Histochemistry (1981) 71 : 195-201
Histochemistry 9 Springer-Verlag 1981
Further Studies on the Precipitative Freeze Dissolution Technique for the Localization of Inorganic Ions F. Van Iren and P. Boers-Van der Sluijs Department of Plant Nutrition, Botanisch Laboratorium der Rijksuniversiteit, Nonnensteeg 3, NL-2311 VJ Leiden, The Netherlands
Summary. The precipitative freeze dissolution technique (PFD) for the localization of inorganic ions (Van Iren and Bange 1978) was tested further by means of experiments with gelatin models. In several cases dislocation artifacts appeared. By careful comparison of various results, it appeared possible to draw meaningful conclusions.
Introduction In 1978 Van Iren and Bange introduced a new technique for the localization of inorganic ions in an attempt to combine the cheap, quick and easy processing of precipitation techniques with the more reliable cryotechniques. In short, small segments of the object are rapidly frozen (quenched) and immersed at subzero temperature in a highly concentrated solution of a reagent that forms an almost insoluble precipitate with the ion to be localized. This solution is kept several tenths of a degreee above its melting point in order to allow dissolution of the tissue ice and concomitant precipitation. Although the danger of eutectic thawing within the frozen tissue was recognized from the beginning, the first results, especially with T1 + but later also with Rb § gave confidence: losses of the ions were low and final distribution appeared much better than after conventional precipitation (Van Iren and Bange 1978; Van Iren and Boers 1980a, b; see also Van Iren et al. 1979). However most of the results were significant only at the cell and tissue level (autoradiography), those at the sulcellular level (electron microscopy) remaining disappointing for several reasons (see Van Iren and Bange 1978). In the meantime we performed several control experiments which will be discussed in this paper. We localized T1 § Rb § and Cs § all three being analogues of K § They are taken up and transported by the K § mechanisms in plant and animal cells, although at different rates (Van Iren et al. 1981). PFD =precipitative freezedissolution; GDA-glutar-dialdehyde, mp=melting point; At= differencebetween melting point and actual temperature of PFD solutions
Abbrevations."
0301-5564/81/0071/0195/$01.40
196 Materials
F. Van Iren and P. Boers-Van der Slnijs and Methods
Plant material always consisted of 10-mm long segments of so called low-salt roots of barley (Hordeum vulgare L., several cultivars), that had absorbed Rb § Cs +, or T1 + for different periods from solutions containing about 0.2 mM of the ion concerned. Root diameter was 0.2 to 0.4 ram. Ions were labelled isotopically (S6Rb, 134Cs, and 2~ respectively) at specific activities of 2.1012 to 8.1013 Bq-mol 1 Some typical ion contents of the tissue are given in Table 2. For further details, see Van Iren et al. 1981). Model studies were performed with gelatin blocks that were prepared by inhibition of the gelatin (20%) in aqueous solutions of T1C1 or RbC1. Dimensions of the blocks were about 5 x 2 x 1 mm. Each root segment or gelatin block was quenched on a small strip of aluminum foil in melting Freon F12 ( - 1 5 8 ~ C). They were collected in small nylon gauze bags in boiling nitrogen, in which they were stored overnight if necessary. After rewarming in boiling Freon ( - 3 0 ~ C) for a few seconds the bags were immersed in 20 ml of PFD solution. The latter was stirred magnetically in a thermostated cuvette and contained a high concentration of a precipitant for the ion to be localized, a fixative (glutar-dialdehyde), and sometimes other additives (see Table 1). The melting point of the various P F D solutions was between - 6 ~ and - 1 8 ~ C. They were thermostated 0.3 ~ C to 1.5 ~ C higher (A t) in order to allow dissolution of the tissue ice and concomitant precipitation of the ions to be studied. For T1 +, NaI or CaI2 was used as a precipitant in most of the experiments (for details see Van Iren and Bange 1978). In later experiments T1 + was precipitated as the extremely insoluble T12Se (K~o~176 6.9 x 10 27, Korenman 1963) in a PFD solution containing 20% (NH4)2S and saturated with black Se powder (cf. Voigt 1966; Giusti and Fiori 1969). In one experiment it was precipitated as T13Co(NO2)6 (K{os~ about 1.5 x 10 15 ; Korenman 1963 ; Sill6n 1964). Rb + and Cs + were precipitated as Rb3Co(NO2)6 (K{o~176 10-17; Stephen and Stephen 1963)and C%Co(NO2)6 (K{~~ 10 16; Sill6n 1964) respectively (for details, see Van Iren and Boers, 1980b). After 15 to 20 min the bags were removed, the segments were taken out, and gradually transferred from the PFD solution to a fixative. Dehydration in a graded ethanol series was preceded by rinsing to remove the fixative and was followed by propylene Table 1. Solutions used in the PFD procedures for the localization of T1 +, Rb + and Cs + Treatment
Iodide (T1 +)
Selenide (T1 +)
Cobaltinitrite (Rb + , Cs + , T1 +)
PFD solution composition
1.3 M NaI or 0.8 MCaI2, 3% GDA, 5 mM ascorbic acid, 50 mM tris-HC1, pH 7.5 about - 6 ~ C 0.3 ~ C
20% (NH~)2S, saturated with Se powder
1.14 M Na3Co 3% GDA, 2% acetic acid
about - 1 7 ~ C 1.5 ~ C
about - 16~ C I~
m.p.
