arena sativa cv. Noble. The Western blots were reacted with rabbit anti-globulin (B) or mouse anti-avenin (C) antiserum. Bound antibodies were visualized with either goat anti-rabbit IgG horseradish peroxidase or goat anti-mouse IgG horseradish peroxidase conjugates, respectively
nent of the protein bodies was a dark-staining area of uniform density, with discrete light-staining inclusions distributed throughout (Fig. 2). These light-staining areas tended to be located near the periphery of the dark staining regions, but were sometimes observed in the central area (Fig. 2). The light- and dark-staining regions have been described previously (Saigo et al. 1983). Additionally, there were two minor components which were sometimes observed within the protein bodies. Membranous whorls were frequently found in several closely-spaced layers within the vacuole (Fig. 3). The fourth component was a highly granular region of electron-dense particles similar in
size to ribosomes (Figs. 2, 3). These ribosome-like particles were often tightly clustered; similar particles were sometimes observed dispersed within the dark-staining matrix of the protein bodies (Fig. 3). Protein aggregates within the R E R were much smaller than the vacuolar protein bodies, and often contained a light-staining central region surrounded by a thin layer of darker-staining material (Fig. 4). This dark-staining material within the R E R was often fibrillar. Many reticulate structures were observed in the cytoplasm; they appeared to be membrane-bound, and were often observed within a network of R E R (Fig. 4) or were contiguous with vacuolar membranes (not shown).
Figs. 2-4. Electron micrographs of developing oat endosperm (8 DAA) post-stained with uranyl acetate and lead citrate. Bars = 0.5 gm Fig. 2. Low-magnification field of 8-DAA endosperm of developing oats. Numerous protein bodies within vacuoles (V), with dark- and light-staining areas, corresponding to the globulins (G) and avenins (A), are observed. A dark-staining granular area within a vacuole is marked with an asterisk. Mitochondria (M), starch granules (S), and R E R (arrows) are present. CW, cell wall. x 15 000 Fig. 3. Protein body within a vacuole (V), demonstrating the four components that are usually observed within the vacuole. G, Globulin. A, avenin. Asterisk, ribosome-like granular structure. Arrowhead, ribosome-like granules dispersed within the darkstaining region. Arrow, membranous whorls, x 42000 Fig. 4. Small protein aggregate within RER. The light-staining avenin (A) is surrounded by a dark-staining fibrillar layer of globulin (arrow). A reticulate structure, frequently observed in the cytoplasm, appears to be interconnected with the R E R (arrowhead). M, Mitochondrion. x 46000
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
319
320
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
Fig. $. Protein body in developing oat endosperm stained with rabbit anti-globulin and mouse anti-avenin followed by goat anti-rabbit colloidal gold (5 nm diameter) and goat anti-mouse colloidal gold (10 nm diameter); this staining protocol allows the detection of the globulin and the avenin proteins on the same section (8-DAA endosperm of developing oats, post-stained with uranyl acetate). T h e globulin is located primarily within the dark-staining areas (G), while the avenin is located in the light-staining areas (A). A dark-staining granular region (asterisk) is stained lightly by the anti-globulin antibody, but not by the anti-avenin antibody. Bar = 0.5 Ixm; x 61000
Localization of globulin. The globulin contained within protein bodies was limited to the dark-staining areas. These regions were labelled uniformly with the small gold conjugates when sections were reacted with a mixture of rabbit anti-globulin and
mouse anti-avenin antibodies, and then 5-nm goat anti-rabbit and 10-nm goat anti-mouse/colloidalgold conjugates (Fig. 5). Few of these 5-nm gold particles, which specifically recognize the globulin protein, were observed in the light-staining regions
C.R. Lendinget al. : Immunolocalizationof oat globulin and avenin proteins in protein bodies of the protein bodies (Figs. 5, 6, 8) when sections were immunostained with both antibodies, or when only rabbit anti-globulin followed by 5-nm goat anti-rabbit/colloidal gold was used (Fig. 7). We infrequently observed small protein bodies within R E R which contained only dark-staining regions; these regions were labelled uniformly with the anti-globulin antibody (Fig. 9). The antibody also immunostained amorphous, dark-staining regions within small, dilated RER (Fig. 11). This material often appeared to be aggregating around already formed light-staining deposits (Figs. 11, 12). The aggregates of dark-staining material observed within R E R had indistinct edges (Figs. 9, 11, 12), whereas the edges of dark-staining deposits in vacuolar protein bodies were distinct (Figs. 6-8). Occasional immunostaining was observed on the granular inclusions (Figs. 5 7), although the density of labelling was low. No cytoplasmic staining was observed when sections were immunostained with the anti-globulin antibody. However, this antibody reacted with discrete regions adjacent to the cell walls (Fig. 10). This staining was consistently observed in the region between the plasma membrane and the cell wall, and may have been associated with plasmodesmata. Because the sections treated with antibodies and colloidal gold were post-stained only with uranyl acetate and not lead citrate, it was difficult to determine the underlying structure accurately. Localization of avenin. The deposition of avenin was complementary to that of the globulin. The light-staining regions of protein bodies were labelled with the larger gold conjugates when sections were reacted with a mixture of rabbit antiglobulin and mouse anti-avenin antibodies, and then a mixture of 5-nm goat anti-rabbit/colloidal gold and 10-nm goat anti-mouse/colloidal gold conjugates (Fig. 5). Few of these 10-nm gold particles, which specifically recognize the avenin protein, were observed in the dark-staining regions of the protein bodies (Figs. 5, 6, 8) when sections were immunostained with both antibodies. Small, light-staining aggregates within RER, similar in size to the light-staining regions observed within the vacuolar protein bodies, also immunostained with the anti-avenin antibody (Figs. 11, 12). When avenin antibodies alone were used to immunostain sections, a small amount of anti-avenin staining was observed in the dark-staining regions of the protein bodies (not shown). Thus, some avenin may be interspersed within the dark-staining region which consists predominantly of globulin. Conversely, little immunostaining of globulin proteins was observed within the light-staining re-
322
gions of protein bodies when sections were reacted with anti-globulin antibody alone (Fig. 7). Avenin immunostaining, unlike the immunostaining for the globulin, was never observed adjacent to cell walls. Controls. In all the immunostained preparations, colloidal gold marker was limited to protein aggregates within the R E R and to vacuolar protein bodies, with the exception of the anti-globulin staining observed adjacent to cell walls; background labelling over the cytoplasm, other organelles, and cell walls was extremely low. When preimmune serum was substituted for either the rabbit anti-globulin or the mouse anti-avenin and the sections were incubated with both types of colloidal gold, few 5-nm or 10-nm gold particles were observed, respectively. Discussion
