Planta
Planta 0990)182:9-21
9 Springer-Verlag1990
Mapping membrane potential differences and dye-coupling in internodal tissues of tomato (Solarium lycopersicum L.) C. van der S c h o o t * and A.J.E. van Bel
Transport PhysiologyResearch Group, Botanical Laboratory, University of Utrecht, Lange Nieuwstraat 106, NL-3512 PN Utrecht, The Netherlands Received 6 November 1989; accepted 2 January 1990 Abstract. Symplastic continuity in internodal tissues of
tomato was investigated by electrophysiological and micro-injection techniques. Recordings of potential differences (pd) were combined with iontophoretic injection of 5,6-carboxyfluorescein (CF) and Lucifer Yellow CH (LYCH). Mapping of pds and dye-coupling respectively labeled the potential sites of major solute uptake and the spatial differentiation in symplast permeability. The cells of the central pith had low negative pds ( - 23 mV) and limited dye-coupling. In the pith periphery, pds ( 43 mV) and dye-coupling were more pronounced. The pith was symplastically isolated from the secondary xylem. Axial contact cells (pd - 5 4 to - 1 0 9 mV) showed a strong dye-coupling within the vessel-enclosing sheath. No dye transfer was observed from the vessel sheaths to other cells. Fibre-tracheids and libriform wood fibres were dye-coupled. Ray cells (pd - 3 9 to - 9 4 mV) transported dye to adjacent fibres (pd - 1 8 to - 4 1 mV). The reverse was not observed. Except for a few cases of limited radial cell-cell transport, no dye movement through rays was observed. Cambium cells were connected mainly axially (ray cell initials) or both axially and tangentially (fusiform initials). Radial dye-coupling of cambium cells was rare. The limitations of the present approach and the significance of the differential symplast permeability for xylem-to-phloem transfer are discussed. Key words: Dye-coupling - Lycopersicum (membrane potential, dye transport) - Membrane potential difference - Transport, symplastic - Xylem, secondary
* Present address and address for correspondence: ATO Agrotechnologie, Postbus 17, NL-6700 AA Wageningen,The Netherlands Abbreviations: LYCH= Lucifer Yellow CH ; CF = 5,6-carboxyfluorescein; dcn = dye-couplingnumber; pd = membrane potential difference; Mr = relative molecularmass
Introduction
The stem is the physiological intermediate between the roots and leaves. A major function of the stem is partitioning of materials by long-distance transport, combined with storage and mobilization. This requires an efficient exchange between the long-distance transport channels and metabolic compartments which, in turn, is facilitated by a well-tuned coherence between the hydrosystem and the living elements. The importance of the vascular architecture for longitudinal and lateral transfer has been described recently for internodal tomato xylem (Van der Schoot and Van Bel 1989b). In internodal xylem, the plasma membrane of the contact cells (or vessel-associated cells) appears to present the main barrier between the bulk of the apoplast fluid (the vessel content) and the adjacent symplast elements. The contact pits at the vessel-contact cell interface are highly permeable, in contrast to the remaining wall areas (Wisniewski et al. 1987a, b). It has been suggested that the vessel-associated cells control the vessel contents and act as a kind of 'vessel-companion cell' (Czaninski 1977). These metabolically active cells (Van der Schoot and Van Bel 1989b) execute carrier-mediated processes energized, at least partly, by the proton-motive force (Van Bel and Van Erven 1979a, b; Van Bel et al. 1979; Sauter 1981, 1982). Solute accumulated by ray contact cells may be readily transferred to other ray cells for storage or for transport to the phloem. It is widely accepted that ray transport occurs via plasmodesmata which constitute a symplastic continuity (e.g. Ziegler 1964, 1965; H611 1975; Sauter 1982). It has not been unequivocally demonstrated, however, that this potential symplastic continuum indeed acts as such. Neither is it clear if the ray is a restricted symplast unit. If this is so, the question arises whether other symplastic units exist in the stem, either fixed or variable in size.