At
: none
(NO2)6,
Fix. l h , 0 ~
a s P F D , but 1 0 0 m M I 1 h, 0 ~
100 m M Co(NO2)~-, 3% GDA, 50 mM Tris-NaOH, p H 6 : l h, 0 ~
Rinses, 0 ~ C
100 mM I , 5 mM ascorbic demineralized water : acid: 3 x 5 min 3 x 5 min
100 mM Co(NO2)~- : 3 x 5 min,
Dehydration
ethanol series with 5 mM ascorbic acid: each step 5 min, 0~ C; propylene oxide: 2 x 10 min, room temp.
ethanol series: each step 5 min, 0 ~ C; propylene oxide: 2 x 10 rain, room temp.
ethanol series: first step with 100 mM Co(NO2)63-, each step 5 min, 0~ C; propylene oxide : 2 • 10 min, room temp.
Embedding
gradually in ERL during 18h
gradually in ERL during 3h
gradually in ERL during 18h
Localization of Inorganic Ions
197
oxide treatment and embedding in ERL (Spurr 1969). Most of the fluids used after the PFD contained additives to prevent redissolution or oxidation of the precipitates. Composition and way of application of all solutions have been summarized in Table 1. Preparation of autoradiograms by means of dry cutting of 3-gm thick sections and application of Kodak AR-10 stripping film as well as determination of losses of the relevant ions from the tissue during the whole procedure were as described by Van Iren and Bange (1978) and Van Iren and Boers (1980b). Ultra-thin sections were cut and floated in the knife trough on a saturated solution of TII, Rb3Co (NO2)6, or on distilled water for electron microscopical observation of root material containing TII, Rb3Co (NOz)6, or T12Se precipitates respectively. After collection of the sections on Pioloform coated copper grids, adhering through solutions were removed by dipping the grids in isopentane (Van Iren and Van der Spiegel 1975).
Results
Losses of Ions from Gelatin Blocks and Root Segments During the Localization Procedures Losses from the tissue of the ion studied were always measured. Table 2 summarizes the losses observed in those experiments from which autoradiograms are shown in this paper. For the same ion and the same P F D procedure relative losses were negatively correlated with the ion content of the tissue. In general losses during the precipitation step of the procedure were less than 2%. During the preparation for plastic embedding some redissolution was inevitable.
Autoradiograms of Gelatin Blocks Gelatin blocks containing 5 to 50 m M of T1 + were subjected to P F D either with selenide or with cobaltinitrite as precipitant. Those containing Rb + were processed with the latter reagent only.
Table 2. Losses of ions from gelatin blocks or root segments during some localization experiments Gelatin blocks Ion
TI +
TI +
Rb +
Rb +
5
50
10
10
50
PFD solution: m.p. (~ A t (o C)
Se 2-19.0 1.5
Se 2 -19.0 1.0
Co a -15.6 1.0
Co a -15.6 1.0
Losses (%) : PFD fix, rinse dehydration embedding
0.7 1.3 0.2 0.8
0.6 0.4 0.2 0.1
1.7 4.6 0.1 0.1
Recovery (%)
96.2
98.9
1
2
Content (~tmol 9g- 1)
Fig. number a CO (NO2)3
T1 +
Root segments T1 +
T1 +
Rb +
Cs +
8
8
17
12
Co a -14.7 1.0
Se 2-16.0 1.0
I -6.5 0.3
Co a -16.0 1.0
Co a -15.0 1.0
0.9 1.6 0.1 0.0
0.8 0.6 0.0 0.0
1.6 3.0 0.1 0.2
0.5 1.2 0.2 2.1
0.5 1.1 0.1 0.4
1.8 3.0 0.2 0.1
93.5
97.4
98.6
95.1
96.0
97.9
94.4
3
5
6
7
8
9
198
F. Van Iren and P. Boers-Van der Sluijs
Figs. 1-6. Autoradiograms of semithin (1-3 gm) sections of gelatin blocks containing different amounts of Rb + or T1+, labelled with 86Rb or Z~ respectively. Arrows indicate edge of gelatin block Fig. 1. 5 mM T1 +, spec. act. 5.101 a Bq. mol- 1, PFD with SeZ-, exposure 100 d Fig. 2. 50 mM T1+, spec. act. 1013 Bq. tool- 1, PFD with Se2-, exposure 14 d Fig. 3. 10 mM TI +, spec. act. 5.1013 Bq.mol- 1, PFD with Co(NOz)63-, exposure 28 d Fig. 4. 50 mM T1+, spec. act. 1013 Bq.mol -x, PFD with Co(NOz)63 , exposure 7 d Fig. 5. 10 mM Rb +, spec. act. 8- 1013 Bq.mol- l, PFD with Co(NOz)63-, exposure 7 d Fig. 6. 50 mM Rb +, spec. act. 2- 1013 Bq.mol- 1, PFD with Co(NOz)63 , exposure 7d
A f t e r P F D w i t h Se 2 - the b l o c k s c o n t a i n i n g 10 m M T1 + y i e l d e d a u t o r a d i o g r a m s w i t h a n e v e n d i s t r i b u t i o n o f silver g r a i n s e x c e p t f o r a slightly h i g h e r o r l o w e r l a b e l l i n g o f the e d g e s o v e r a d i s t a n c e o f 5 to 10 g m (Fig. 1). W i t h 50 m M T1 + (Fig. 2) a single h e a v i l y l a b e l l e d d o t was f o u n d o c c a s i o n a l l y . W i t h C 0 ( N O 2 ) 6 3 ; as p r e c i p i t a n t b o t h TI + a n d T b + c o n t a i n i n g b l o c k s rev e a l e d a u t o r a d i o g r a m s w i t h i n h o m o g e n e o u s d i s t r i b u t i o n o f label. O f t e n m o s t
