The number of stressors reported to induce the expression of heat shock proteins (HSPs) in plants is extensive (Rees, MNL60:92, 1986; MNL61:69, 1989; Plant Physiol. 90:1256-1261, 1990). In many cases, the synthesis of HSPs has been implicated in the acquisition of stress tolerance. Further evidence suggests that some of the HSPs are developmentally regulated (Bouchard and Walden, MNL64:122, 1990; Atkinson, this issue). In spite of a concerted effort to understand the heat shock response little is known yet about the role these proteins play within the cell. We report on the preparation of polyclonal antibodies against the HSP 18 family and preliminary results on the intracellular localization of these proteins in maize radicles.

The low molecular weight 18kDa family of HSPs was isolated and purified from plumules of etiolated, 4 day-old Oh43 seedlings which had been heat-shocked at 42 C for three hours (as described by Baszczynski, Can. J. Genet. Cytol. 28:1076-1087, 1986). The 18kDa peptides were separated on 7.5%-17.5% gradient SDS polyacrylamide slab gels, eluted and concentrated. A small aliquot of each concentrated protein sample was electrophoresed on 1-D slab gels and samples containing only 18kDa peptides were pooled. An aliquot of the pooled peptides was separated further by 2-D IEF-SDS PAGE and visualized by Coomassie Blue staining and fluorography. The pooled peptides were then emulsified in complete Freund's adjuvant and injected subcutaneously into rabbits. Injections and boosts were performed as described previously by Rees (1989). Crude sera was purified by chromatography on a protein A sephadex CL-4B column (Pharmacia). The IgG-containing column fractions were tested for their antigenic specificity by immunoblotting using nitrocellulose and a Bio-rad alkaline phosphatase conjugate substrate kit to detect antibody binding. As well, immunoprecipitations of radiolabelled proteins extracted from heat-shocked and nonheat-shocked radicles was carried out according to the method of Baszczynski (1986), and immunoprecipitated peptides were analyzed by 2-D PAGE and fluorography. The resulting fluorograms and Western blots demonstrated the specificity of the IgG for the 18kDa family of HSPs. The HSP 18 antibodies showed no detectable cross-reactivity with other heat shock or nonheat shock peptides.

For the purpose of immunocytological localization, heat-shocked and nonheat-shocked Oh43 root tips were cut into 1mm cubic sections and fixed for one hour in 2% gluteraldehyde and 1% osmium tetroxide in 0.2M sodium cacodylate (pH 6.8) at 4 C. The fixed roots were embedded in Epon/Araldite. Gold sections were cut and placed on 400 mesh nickel grids or formvar/cabon-coated slot grids. Sections were treated with sodium periodate, blocked with 1% blotto (in PBS) and incubated with 1:10 IgG:PBS for 1 hour at room temperature. After rinsing in PBS the grids were incubated in protein A-gold conjugate for an hour, rinsed in PBS then ddH2O and viewed on a Phillips CM 10 electron microscope at 60kV. Background level of colloidal gold labelling was determined by incubating serial sections either with preimmune serum or immune serum preabsorbed with 18kDa peptides (see Table 1). Pictures were taken, at 15,000 times magnification, of cortex root cells 1 to 2mm from the root tips. Cells of approximately equal size were chosen from three different heat-shocked and nonheat-shocked roots. Whole cell colloidal gold counts revealed that nonspecific labelling was significantly low (see Table 1), with nonspecific binding appearing randomly throughout the cell.

A definite pattern of HSP 18 distribution was evident in the heat-shocked root cells. Gold particles were clustered within the cytoplasm in groups ranging in size from 12 to 30 particles. This cluster distribution was consistent with observations made by autoradiographic detection of HSPs (Neumann, Eur. J. Cell Biol. 334:254-264, 1984). The clustering has been interpreted as representing the heat shock granules reported and characterized by Nover (Mol. Cell Biol. 3:1648-1655, 1983). A number of other organelles showed high levels of gold labelling (see Table 2). Label was detected in both the nucleus and nucleolus. Gold was also evident close to and on the E.R. and golgi membranes. We found no other cellular structures with a significant increase in the amount of detected HSPs during heat shock. However, increased levels of label were observed along short regions of the plasma wall, primary cell wall and across the middle lamella. Gold was found in discrete, irregular groupings along the cell wall, suggesting that there may be movement of HSPs between cells via the plasmodesmata.

Table 1. Whole cell quantitation of gold particles in heat-shocked and nonheat-shocked (control) root cells. The number of particles counted in control cells incubated with immune serum was assigned a value of 1.
 

Treatment	Pre-immune	Serum immune	HSP absorbed immune
Heat Shock	0.4±0.08*	36.7±6.1	0.8±0.43
Control	0.5±0.20	1±0.61	0.4±0.16

*mean of three cells ± standard deviation

Table 2. Distribution and relative quantity of the HSP 18 family within the cell. The numbers of gold particles within an organelle range from none (-) to greater than 30 particles per square µm (++++).
 

Organelle	Heat Shock	Control
cytoplasm	++++	-
nucleus	++	+
nucleolus	++	+
vacuole	++	++
mitochondria	-	-
golgi	+++	-
endoplasmic reticulum	+++	-
plasma membrane	++	-
cell wall	+	-
middle lamella	++	-

These primary results suggest that, after three hours of heat shock, HSP 18 peptides distribute in a predictable pattern within the cytoplasm and that the HSPs may be transported intercellularly. 

Return to the MNL 67 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage