Introduction

Calmodulin (CAM) is a heat-stable, acidic and multifunctional calcium-binding protein which modulates a number of enzymic activities and many cellular processes in a calcium-dependent manner. In plants, CaM has been reported to regulate NAD kinase, quinate oxidoreductase, Ca2+-ATPase, protein kinases and H+-ATPase1. In addition, plant CaM may be involved in gravitropic responses, phytochrome induced responses, and hormone-mediated processes2, 3.

The studies of the cellular and subcellular distribution of CaM in situ are essential for understanding the function of CaM in cells. Precise knowledge of the subcellular localization of CaM may provide some clue to its possible function in the biochemical mechanisms where its actions have been demonstrated in vitro. Although many biochemical studies of plant CaM have been done, relatively little is known about its subcellular location in plant cells. In this paper, the experimental results of the subcellular distribution of CaM in corn root cells by immunoelectron microscopy are reported.

Materials and methods

Chemicals

Peroxidase-labeled goat anti-rabbit IgG was purchased from Bio-Rad . 3,3′-diamino-benzidine-4HCl (DAB) from Sigma, paraformaldehyde and glutaraldehyde from Polysciences, sodium metaperiodate from Fisher. Protein A-gold conjugates (15 nm) was purchased from Institute of Medical Sciences, Chinese Academy of Military Medical Sciences(Beijing). All other chemicals were of reagent grade.

Plant material

Corn seeds (Zea mays L.) were surface-sterilized in 1 % sodium hypochlorite for 30 min and then rinsed at least 3 times with sterile distilled water. After soaked in the water for 24 h, seeds were placed on moist filter paper in a covered dish and germinated for 48 h in the dark at 20 °C. 1.5 mm terminals of corn roots were excised for immunoelectron microscopic observation.

Preparation and characterization of anti-CaM antibody

CaM was purified to homogeneity from wheat coleoptiles by the method of Biro et al4. Anti-CaM serum was raised in rabbits. Briefly, 0.5 ml wheat CaM (0.8 mg) in PBS was emulsified with equal volume of either complete Freund's, adjuvant (initial injection) or incomplete Freund's adjuvant (subsequent injections). Male New Zealand white rabbits were injected subcutaneously into 6-8 sites on the back. The boost injections were repeated at 2 wk intervals. Ten days after the first boost, the serum was obtained and tested for immunoreaetivity with CaM by ELISA. A series of boosts and tests was performed until a sufficient titer was achieved.

Specificity of the antiserum toward CaM was tested by immunostaining of Western blots of corn root proteins. Briefly, corn root proteins were separated on a 15% SDSpolyacrylamide gel using the procedure of Laemmli5, and then electrophoretieally transferred to nitrocellulose membrane (0.45 μm pore size) by the method of Towbin et al6 at 200 mA for 45 min at 4°C. Amido black staining of the control strip revealed the band pattern of the transferred proteins. The blotted nitrocellulose membrane was blocked with 2% (W /V) nonfat dry milk in PBS (pH 7.4) for 2 h as described in Johnson et al.7, and then incubated with anti-CaM serum in PBS containing 0.05% Tween-20 (PBST) for 3 h. After being washed in PBST , the membrane was incubated with peroxidase-labeled goat anti-rabbit IgG in PBS for 1 h. After washing, the immunoblots were visualized in a solution containing 0.05% (W/ V) DAB and 0.01% H20 2 in 0.05 M Tris-HCl buffer (pH 7. 6)

Immunoperoxidase labeling

Corn roots were fixed in 4% paraformaldehyde and 0.1% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) for 6 h at 4°C. Longitudinal slices (30 μm) of the roots were prepared with a vibratome. The immunostaining procedures employed here were the same as those described by Lin et al.8. After immunostaining, the slices were post fixed in 1% OsO4 in 0.1 M phosphate buffer (pH 7.4) for 1 h at room temperature, dehydrated through ethanol series and propylene oxide, and then fiat-embedded in Epon 812. The positive immunoreaction areas were marked under a light microscope and mounted on blank blocks. Superficial ultrathin(700 Å) sections, cut Parallel to the tissue surface originally exposed to the immunological reagents, were examined in a Hitachi H-600 electron microscope without furtherstaining. Control experiments included: a) the anti-CaM serum was omitted; b) use of preimmune rabbit serum instead of anti-CaM serum; c) anti-CaM serum absorbed with excess of purified wheat CaM.