Localization of Inorganic Ions
199
Figs. 7, 8 and 9. Autoradiograms of semithin (3 gm) sections of roots after absorption during 4 h from 0.2 or 0.3 mM T1+, Rb + or Cs +, labelled with 2~ 86Rb, or 134Cs respectively. Sections were cut at about 50 mm from the root tip Fig. 7. TI +, spec. act. 7.1012 Bq-mol 1, PFD with CaI2, exposure 3 d Fig. 8. Rb § spec. act. 4.10 ~z Bq.mo1-1 PFD with Co(NO2)63-, exposure 10 d Fig. 9. Cs +, spec. act. 1012 Bq.mo1-1, PFD with Co(NO2)63-, exposure 30 d o f the label was f o u n d in patches, 3 to 10 btm in d i a m e t e r (Figs. 3-6). W i t h R b § s o m e t i m e s a h o m o g e n e o u s l y labelled zone s e p a r a t e d the d o t t e d centre o f the b l o c k f r o m the o u t e r m e d i u m (Fig. 6).
Autoradiograms of Root Sections M a n y e x p e r i m e n t s were p e r f o r m e d in which b a t c h e s o f excised r o o t s were all o w e d to t a k e up R b § Cs +, or T1 § for different p e r i o d s (several m i n u t e s to 5 h). The results o f s h o r t e r u p t a k e times have been discussed elsewhere (Van I r e n et al. 1981). H o w e v e r there were no significant differences with the results o f 4 h o f u p t a k e t h a t will be s h o w n in this p a p e r , except for the a m o u n t o f label. W i t h i n the e p i d e r m i s a n d the cortex all l o c a l i z a t i o n results for the three ions s t u d i e d were alike (Figs. 7-9). M o s t o f the label was f o u n d near the periphery o f the cells. In these m a t u r e p l a n t cells the central vacuole occupies 80 to 90% o f the volume, the c y t o p l a s m a n d the wall t o g e t h e r being only 1 to 2 btm thick. It has been d e m o n s t r a t e d before (Van Iren a n d Bange 1978; Van Iren et al. 1981) that m o s t o f the precipitates are in the c y t o p l a s m , the cell wall a n d the vacuole c o n t a i n i n g little o r no precipitate ( E l e c t r o n m i c r o s c o p i c a l observations). F r o m kinetic d a t a it was c o n c l u d e d t h a t up to a b o u t 3 h o f
200
F. Van Iren and P. Boers-Van der Sluijs
uptake the vacuolar ion concentration is very low relative to that in the cytoplasm. Also the cell wall should be virtually free of ions at the end of the uptake procedure. So the localization results are in line with the interpretation of kinetic data (see Van Iren et al. 1981). The endodermis that forms the boundary layer between the cortex and the stele was labelled more than the cortex in most of the autoradiograms. However, with Rb § and Cs + after precipitation with cobaltinitrite (Figs. 8 and 9) several heavily labelled dots appeared. These were absent with T1 + after the usual precipitation with I - (Fig. 7) or Se 2-. This difference was even more pronounced within the stele: with T1 § no stelar tissue was labelled preferentially but with Rb § or Cs + heavily labelled dots were found, often near the xylem vessel walls. As a control we subjected several root segments, loaded with T1 § to P F D with either selenide or cobaltinitrite as precipitant. Again the same difference appeared. Contrary to the selenide precipitation, cobaltinitrite gave rise to massive patches of label in the central region of the root, also often near the xylem vessel walls. Electron Microscopical Observations on Root Material Details of ultrastructural localization of T1 § after P F D with I - have been published before (Van Iren and Bange 1978). With Se 2-, precipitates appeared much finer and were discernible only at higher magnification. On the other hand precipitation of Rb + with cobaltinitrite appeared less fine. Some precipitate grains were found in the cell walls, viz. in the primary wall and in the corners of the extracellular spaces. Despite application of several modifications, preservation of ultrastructure was not improved relative to that published before (Van Iren and Bange 1978). Discussion Especially the gelatin controls reveal that there is no reason to assume dislocation of T1 § during the preparation of autoradiograms by use of P F D with selenide, at least at the tissues level. We did not study