The immunocytochemical staining which we have performed demonstrates that the previously observed dark- and light-staining regions of oat endosperm protein bodies contain the globulins and avenins, respectively. Whereas the morphology of these organelles has been examined in previous studies of developing oat endosperm (Saigo et al. 1983; Bechtel and Pomeranz 1981), the location of globulin and prolamine proteins was not determined. The sites of aggregation of the globulin and avenin proteins appear to be spatially distinct. Avenin proteins seem to aggregate directly within the RER and then migrate to the vacuole; the sizes of the avenin aggregates within both the R E R and vacuolar protein bodies are similar. Although some globulin aggregation occurs within the RER, usually onto avenin deposits, the main aggregation of the globulin appears to take place within the vacuole. Changes in ionic strength and/or pH within the vacuole, or posttranslational modification of the globulin, may be responsible for this phenomenon. Rice is similar to oats in that it contains large amounts of globulin-like proteins (glutelins) (Takaiwa et al. 1986) and small amounts of prolamine proteins; however, rice prolamines aggregate and remain in the R E R (Krishnan et al. 1986). The glutelins aggregate separately from the prolamine component and form protein aggregates within vacuolar membranes after processing through the Golgi (Krishnan et al. 1986). In oats, it appears that the avenins, and to some extent the globulins, aggregate within the RER; however, both proteins subsequently enter the vacuole, where they form protein bodies. The spatial separation of these
322
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
events could explain the two populations of avenins observed by Burgess and Miflin (1985); the avenin aggregates within the R E R may have a different buoyant density than that of vacuolar protein bodies, which consist largely of globulin. Anti-globulin staining often appears to be associated with small, dark patches adjacent to the dilated regions that emerge from plasmodesmata. Because of the low contrast of these sections, we cannot determine with certainty the structure that is immunostained with the anti-globulin antibody. However, this globulin staining is consistently observed, whereas no staining is observed with either preimmune serum or the anti-avenin antibody. Small amounts of globulin may become trapped within this region during the period of rapid cell division which occurs in the transition from the free-nuclear (syncytial) stage to the cellular stage (Saigo et al. 1983). The occurrence of this globulin near the plasmodesmata may explain the results of Langston-Unkefer and Gade (1983), who isolated a globulin from the cell-wall fraction of oat endosperm cells. The membranous structures which are appressed to the protein bodies do not react with either the avenin or globulin antibodies (Fig. 6), while the clustered granular structures are lightly labelled by the anti-globulin antibodies (Figs. 5-7). We do not know the composition or origin of either of these vacuolar components. Although Sai-
go et al. (1983) observed membranous whorls within vacuoles, these vacuoles did not contain any granular material; their micrograph magnifications are too low to allow a direct comparison to our results. The membrane-like region is similar to that observed within protein bodies of developing barley endosperm by Cameron-Mills and von Wettstein (1980). The dispersed, granular inclusions observed within some protein bodies (Fig. 3) are similar in size to that of ribosomes. Perhaps these inclusions are ribosomes from the surface of R E R that are internalized during the formation of protein bodies. Golgi apparati were not observed in the developing oat endosperm we examined. However, this study was performed using methods that retain maximal antigenicity of the storage proteins. There tends to be a compromise between preservation of antigenicity and preservation of structure during fixation for immunocytochemistry. Thus, the ultrastructural preservation in this study may not have been adequate for visualization of Golgi apparati, since these organelles are particularly susceptible to fixation artifacts. Saigo et al. (1983) also observed few Golgi apparati in developing oat endosperm, but noted the presence of direct connections between RER and smooth vacuolar membranes. These authors speculated that direct transport between the R E R and the vacuole was a possible pathway for protein-body formation in oats. A1-
Figs. 6-12. Cells stained with mouse anti-avenin and rabbit anti-globulin followed by goat anti-mouse colloidal gold (10 nm diameter) and goat anti-rabbit coUoidal gold (5 nm diameter); this staining protocol allows the detection of the avenin and the globulin proteins on the same section (8-DAA endosperm of developing oats, post-stained with uranyl acetate). Figure 7 is stained only with rabbit anti-globulin followed by goat anti-rabbit colloidal gold (5 nm diameter). Bars =0.5 gm Fig. 6. The globulin is located primarily within the dark-staining areas (G), while the avenin is located in the light-staining areas (A). No staining by either antibody is observed over the membranous regions of the protein body (arrow). A dark-staining granular region (arrowhead) is stained lightly by the anti-globulin antibody, x 70 000 Fig. 7. Section immunostained with anti-globulin antibody, demonstrating the sparse staining observed for this protein over both the avenin region (A) and the membranous regions (arrow). Light staining is observed over the granular regions (arrowheads). x 72 000 Fig. 8. A dark-staining region (arrow) within the avenin (A) is immunostained by the anti-globulin antibody; globulin (G) also surrounds the avenin inclusion, x 52000 Fig. 9. Globulin is observed directly within a membrane of RER. Small, amorphous aggregates of material immunostain with the anti-globulin antibody and do not have the smooth outlines characteristic of larger globulin aggregates that are observed within vacuoles, x 62000 Fig. 10. Immunolabelling of anti-globulin antibody over discrete patches near the cell wall. These patches seem to overlie a material that stains lightly with uranyl acetate. Staining is often associated with ptasmodesmata (arrow). x 59000 Fig. 11. A deposit of both avenin (A) and globulin (arrow) within a region of RER. The globulin appears to be amorphous during its initial stages of synthesis, forming a more distinct boundary as it aggregates, x 62000 Fig. 12. Small aggregates of both avenin and globulin within the RER. x 62000
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
323
324
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
though we have made similar observations, we can make no conclusive statements regarding involvement of the Golgi apparatus in protein-body formation. Their apparent absence does not rule out possible involvement in transport, particularly of the globulins. The composition of the reticulate structures which are observed in the cytoplasm, often within a network of R E R (Fig. 4), is unknown. The frequent continuity of these structures with both R E R and vacuolar membranes (not shown) indicates a role in the formation of protein bodies, although no immunostaining is observed with either the anti-globulin or anti-avenin antibodies. To conclude, we have demonstrated that globulin and avenin proteins in developing oat endosperm are located within vacuolar protein bodies. Most of the globulin seems to aggregate within the vacuolar protein bodies, whereas the avenins appear to aggregate within the RER. Although Golgi apparati were seldom observed, we cannot rule out their role in the processing of these proteins. We are currently using pulse-chase experiments to label both the avenin and globulin radioactively, to elucidate the transport pathway for these proteins. We would like to acknowledge Dr. Ed Halk (Lubrizol Genetics Corp., Madison, Wis., USA) for preparation of the mouse antiavenin antibodies. We also thank Dr. Mark Shotwell for his helpful suggestions regarding this manuscript. This research was supported by grant No. 87-CRCR-1-2356 from the U.S. Department of Agriculture Competitive Grants Research Organization to BAL. We acknowledge the Electron Microscopy Center at Purdue University for the use of its facilities. This is Journal Paper No. 11842 from the Purdue Agricultural Experiment Station. -