Immunogold labeling

Corn roots were fixed in 4% paraformaldehyde and 0.5% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) for 6 h at 4°C. Longitudinal slices (60 μm) of the roots were prepared with a vibratome . The slices were postfixed, dehydrated, and fiat-embedded as described above. Thin sections (700 Å) were cut with glass knives and mounted on nickel grids (300 mesh). All the following operations were performed at room temperature, unless otherwise stated. The sections were immersed in distilled water for 10 min and then etched with a saturated aqueous solution of sodium meta periodate for 30 min to expose the antigenic sites9. The sections were then washed in distilled water 3 times and treated with 20 mM glycine in PBS for 15 min to inactivate the nonreacted aldehyde groups10.

The sections were washed 3 times with PBS and then blocked in 1% nonfat dry milk in PBS for 1 h. The sections were incubated in antiCaM serum diluted 1:30 with PBS for 24 h at 4°C, washed 4 times with PBS, and incubated with protein A-gold conjugates (15 nm) diluted 1:30 in PBS for 1 h. After the sections were washed in PBS 3 times and twice in distilled water, they were dried and poststained with saturated uranyl acetate in 50% (V / V) ethanol (2 min) and lead citrate (30 s). The sections were examined and photographed with a Hitachi H-600 electron microscope. Three control experiments were also performed as described in immunoperoxidase labeling.

Quantitation of gold labeling

The density of labeling of each cell compartment was obtained by determining the number of gold particles per μm2 of the cellular compartment. 20 micrographs were analyzed11 at a final magnification of 30,000 ×.

Results

The specificity of the antibody probe is most crucial for immunocytochemical localization. We analyzed the specificity of anti-wheat CaM serum by Western blots. As shown in Fig 1, anti-wheat CaM serum reacted only with one protein band identified as Ca M (with an electrophoretic mobility in the range of corn CaM) in the corn root proteins (Lane 5) and with purified corn CaM (Lane 4). No cross-reaction with other polypeptides was observed. In addition, anti-CaM serum absorbed with purified wheat CaM could not react with any protein from the corn roots on Western blots (data not shown). These results implied that CaM in corn root cells can be specifically detected by the anti-wheat CaM serum.

Fig 1
figure 1

The specificity of anti-CaM antibody to CaM from corn roots. Protein samples were subjected to SDS-PAGE and transferred to nitrocellulose (Western blo t). The Western blots were either stained with 0. 1 % Amido black (lanes 1-3) or immunostained with anti-CaM antibody (lanes 4 and 5). Lane 1, Molecular-weight markers (Values in kD); lanes 2 and 4 purified corn CaM (5 μg); lanes 3 and 5, corn roots proteins (100 μg) . In the lane 5. the only band that was immunostained by anti-CaM antib od y was one with the same mobility as purified corn CaM.

Full size image

Ultrathin sections of Epon 812-embedded corn roots were used to localize CaM by protein A-immunogold technique. Good ultrastructural preservation with the retention of antigenic determinants was obtained by the methods used. In the meristematic cells, gold particles were associated with the plasma membrane, cytoplasmic face of RER, polyribosomes, mitochondria as well as with the cell wall (Fig 2 and 3). Characteristically, the starch grains of the amyloplasts was densely labeled in the root-cap cells (Fig 4). Little gold labeling could be observed in the control (Fig 5).

Fig 2
figure 2

Meristematic cell of corn root. Gold particles are localized on the plasma mem-branes (PM), polyribosomes (arrow), and cytoplasmic, face of rough endoplasmic reticulum (RER). Some gold particles are present in the cell wall (C W) . × 30,000.

Full size image
Fig 3
figure 3

Meristematic cell of corn root. Gold particles are localized on the cytoplasmic face of RER. mitochondrion (M), and polyribosomes (arrow). × 30.000.

Full size image
Fig 4
figure 4

Corn root-cap cell. The starch grain (S) in the amyloplast is densely labeled with gold particles. × 35,000.

Full size image
Fig 5
figure 5

Control. Corn root-cap cell. No labeling is seen in the cell treated with CaM-absorbed antiserum. × 25,000.

Full size image

The distribution of CaM in corn root cells was also investigated by immunoperoxidase technique. In the meristematic cells, the peroxidase reaction deposits were associated with the inner surface of plasma membrane and possibly with plasmodesmata (Fig 6). The deposits were also observed on the cytoplasmic face of ER in these cells (data not shown). No peroxidase reaction deposits could be seen in the control (Fig 7).