precipitation of T1 § with I in gelatin. However the results in root material did not differ from those with Se 2-. Thus the same reability for P F D with the two precipitants is indicated. When cobaltinitrite was used to precipitate either T1 § or Rb + or Cs § dislocation appeared possible. Since the overall distribution of T1 § Rb § and Cs § within the cortex of our material was always the same, irrespective of the precipitant used, these results can be regarded as reliable. The dense patches within the stele that are found only after precipitation with cobaltinitrite, must be regarded as artifacts. Probably the explanation of the different reliability of these localizations is to be found in the phenomenon of eutectic thawing. This phenomenon has already been suspected to interfere with effective immobilization of ions in frozen tissue (Van Iren and Bange 1978). In cortex the process of precipitative freeze dissolution can proceed relatively unhampered and be
Localization of Inorganic Ions
201
completed within seconds. Thus, the time available for the formation of areas of eutectic thawing, their interconnection, and the diffusion of ions within them is very limited in this tissue. The adaxial endodermal walls, which are (at least partially) thickened and suberized in (semi-) mature tissue, together with the Casparian strips certainly have a larger resistance to the passage of the PFD front. Therefore, within the stele dislocation due to eutectic thawing might occur in the case of the relatively large and slowly diffusing cobaltinitrite ion but appears virtually absent with the small selenide or iodide ions. The endodermis may have an intermediate position. It seems feasible that its dense labelling, especially in the passage cells, as found for T1 + with selenide and iodide, was reinforced in the case of cobaltinitrite by dislocation of ions from the stele. Alternatively, differences in the probability of heterogeneous nucleation and in rates of homogeneous nucleation could underly the differences in localization as well. Anyway these results reduce the optimism uttered by Van Iren and Bange (1978) with respect to the more general applicability of the PFD method, which then seemed justified on the basis of the results with T1 +. They also illustrate the hazard in regarding a certain localization result as reliable without comparison of the results of different techniques or variations in the same technique or without careful control experiments. By use of such comparisons we were able to draw meaningful conclusions (see Van Iren and Boers 1980a, b; Van Iren et al. 1981). References
Giusti GV, Fiori A (1969) The histochemical detection of thallium in paraffin embedded tissues by the sulfide-selenium-silver method. Stain TechnoI 44:263-267 Korenman IM (1963) Analytical chemistry of thallium. Israel Program Scientific Translations, Jerusalem Sill6n GL (1964) Stability constants of metal ion complexes. Section I: inorganic ligands. Special publication no. 17 of the Chemical Society, London Spurt AR (i969) A low viscosity epoxy resin embedding medium for electron microscopy. J Ultrastruct Res 26:31-43 Stephen H, Stephen T (1963) Solubilities of organic and inorganic compounds. Pergamon Press, Oxford London New York Paris Van Iren F, Bange GG J (1978) Localization of inorganic ions by precipitative freeze dissolution. Histochemistry 55 : 81-96 Van Iren F, Van der Spiegel A (1975) Subcellular localization of inorganic ions in plant cells by in vivo precipitation. Science 187:1210-1211 Van Iren F, Boers-Van der Sluijs FP (1980a) Localization of rubidium in barley roots: a specific role of the root surface in primary absorption. In: Spanswick RM, Lucas WJ, Dainty J (eds) Plant membrane transport. Elsevier/North Holland, Amsterdam, pp 431-432 Van Iren F, Boers-Van der Sluijs FP (1980b) Symplasmic and apoplasmic radial ion transport in plant roots: cortical plasmalemmas loose absorption capacity during differentiation. Planta 148:130-137 Van Iren F, Boers-Van der Sluijs FP (1980b) Symplasmic and apoplasmic radial ion transport potassium translocation from the root to the shoot of law-salt barley plants. Kinetic and localization studies. PhysioI Plant (in press) Voigt GE (1966) Histochemical demonstration of T1. Acta Med Leg Soc 19:17 18 Received December 22, 1980