References Bechtel, D.B., Pomeranz, Y. 0981) Ultrastructure and cytochemistry of mature oat (Arena sativa L.) endosperm. The aleurone layer and starchy endosperm. Cereal Chem. 58, 61-69 Brinegar, A.C., Peterson, D.M. (1982) Separation and characterization of oat globulin polypeptides. Arch. Biochem. Biophys. 219, 71-79 Burgess, S.R., Miflin, B.J. (1985) The localization of oat (Arena sativa L.) seed globulins in protein bodies. J. Exp. Bot. 36, 945-954 Cameron-Mills, V., von Wettstein, D. (1980) Protein body formation in the developing barley endosperm. Carlsberg Res. Commun. 45, 577-594 Campbell, W.P., Lee, J.W., O'Brien, T.P., Smart, M.G. (1981) Endosperm morphology and protein body formation in developing wheat grain. Aust. J. Plant Physiol. 8, 5-19
Craig S, Miller, C. (1984) LR White resin and improved on-grid immunogold detection of vicilin, a pea seed storage protein. Cell Biol. Int. Rep. 8, 879-886 Derbyshire, E., Wright, D.J., Boulter, D_ (1976) Legumin and vicilin, storage proteins of legume seeds. Phytochemistry 15, 3-24 Hayat, M.A. (1986) Basic techniques for transmission electron microscopy. Academc Press, Orlando, F1, USA Greenwood, J.S., Chrispeels, M.J. (1985) Immunocytochemical localization of phaseolin and phytohemagglutinin in the endoplasmic reticulum and Golgi complex of developing bean cotyledons, Planta 164, 295-302 Kim, S.I., Charbonnier, L., Mosse, J. (1978) Heterogeneity of avenin, the oat prolamin. Fractionation, molecular weight and amino acid composition. Biochim. Biophys. Acta 537, 22-30 Kim, W.K., Franceschi, V.R., Krishnan, H.B., Okita, T.W. (1988) Formation of wheat protein bodies: involvement of the Golgi apparatus in gliadin transport. Planta 176, 173182 Krishnan, H.B., Franceschi, V.R., Okita, T.W. (1986) Immunochemical studies on the role of the Golgi complex in protein body formation in rice seeds. Planta 169, 471 480 Langston-Unkefer, P.J., Gade, W. (1983) A seed storage protein with possible self-affinity though lectin-like binding. Plant Physiol. 74, 675-680 Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685 Larkins, B.A., Hurkman, W.J. (1978) Synthesis and deposition of zein in protein bodies of maize endosperm. Plant Physiol. 62, 256-263 Luthe, D.S. (1983) Storage protein accumulation in developing rice (Oryza sativa L.) seeds. Plant Sci. Lett. 32, 147-158 Lending, C.R., Kriz, A.L., Larkins, B.A. Bracket, C.E. (1988) Structure and immunocytochemical localization of zeins. Protoplasma 143, 51-62 Parker, M.L., Hawes, C.R. (1982) The Golgi apparatus in developing endosperm of wheat (Triticum aestivum L.). Planta 154, 277-283 Saigo, R.H., Peterson, D.M., Holly, J. (1983) Development of protein bodies in oat starchy endosperm. Can. J. Bot. 61, 1206-1215 Slot, J.W., Geuze, H.J. (1981) Sizing of protein A-colloidal gold probes for immunoelectron microscopy. J. Cell Biol. 90, 533-536 Slot, J.W., Geuze, H.J. (1985) A new method of preparing gold probes for multiple labeling cytochemistry. Eur. J. Cell Biol. 38, 87-93 Takaiwa, F., Kikuchi, S , Oono, K. (1986) The structure of rice storage protein glutelin precursor deduced from cDNA. FEBS 206, 33-36 Taylor, J.R.N., Schussler, L., Liebenberg, N.v.d.W. (1984) Protein body formation in starchy endosperm of developing Sorghum bicolor (L.) Moench seeds. S. Afr. J. Bot. 51, 35M0 Towbin, H., Straehlin, T., Gordon, T. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets. Proc. Natl. Acad. Sci. USA 76, 4350-4354 Walburg, G., Larkins, B.A. (1983) Oat seed globulin. Subunit characterization and demonstration of its synthesis as a precursor. Plant Physiol. 72, 161 165 Received 5 December 1988; accepted 27 February 1989
P l a n t a 9 Springer-Verlag 1989
Immunolocalization of avenin and globulin storage proteins in developing endosperm of Avena sativa L. Craig R. Lending, Ruth S. Chesnut, Katy L. Shaw, and Brian A. Larkins* Department of Botany and Plant Pathology, Purdue University, West Lafayette, IN 47907, USA
Abstract. The seed storage proteins of oats (Arena sativa L.) are synthesized and assembled into vacuolar protein bodies in developing endosperm tissue. We used double-label immunolocatization to study the distribution of these proteins within protein bodies of the starchy endosperm. When sections of developing oat endosperm sampled 8 d after anthesis were stained with uranyl acetate and lead citrate, the vacuolar protein bodies consisted of light-staining regions which were usually surrounded by a darker-staining matrix. Immunogold staining of this tissue demonstrated a distinct segregation of proteins within protein bodies; globulins were localized in the dark-staining regions and prolamines were localized in the light-staining regions. We observed two additional components of vacuolar protein bodies: a membranous component which was often appressed to the outside of the globulin, and a granular, dark-staining region which resembled tightly clustered ribosomes. Neither antibody immunostained the membranous component, but the granular region was lightly labelled with the anti-globulin antibody. Anti-globulin immunostaining was also observed adjacent to cell walls and appeared to be associated with plasmodesmata. Immunostaining for both antigens was also observed within the rough endoplasmic reticulum. Based on the immunostaining patterns, the prolamine proteins appeared to aggregate within the rough endoplasmic reticulum while most of the globulin appeared to aggregate in the vacuole. * To whom correspondence should be addressed Present address: Department of Plant Sciences, University of
Arizona, Tucson, AZ 85721, USA Abbreviations: D A A = d a y s after anthesis; IgG=immunoglobulin G; Mr = apparent molecular mass; RER = rough endoplasmic reticulum; SDS-PAGE=sodium dodecyl sulfate polyacrylamide gel electrophoresis
Key words: Arena (seed storage proteins) - Avenin - Endoplasmic reticulum - Globulin - Protein body Storage protein
Introduction
Developing angiosperm seeds store nitrogen in the form of storage proteins which accumulate in the embryo, the endosperm, or both. In many dicotyledonous plants such as legumes, the principal storage proteins are globulins. These proteins are synthesized predominantly in the cotyledons of the developing embryo (Derbyshire et al. 1976). Globulins are saline-soluble storage proteins that are distinguished on the basis of sedimentation coefficients as 7S or 11S. Both types of globulins are composed of multiple subunits that are synthesized on rough endoplasmic reticulum (RER). Immunolocalization data (Craig and Miller 1984; Greenwood and Chrispeels 1985) support the hypothesis that the products are then transported to vacuoles by way of the Golgi system, where they form insoluble aggregates which have classically been called protein bodies. In contrast to dicotyledons, most cereals have alcohol-soluble storage proteins, called prolamines, which are synthesized in endosperm tissue. Prolamines are synthesized on RER but vary with regard to transport and association into protein bodies. In maize and sorghum (Larkins and Hurkman 1978; Taylor et al. 1984), the prolamine proteins remain within the RER where they aggregate into protein bodies. Wheat and barley prolamines are transported from the RER, presumably by way of the Golgi-system (Kim et al. 1988), and form protein bodies within vacuoles. Protein bodies in