Fig 6
figure 6

Meristematic cell of corn root. Peroxidase reaction deposits are shown to be attached to the inner surface of plasma membranes(PM). It seems reaction deposits are also on the Plasmodesma (arrow). ×20,000.

Full size image
Fig 7
figure 7

Control. Meristematic cell of corn root. No specific reaction deposits are seen in the cell treated with preimmune rabbit serum, ×25,000.

Full size image

Since the immunoreactive sites for CaM were widely distributed in corn root cells, we have tried to evaluate the relative amount of CaM in different cellular compartments by calculating the labeling density of some cell compartments. As shown in Tab 1, the labeling density of amyloplasts was the highest among those of all cellular compartments tested in the root-cap cells. The density of labeling over the cell wall was the lowest among those cellular compartments tested in both the root-cap cells and the meristematic cells. However, since the background labeling obtained with CaM-absorbed antiserum was subtracted from the labeling obtained with anti-CaM serum, the cell wall labeling was very likely to represent bona fide CaM antigenic sites.

Tab 1 Density of labeling over various cell compartments (n=20)
Full size table

Discussion

Using immumoelectron microscopy, we have determined the subcellular distribution of CaM in corn root cells. Specific immunolabeling for CaM with good ultrastructural preservation was obtained with Epon 812-embedded corn root ceils. Similarly, Hatase et al12 also showed that there was specific immunogold labeling for CaM in Epon 812-embedded mitochondria isolated from rat liver. These results indicated that CaM antigenicity in Epon 812-embedded tissues could be retained. Western blot analysis showed that the anti-wheat CaM serum we raised did not recognize proteins other than CaM in corn roots. This result indicated the specificity of anti-wheat CaM serum towards CaM. Moreover, control experiments demonstrated the specificity of immunogold or immunoperoxidase labeling of CaM in corn root cells. Thus, the immunocytochemical methods employed here rendered us confident in the reliability of CaM antigenic site distribution.

CaM was shown in situ to be present in nuclei, cytoplasm, mitochondria, and cell wall(Tab 1). These results were in accord with other reports of CaM distribution by RIA in isolated subcellular fractions from several plants4, 13, 14. In addition, CaM was also precisely localized on the plasma membrane, cytoplasmic face of RER and poly-ribosomes of the meristematic cells. These findings are in agreement with the CaM localization reported in animal cells15, 16. Using biochemical methods, Collinge and Trewavas 17 demonstrated that CaM was associated with pea plasma membrane preparations and that CaM may be projected from the inner surface of the membrane. Our immunoelectron microscopic studies (Fig 2 and 6) confirmed their findings. Membrane-bound CaM may play a role in controlling the intracellular calcium level through stimulating membrane Ca2+-ATPase activity1.

There are plenty of evidences that calcium plays an important role in the gravitropic response of roots and shoots2, 18. And also there is some clue that CaM is implicated in the gravitropic mechanism19. In the immunocytochemical localization studies of CaM, Lin and Sun20 found an intense staining in the root cap of corn, and Dauwalder et al.21, observed a strong staining of amyloplasts in the columella of pea rootcap. Our immunogold labeling results showed that the starch grains in amyloplasts of root-cap cells were densely labeled (Fig 4). Generally speaking, the amyloplasts of root-cap cells are thought to be the gravity-sensing organelle in the gravitropic response of roots22. Moreover, CaM could regulate the phosphorylation of several proteins in amyloplasts23. Taking these data together, it is quite possible that CaM associated with amyloplasts in the root-cap cells plays a role in the sensing of plant roots to gravitropic stimuli.

The cell wall of rootcap cells and meristematic cells was lightly labeled for CaM(Tab 1). This result was consistent with the findings using ultralcentrifugation and RIA4, 14. That Dauwalder et al.21 could not observe the staining of cell wall may be due to the very limited resolution of the immunofluorescence technique used and the very low concentration of CaM in the cell wall. in out laboratory, cell wall CaM was purified from wheat etiolated coleoptiles and several CaM-binding proteins in wheat cell wall were also identified24, 25. Further work is currently in progress to investigate the possible function of cell-wall CaM in plant cell regulation.

Finally, CaM may be associated with plasmodesmata in the meristematic cells (Fig 6). But this result requires further confirmatiom