316
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
wheat have also been observed within membranes of RER, but whether or not they remain there is unknown (Campbell et al. 1981; Parker and Hawes 1982). Oat and rice seeds are unusual among cereals in that their endosperms contain relatively small amounts of prolamine proteins. The predominant storage proteins in these cereals are structurally similar to the 11S globulins of dicotyledons; however, these proteins are much less soluble in saline solutions (Brinegar and Peterson 1982; Luthe 1983). Protein bodies that form in rice endosperm have distinctive morphologies; some protein bodies are found within smooth membranes of vacuoles, whereas others form directly within membranes of the RER. Recent histochemical studies have shown that the globulin-like (glutelin) protein of rice (Takaiwa et al. 1986) is found within the vacuole whereas the prolamines aggregate into protein bodies within RER (Krishnan et al. 1986). Previous electron-microscopic studies of developing starchy endosperm of oat have shown that protein bodies are located primarily within vacuolar membranes (Saigo et al. 1983). These protein bodies generally have two regions with different staining properties when prepared for conventional electron microscopy and post-stained with uranyl acetate and lead citrate. The more predominant dark-staining regions have small locules of lightstaining material within their boundaries. Bechtel and Pomeranz (1981) used digestion with various proteases to demonstrate that these two components were different. The light-staining inclusions were assumed by Saigo et al. (1983) to be lipid bodies, although these authors speculated that the inclusions might instead be the prolamine component. Burgess and Miflin (1985) isolated oat endosperm protein bodies by sucrose density gradient ultracentrifugation and determined their polypeptide composition. Two overlapping populations of protein bodies corresponding to globulins and prolamines were observed, although the difference in their densities was not sufficient to allow their resolution. These protein bodies were not examined microscopically to determine their ultrastructural characteristics. To determine the organization and distribution of both the prolamine (avenin) and globulin proteins, we have used double-label immunogold staining with antibodies directed against these proteins to examine their distribution within developing oat starchy endosperm. The globulins occur in the dark-staining regions of the protein bodies, whereas the avenins are found in the light-staining inclusions.
Materials and methods Chemicals and plant material Bovine serum albumin (BSA), Tris (2-amino-2-[hydroxymethyl]-l,3-propanediol), Tween 20 (polyoxyethylenesorbitan monolaurate), Coomassie Brilliant Blue R, chloroauric acid, goat anti-mouse IgG (immunoglobulin G)-horseradish peroxidase, and goat anti-rabbit IgG were from Sigma Chemical Co., St. Louis, Mo., USA. Goat anti-rabbit IgG-horseradish peroxidase was from Bio-Rad, Richmond, Cal., USA, goat antimouse IgG was from Boehringer Mannheim Biochemicals, Indianapolis, Ind., USA, and LR White resin was from Structure Probe, West Chester, Penn., USA. All other chemicals were reagent grade. Oats (Arena sativa L. cv. Noble, a gift of Dr. Gregory Shaner, Department of Botany and Plant Pathology, Purdue University) were cultivated in a greenhouse with supplemental lighting for 16 h per day. The daytime photon lluence rate was 500 g m o l . m - 2 . s -1 PPFD (photosynthetic photon flux density), and the daytime and nighttime temperatures were maintained at 20 25 ~ C and 18-23 ~ C, respectively. The day of anthesis was noted for individual primary florets, and developing seeds were harvested 8 d after anthesis (DAA). The seeds were harvested and immediately fixed for electron microscopy. Fixation and embedment of seeds Oat seeds (caryopses) were prepared for electron microscopy and immunoeytochemistry by fixation in a solution that contained 1% (v/v) glutaraldehyde and 4% (w/v) freshly prepared formaldehyde in PB (phosphate buffer: 50 m M KHzPO4/ Na2HPO4, pH 7.0). The seeds were sliced into 1-mm-thick sections after they were immersed in the fixative; all steps prior to infiltration with plastic were performed on ice or at 4 ~ C. After overnight fixation, the tissue slices were washed three times (5 min each) with PB and post-fixed for 2 h in 2% (w/v) osmium tetroxide in PB. Samples were washed three times with PB and two times with distilled water (5 min each), and were then dehydrated in a graded ethanol series: 15 min each in 10%, 30%, 50%, 70%, 90%, 100% ethanol. After an additional 15 min in fresh 100% ethanol, the samples were infiltrated with 50% LR White resin in ethanol for 15 rain and then placed in 100% LR White resin for 15 min. Fresh resin was added, and the samples were infiltrated overnight; they were polymerized in fresh resin in open molds at 60 ~ C inside a sealed oven purged with nitrogen. After 24 h, samples were removed from the molds, and thin sections were cut with an Ultracut E ultramicrotome (Reichert-Jung, Vienna, Austria). Sections were collected on copper grids coated with formvar and carbon (Hayat 1986) for immunocytochemical staining. Some sections were not immunostained, but were stained with uranyl acetate and lead citrate for examination. Grids were examined with a Philips (Mahwah, N.J., USA) EM-400 electron microscope. Extraction of oat proteins Total seed protein. Oat seeds were ground to a fine flour and stirred in ice-cold acetone (3 ml/g) for 1-2 h at 4 ~ C to remove lipids. Total sodium dodecyl sulfate (SDS)-soluble oat seed protein was obtained by mixing 10 mg of the defatted oat flour in 1 ml of SDS sample buffer (62.5 mM Tris, pit 6.8, 2.3% SDS, 10% glycerol, 5% 2-mercaptoethanol). Insoluble material was pelleted by centrifugation at 14000-g for 1 rain, and the supernatant was used for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
Avenins. Avenins were extracted from defatted oat flour according to the procedure of Kim et al. (1978). Defatted oat flour was stirred in 10 volumes of 52% (v/v) ethanol for 1.5~4 h at room temperature. Insoluble material was removed by centrifugation at 10000-g (ray) for 20 min. Avenins were precipitated from the supernatant by the addition of two volumes of ice-cold 1.5% (w/v) NaC1, collected by centrifugation at 7000 .g for 30 rain, and freeze-dried. Avenin used for antibody production was redissolved in 52% ethanol and dialyzed against distilled water for 16 h at 4~ C, centrifuged at 4500.g for 30 rain, and the pellet was freeze-dried. Globulins. Defatted oat flour was stirred in 10 volumes of 50 mM Tris-HC1 and 1 M NaC1, pH 8.5, (Brinegar and Peterson 1982) for 3.5 h at room temperature to extract oat globulin. Insoluble material was removed by centrifugation at 9000.g for 10 rain. The supernatant was dialyzed against distilled water at 4~ C. Globulins were collected by centrifugation at 9000.g for 25 min, and freeze-dried.
317
A gel similar to those used for Western blots was stained with Coomassie Brilliant Blue R to visualize all proteins present in the extracts. For this gel, 30 gl of total SDS-soluble protein extract, 20 gg of o a t globulin, 40 gg of avenin, and 7 gg of BSA were subjected to SDS-PAGE in a 12.5% acrylamide gel.
Preparation of colloidal-gold conjugate Colloidal-gold conjugates for immunocytochemical stainings were prepared as described by Slot and Geuze (1985). Goat anti-rabbit IgG was complexed to colIoidal gold that had a mean diameter of 5 nm, and goat anti-mouse IgG was complexed to colloidal gold that had a mean diameter of 10 nm. The conjugates were separated from unbound antibodies by centrifugation in a 10%-30% linear glycerol gradient that contained 20 mM Tris-HC1 pH 8.2, 150 mM NaC1, 1% (w/v) BSA, and 20 mM sodium azide (Slot and Geuze 1981). The gold conjugates were stored at 4~ C and were diluted to a pale pink color immediately before use.
Immunoeytochemieal staining Preparation of antisera and IgG purification Rabbit anti-globulin. Antibodies to purified globulin were prepared as described by Walburg and Larkins (1983).
Mouse anti-avenin. Two 30-d-old Swiss mice (Charles River, Wilmington, Mass., USA) were injected intraperitoneally with 100 gg of avenin emulsified in Freund's Complete Adjuvant. The mice received a second injection of 125 gg of avenin emulsified in Ribi Adjuvant (Ribi Immunochem, Hamilton, Mont. USA) at 66 d after the initial injection, followed by a final injection of 330 gg of avenin emulsified in Freund's complete adjuvant at 135 d after the initial injection. Sarcoma 180 cells (No. ATCC CCL-8; American Type Culture Collection, Rockville, Md., USA) were injected intraperitoneally at 155 d after the initial injection (5.106 cells/mouse). A total of 85 ml of ascites fluid was collected over a period of several weeks, and this was stored at 80~ after cellular debris had been removed by centrifugation.
Gel electrophoresis and Western blots Proteins were separated by SDS-PAGE according to Laemmli (1970) in separating gels that contained 12.5% (w/v) polyacrylamide. Total SDS-soluble protein (150 nl of extract) and 0.2 gg each of oat globulin, avenin, and BSA were subjected to SDSPAGE. For Western-blot analysis, proteins were transferred from SDS gels to nitrocellulose filters (Sehleieher and Schuell, Keene, N.H., USA) with the glycine electrode buffer of Towbin et al. (1979) in a BioRad TransBlot apparatus. Electroblotting was performed at 250 mA for 2 h at 4~ C. After transfer, the filters were agitated for 5 min in TTBS (10 mM Tris-HC1, pH 7.4, 140 mM NaC1, 0.15% [v/v] Tween 20), and were then incubated for an additional 5 min in fresh TTBS. The filters were transferred to a solution of either avenin or globulin antiserum that had been diluted 1 : 1000 in TTBS, and were incubated overnight with agitation. After the filters were washed in three changes of TTBS over a period of 30 rain, protein bands reacting with avenin- or globulin-specific antibodies were identified by reaction with either goat anti-mouse IgG conjugated with horseradish peroxidase or goat anti-rabbit IgG conjugated with horseradish peroxidase (BioRad). Bound antibodies were visualized by reaction with 4-chloro-l-naphthol according to the manufacturer's specifications.
Thin sections were immunostained as described by Lending et al. (1988), with modifications to enable dual-labelling of the same section. The primary-antibody incubation was performed with a mixture of mouse anti-avenin serum and rabbit antiglobulin serum which were diluted 1 : 50 and 1 : 100, respectively. Both of the colloidal-gold conjugates were used simultaneously at a 1:10 dilution. Immunocytochemical controls were performed by replacing either of the two primary antibodies with a similar dilution of homologous preimmune serum.
Results Preparation and characterization o f antibodies. W e e x t r a c t e d a v e n i n f r o m o a t seeds a n d p u r i f i e d it f r o m c r o s s - c o n t a m i n a t i n g p r o t e i n s . W e u s e d this p r e p a r a t i o n to p r o d u c e m o u s e a n t i - a v e n i n a n t i b o d i e s ; t h e r a b b i t a n t i b o d i e s d i r e c t e d a g a i n s t the p u r i f i e d g l o b u l i n w e r e p r e v i o u s l y d e s c r i b e d (Walb u r g a n d L a r k i n s 1983). T h e a n t i b o d i e s r e a c t e d specifically w i t h the a n t i g e n to w h i c h t h e y were p r o d u c e d (Fig. 1). A n t i b o d i e s d i r e c t e d a g a i n s t the g l o b u l i n s r e a c t e d specifically w i t h the p o l y p e p t i d e s of Mr ( a p p a r e n t m o l e c u l a r mass) 22000 a n d 36000; no cross-reactivity was observed with the o t h e r p r o t e i n c o m p o n e n t s (Fig. 1B, l a n e s I a n d 2). A n t i b o d i e s a g a i n s t the a v e n i n f r a c t i o n w e r e specific for several p o l y p e p t i d e s o f M r 23 000 t o 29 000 (Fig. 1 C, l a n e s 1 a n d 3).
Endosperm
ultrastructure. The subaleurone ( s t a r c h y e n d o s p e r m ) cells o f o a t seeds 8 D A A c o n t a i n e d n u m e r o u s p r o t e i n b o d i e s w h i c h were l o c a t ed w i t h i n v a c u o l e s (Fig. 2). T h e s e cells also c o n tained m i t o c h o n d r i a , large starch granules, a n d abundant RER. P r o t e i n b o d i e s were f r e q u e n t l y f o u n d to c o n t a i n f o u r d i s t i n c t c o m p o n e n t s , a l t h o u g h the p r o p o r t i o n s o f these w e r e v a r i a b l e . T h e m a j o r c o m p o -
318
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
Fig. 1A-C. Immunospecificity of avenin and globulin antisera. A Total SDS-soluble protein extract (30 gl) from defatted oat flour (lane 1), or 20 !-tg of oat globulin (lane 2), or 40 I-tg of avenin (lane 3) or 7 gg of BSA (lane 4) was separated by SDS-PAGE and stained with Coomassie Blue. B, C Total SDS-soluble protein extract (150 nl) from defatted oat flour (lane I), or 0.2 gg of oat globulin (lane 2), or 0.2 gg of avenin (lane 3) or 0.2 gg of BSA (lane 4) was separated by SDS-PAGE and transferred to nitrocellulose for immunodetection. All proteins except BSA are from Arena sativa cv. Noble. The Western blots were reacted with rabbit anti-globulin (B) or mouse anti-avenin (C) antiserum. Bound antibodies were visualized with either goat anti-rabbit IgG horseradish peroxidase or goat anti-mouse IgG horseradish peroxidase conjugates, respectively
nent of the protein bodies was a dark-staining area of uniform density, with discrete light-staining inclusions distributed throughout (Fig. 2). These light-staining areas tended to be located near the periphery of the dark staining regions, but were sometimes observed in the central area (Fig. 2). The light- and dark-staining regions have been described previously (Saigo et al. 1983). Additionally, there were two minor components which were sometimes observed within the protein bodies. Membranous whorls were frequently found in several closely-spaced layers within the vacuole (Fig. 3). The fourth component was a highly granular region of electron-dense particles similar in
size to ribosomes (Figs. 2, 3). These ribosome-like particles were often tightly clustered; similar particles were sometimes observed dispersed within the dark-staining matrix of the protein bodies (Fig. 3). Protein aggregates within the R E R were much smaller than the vacuolar protein bodies, and often contained a light-staining central region surrounded by a thin layer of darker-staining material (Fig. 4). This dark-staining material within the R E R was often fibrillar. Many reticulate structures were observed in the cytoplasm; they appeared to be membrane-bound, and were often observed within a network of R E R (Fig. 4) or were contiguous with vacuolar membranes (not shown).
Figs. 2-4. Electron micrographs of developing oat endosperm (8 DAA) post-stained with uranyl acetate and lead citrate. Bars = 0.5 gm Fig. 2. Low-magnification field of 8-DAA endosperm of developing oats. Numerous protein bodies within vacuoles (V), with dark- and light-staining areas, corresponding to the globulins (G) and avenins (A), are observed. A dark-staining granular area within a vacuole is marked with an asterisk. Mitochondria (M), starch granules (S), and R E R (arrows) are present. CW, cell wall. x 15 000 Fig. 3. Protein body within a vacuole (V), demonstrating the four components that are usually observed within the vacuole. G, Globulin. A, avenin. Asterisk, ribosome-like granular structure. Arrowhead, ribosome-like granules dispersed within the darkstaining region. Arrow, membranous whorls, x 42000 Fig. 4. Small protein aggregate within RER. The light-staining avenin (A) is surrounded by a dark-staining fibrillar layer of globulin (arrow). A reticulate structure, frequently observed in the cytoplasm, appears to be interconnected with the R E R (arrowhead). M, Mitochondrion. x 46000
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
319
320
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
Fig. $. Protein body in developing oat endosperm stained with rabbit anti-globulin and mouse anti-avenin followed by goat anti-rabbit colloidal gold (5 nm diameter) and goat anti-mouse colloidal gold (10 nm diameter); this staining protocol allows the detection of the globulin and the avenin proteins on the same section (8-DAA endosperm of developing oats, post-stained with uranyl acetate). T h e globulin is located primarily within the dark-staining areas (G), while the avenin is located in the light-staining areas (A). A dark-staining granular region (asterisk) is stained lightly by the anti-globulin antibody, but not by the anti-avenin antibody. Bar = 0.5 Ixm; x 61000
Localization of globulin. The globulin contained within protein bodies was limited to the dark-staining areas. These regions were labelled uniformly with the small gold conjugates when sections were reacted with a mixture of rabbit anti-globulin and
mouse anti-avenin antibodies, and then 5-nm goat anti-rabbit and 10-nm goat anti-mouse/colloidalgold conjugates (Fig. 5). Few of these 5-nm gold particles, which specifically recognize the globulin protein, were observed in the light-staining regions
C.R. Lendinget al. : Immunolocalizationof oat globulin and avenin proteins in protein bodies of the protein bodies (Figs. 5, 6, 8) when sections were immunostained with both antibodies, or when only rabbit anti-globulin followed by 5-nm goat anti-rabbit/colloidal gold was used (Fig. 7). We infrequently observed small protein bodies within R E R which contained only dark-staining regions; these regions were labelled uniformly with the anti-globulin antibody (Fig. 9). The antibody also immunostained amorphous, dark-staining regions within small, dilated RER (Fig. 11). This material often appeared to be aggregating around already formed light-staining deposits (Figs. 11, 12). The aggregates of dark-staining material observed within R E R had indistinct edges (Figs. 9, 11, 12), whereas the edges of dark-staining deposits in vacuolar protein bodies were distinct (Figs. 6-8). Occasional immunostaining was observed on the granular inclusions (Figs. 5 7), although the density of labelling was low. No cytoplasmic staining was observed when sections were immunostained with the anti-globulin antibody. However, this antibody reacted with discrete regions adjacent to the cell walls (Fig. 10). This staining was consistently observed in the region between the plasma membrane and the cell wall, and may have been associated with plasmodesmata. Because the sections treated with antibodies and colloidal gold were post-stained only with uranyl acetate and not lead citrate, it was difficult to determine the underlying structure accurately. Localization of avenin. The deposition of avenin was complementary to that of the globulin. The light-staining regions of protein bodies were labelled with the larger gold conjugates when sections were reacted with a mixture of rabbit antiglobulin and mouse anti-avenin antibodies, and then a mixture of 5-nm goat anti-rabbit/colloidal gold and 10-nm goat anti-mouse/colloidal gold conjugates (Fig. 5). Few of these 10-nm gold particles, which specifically recognize the avenin protein, were observed in the dark-staining regions of the protein bodies (Figs. 5, 6, 8) when sections were immunostained with both antibodies. Small, light-staining aggregates within RER, similar in size to the light-staining regions observed within the vacuolar protein bodies, also immunostained with the anti-avenin antibody (Figs. 11, 12). When avenin antibodies alone were used to immunostain sections, a small amount of anti-avenin staining was observed in the dark-staining regions of the protein bodies (not shown). Thus, some avenin may be interspersed within the dark-staining region which consists predominantly of globulin. Conversely, little immunostaining of globulin proteins was observed within the light-staining re-
322
gions of protein bodies when sections were reacted with anti-globulin antibody alone (Fig. 7). Avenin immunostaining, unlike the immunostaining for the globulin, was never observed adjacent to cell walls. Controls. In all the immunostained preparations, colloidal gold marker was limited to protein aggregates within the R E R and to vacuolar protein bodies, with the exception of the anti-globulin staining observed adjacent to cell walls; background labelling over the cytoplasm, other organelles, and cell walls was extremely low. When preimmune serum was substituted for either the rabbit anti-globulin or the mouse anti-avenin and the sections were incubated with both types of colloidal gold, few 5-nm or 10-nm gold particles were observed, respectively. Discussion
The immunocytochemical staining which we have performed demonstrates that the previously observed dark- and light-staining regions of oat endosperm protein bodies contain the globulins and avenins, respectively. Whereas the morphology of these organelles has been examined in previous studies of developing oat endosperm (Saigo et al. 1983; Bechtel and Pomeranz 1981), the location of globulin and prolamine proteins was not determined. The sites of aggregation of the globulin and avenin proteins appear to be spatially distinct. Avenin proteins seem to aggregate directly within the RER and then migrate to the vacuole; the sizes of the avenin aggregates within both the R E R and vacuolar protein bodies are similar. Although some globulin aggregation occurs within the RER, usually onto avenin deposits, the main aggregation of the globulin appears to take place within the vacuole. Changes in ionic strength and/or pH within the vacuole, or posttranslational modification of the globulin, may be responsible for this phenomenon. Rice is similar to oats in that it contains large amounts of globulin-like proteins (glutelins) (Takaiwa et al. 1986) and small amounts of prolamine proteins; however, rice prolamines aggregate and remain in the R E R (Krishnan et al. 1986). The glutelins aggregate separately from the prolamine component and form protein aggregates within vacuolar membranes after processing through the Golgi (Krishnan et al. 1986). In oats, it appears that the avenins, and to some extent the globulins, aggregate within the RER; however, both proteins subsequently enter the vacuole, where they form protein bodies. The spatial separation of these
322
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
events could explain the two populations of avenins observed by Burgess and Miflin (1985); the avenin aggregates within the R E R may have a different buoyant density than that of vacuolar protein bodies, which consist largely of globulin. Anti-globulin staining often appears to be associated with small, dark patches adjacent to the dilated regions that emerge from plasmodesmata. Because of the low contrast of these sections, we cannot determine with certainty the structure that is immunostained with the anti-globulin antibody. However, this globulin staining is consistently observed, whereas no staining is observed with either preimmune serum or the anti-avenin antibody. Small amounts of globulin may become trapped within this region during the period of rapid cell division which occurs in the transition from the free-nuclear (syncytial) stage to the cellular stage (Saigo et al. 1983). The occurrence of this globulin near the plasmodesmata may explain the results of Langston-Unkefer and Gade (1983), who isolated a globulin from the cell-wall fraction of oat endosperm cells. The membranous structures which are appressed to the protein bodies do not react with either the avenin or globulin antibodies (Fig. 6), while the clustered granular structures are lightly labelled by the anti-globulin antibodies (Figs. 5-7). We do not know the composition or origin of either of these vacuolar components. Although Sai-
go et al. (1983) observed membranous whorls within vacuoles, these vacuoles did not contain any granular material; their micrograph magnifications are too low to allow a direct comparison to our results. The membrane-like region is similar to that observed within protein bodies of developing barley endosperm by Cameron-Mills and von Wettstein (1980). The dispersed, granular inclusions observed within some protein bodies (Fig. 3) are similar in size to that of ribosomes. Perhaps these inclusions are ribosomes from the surface of R E R that are internalized during the formation of protein bodies. Golgi apparati were not observed in the developing oat endosperm we examined. However, this study was performed using methods that retain maximal antigenicity of the storage proteins. There tends to be a compromise between preservation of antigenicity and preservation of structure during fixation for immunocytochemistry. Thus, the ultrastructural preservation in this study may not have been adequate for visualization of Golgi apparati, since these organelles are particularly susceptible to fixation artifacts. Saigo et al. (1983) also observed few Golgi apparati in developing oat endosperm, but noted the presence of direct connections between RER and smooth vacuolar membranes. These authors speculated that direct transport between the R E R and the vacuole was a possible pathway for protein-body formation in oats. A1-
Figs. 6-12. Cells stained with mouse anti-avenin and rabbit anti-globulin followed by goat anti-mouse colloidal gold (10 nm diameter) and goat anti-rabbit coUoidal gold (5 nm diameter); this staining protocol allows the detection of the avenin and the globulin proteins on the same section (8-DAA endosperm of developing oats, post-stained with uranyl acetate). Figure 7 is stained only with rabbit anti-globulin followed by goat anti-rabbit colloidal gold (5 nm diameter). Bars =0.5 gm Fig. 6. The globulin is located primarily within the dark-staining areas (G), while the avenin is located in the light-staining areas (A). No staining by either antibody is observed over the membranous regions of the protein body (arrow). A dark-staining granular region (arrowhead) is stained lightly by the anti-globulin antibody, x 70 000 Fig. 7. Section immunostained with anti-globulin antibody, demonstrating the sparse staining observed for this protein over both the avenin region (A) and the membranous regions (arrow). Light staining is observed over the granular regions (arrowheads). x 72 000 Fig. 8. A dark-staining region (arrow) within the avenin (A) is immunostained by the anti-globulin antibody; globulin (G) also surrounds the avenin inclusion, x 52000 Fig. 9. Globulin is observed directly within a membrane of RER. Small, amorphous aggregates of material immunostain with the anti-globulin antibody and do not have the smooth outlines characteristic of larger globulin aggregates that are observed within vacuoles, x 62000 Fig. 10. Immunolabelling of anti-globulin antibody over discrete patches near the cell wall. These patches seem to overlie a material that stains lightly with uranyl acetate. Staining is often associated with ptasmodesmata (arrow). x 59000 Fig. 11. A deposit of both avenin (A) and globulin (arrow) within a region of RER. The globulin appears to be amorphous during its initial stages of synthesis, forming a more distinct boundary as it aggregates, x 62000 Fig. 12. Small aggregates of both avenin and globulin within the RER. x 62000
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
323
324
C.R. Lending et al. : Immunolocalization of oat globulin and avenin proteins in protein bodies
though we have made similar observations, we can make no conclusive statements regarding involvement of the Golgi apparatus in protein-body formation. Their apparent absence does not rule out possible involvement in transport, particularly of the globulins. The composition of the reticulate structures which are observed in the cytoplasm, often within a network of R E R (Fig. 4), is unknown. The frequent continuity of these structures with both R E R and vacuolar membranes (not shown) indicates a role in the formation of protein bodies, although no immunostaining is observed with either the anti-globulin or anti-avenin antibodies. To conclude, we have demonstrated that globulin and avenin proteins in developing oat endosperm are located within vacuolar protein bodies. Most of the globulin seems to aggregate within the vacuolar protein bodies, whereas the avenins appear to aggregate within the RER. Although Golgi apparati were seldom observed, we cannot rule out their role in the processing of these proteins. We are currently using pulse-chase experiments to label both the avenin and globulin radioactively, to elucidate the transport pathway for these proteins. We would like to acknowledge Dr. Ed Halk (Lubrizol Genetics Corp., Madison, Wis., USA) for preparation of the mouse antiavenin antibodies. We also thank Dr. Mark Shotwell for his helpful suggestions regarding this manuscript. This research was supported by grant No. 87-CRCR-1-2356 from the U.S. Department of Agriculture Competitive Grants Research Organization to BAL. We acknowledge the Electron Microscopy Center at Purdue University for the use of its facilities. This is Journal Paper No. 11842 from the Purdue Agricultural Experiment Station. -
References Bechtel, D.B., Pomeranz, Y. 0981) Ultrastructure and cytochemistry of mature oat (Arena sativa L.) endosperm. The aleurone layer and starchy endosperm. Cereal Chem. 58, 61-69 Brinegar, A.C., Peterson, D.M. (1982) Separation and characterization of oat globulin polypeptides. Arch. Biochem. Biophys. 219, 71-79 Burgess, S.R., Miflin, B.J. (1985) The localization of oat (Arena sativa L.) seed globulins in protein bodies. J. Exp. Bot. 36, 945-954 Cameron-Mills, V., von Wettstein, D. (1980) Protein body formation in the developing barley endosperm. Carlsberg Res. Commun. 45, 577-594 Campbell, W.P., Lee, J.W., O'Brien, T.P., Smart, M.G. (1981) Endosperm morphology and protein body formation in developing wheat grain. Aust. J. Plant Physiol. 8, 5-19
Craig S, Miller, C. (1984) LR White resin and improved on-grid immunogold detection of vicilin, a pea seed storage protein. Cell Biol. Int. Rep. 8, 879-886 Derbyshire, E., Wright, D.J., Boulter, D_ (1976) Legumin and vicilin, storage proteins of legume seeds. Phytochemistry 15, 3-24 Hayat, M.A. (1986) Basic techniques for transmission electron microscopy. Academc Press, Orlando, F1, USA Greenwood, J.S., Chrispeels, M.J. (1985) Immunocytochemical localization of phaseolin and phytohemagglutinin in the endoplasmic reticulum and Golgi complex of developing bean cotyledons, Planta 164, 295-302 Kim, S.I., Charbonnier, L., Mosse, J. (1978) Heterogeneity of avenin, the oat prolamin. Fractionation, molecular weight and amino acid composition. Biochim. Biophys. Acta 537, 22-30 Kim, W.K., Franceschi, V.R., Krishnan, H.B., Okita, T.W. (1988) Formation of wheat protein bodies: involvement of the Golgi apparatus in gliadin transport. Planta 176, 173182 Krishnan, H.B., Franceschi, V.R., Okita, T.W. (1986) Immunochemical studies on the role of the Golgi complex in protein body formation in rice seeds. Planta 169, 471 480 Langston-Unkefer, P.J., Gade, W. (1983) A seed storage protein with possible self-affinity though lectin-like binding. Plant Physiol. 74, 675-680 Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685 Larkins, B.A., Hurkman, W.J. (1978) Synthesis and deposition of zein in protein bodies of maize endosperm. Plant Physiol. 62, 256-263 Luthe, D.S. (1983) Storage protein accumulation in developing rice (Oryza sativa L.) seeds. Plant Sci. Lett. 32, 147-158 Lending, C.R., Kriz, A.L., Larkins, B.A. Bracket, C.E. (1988) Structure and immunocytochemical localization of zeins. Protoplasma 143, 51-62 Parker, M.L., Hawes, C.R. (1982) The Golgi apparatus in developing endosperm of wheat (Triticum aestivum L.). Planta 154, 277-283 Saigo, R.H., Peterson, D.M., Holly, J. (1983) Development of protein bodies in oat starchy endosperm. Can. J. Bot. 61, 1206-1215 Slot, J.W., Geuze, H.J. (1981) Sizing of protein A-colloidal gold probes for immunoelectron microscopy. J. Cell Biol. 90, 533-536 Slot, J.W., Geuze, H.J. (1985) A new method of preparing gold probes for multiple labeling cytochemistry. Eur. J. Cell Biol. 38, 87-93 Takaiwa, F., Kikuchi, S , Oono, K. (1986) The structure of rice storage protein glutelin precursor deduced from cDNA. FEBS 206, 33-36 Taylor, J.R.N., Schussler, L., Liebenberg, N.v.d.W. (1984) Protein body formation in starchy endosperm of developing Sorghum bicolor (L.) Moench seeds. S. Afr. J. Bot. 51, 35M0 Towbin, H., Straehlin, T., Gordon, T. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets. Proc. Natl. Acad. Sci. USA 76, 4350-4354 Walburg, G., Larkins, B.A. (1983) Oat seed globulin. Subunit characterization and demonstration of its synthesis as a precursor. Plant Physiol. 72, 161 165 Received 5 December 1988; accepted 27 February 1989