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes
10
~....,, ~
m.e./i.e,
Fig. 1. Preparation of plant material. A part of the third internode was excised (a) and a tangential incision was made (al) along which the internode was split (a2), slices with a fracture face were excised with a razor blade (a3), and the slices kept in a bathing medium (e). Alternatively, discs were cut from the internode (b) and kept in the bathing medium (c). Bathing time varied between 1 and 5 h. Tissues were fixed in a perspex bathing chamber and the target cells were impaled under the microscope, r.e., reference electrode; m.e./i,e., measuring/injection electrode
t l_ s
The subdivision of the symplast into 'domains' (Erw e e a n d G o o d w i n 1985) w o u l d g r e a t l y c o n t r i b u t e to a n efficient and selective interchange of materials between s e v e r a l c o m p a r t m e n t s in t h e s t e m . T h e p r e s e n t s t u d y d e s c r i b e s m e m b r a n e - p o t e n t i a l rec o r d i n g s a n d d y e - c o u p l i n g m e a s u r e m e n t s in t h e p i t h - t o cortex region of tomato internodes. The experiments w e r e d e s i g n e d t o i d e n t i f y s y m p l a s t d o m a i n s a n d to d e m o n s t r a t e t h e i r c a p a c i t y f o r a c t i v e u p t a k e . Less a t t e n t i o n is d e v o t e d to t h e sieve t u b e - c o m p a n i o n cell c o m p l e x e s , the electrophysiology and plasmodesmatal coupling of w h i c h h a v e b e e n d e a l t w i t h e l s e w h e r e (Van d e r S c h o o t a n d V a n Bel 1989c).
Material and methods
Preparation of plant material. Tomato plants (Solanum lycopersieum L. cv. Moneymaker) were cultivated as described before (Van der Schoot and Van Bel 1989a). Pot-grown plants with 9 or 10 leaves were used. Tissue slices were excised from the third internode, counted from the cotyledons (Fig. 1). Internode discs of 2-3 mm in thickness were transversely excised and kept in a standard bathing medium (125 m o l . m -3 mannitol, 10 m o l . m 3 NaOH-2-(Nmorpholino)ethane sulfonic acid (Mes) buffer, 0.5 mol-m 3 KC1, 0.5 m o l . m -3 MgC12 and 0.5 mol-m -3 CaC12; pH 5.7) until use (Fig. 1 b, c). Before experimental manipulation, the discs were clamped in a perspex bathing chamber filled with the standard medium. Alternatively, strips were torn along a tangential incision from 3-cm-long internode segments (Fig. 1, al-3). This caused the cells in the fracture face to break along the middle lamella and yielded a fairly regular radial surface as the wood split up along the ray-fibre interface. The convex ' b a c k ' of such split internodes (peripheral stem part) was removed with a razor-blade (Fig. 1, a3). The internode strips were fixed with plastic tape onto microscope slides and fastened in a perspex bathing chamber filled with the standard medium. The wood cells at the fracture face remained viable and thus passage through only one wall was required for microelectrode impalement.
Bathing period and effects of cutting and ageing. As in a previous study (Van der Schoot and Van Bel 1989 c), a minimum acclimatization period of I h was given after cutting. The influence of wounding and ageing on the membrane potential difference (pd) of pith cells (e.g. Mertz and Higinbotham 1976; Pierce and Hendrix 1981)
was determined to check the reproducibility of the measurements in the working time-span (Fig. 2). Tissues were kept in the standard medium up to 24 h. In five subsequent 2-h periods ((~2 h, 3-5 h, 6-8 h, 16-18 h, 22-24 h) pds were measured in pith cells of the phloem-containing pith zone and the inner pith (Fig. 2). The pith cells, selected on grounds of their good accessibility for electrodes, did not show a significant increase of pd during the first 5-7 h but there was a strong increase in the subsequent 15 h (Fig. 2). All electrophysiological and dye-coupling experiments were, therefore, carried out between I and 6 h after excision.
Electrophysiology and iontophoretic injection. Measurement of pds was carried out with conventional electrophysiological techniques as described elsewhere (Van der Schoot and Van Bel 1989c). Glass microelectrodes with an inner filament were fabricated using a vertical pipette-puller. Outer tip diameters were less than I ~tm. For pd measurements, the microelectrodes were backfilled with 100 m o l . m -3 KC1 (3000 mol.mo1-3 KCI in the holder). For dye injection the shanks of the microelectrodes were backfilled with a 0.5% solution of 5,6-carboxyfluorescein (CF) in 0.1% LiC1, or a 0.5% solution of the dilithium salt of Lucifer Yellow CH (LYCH); in either case, filling was completed with LiC1 (3000 mol. m-3). Prior to dye injection, pds were measured to ascertain the viability of the impaled cells. Each measurement was corrected for the electrode potential. Dye injection was carried out by pulsing an intermittent current (usually - 2 to - 2 0 nA) through the injection electrode for 5-60s. Higher currents (--50 to --100nA),
E (~
pith ceils
-150-125-
o
-100 ~D
-~
"[3
-
75-
-
50-
-25[1.
!
i l l t r
0
1
3
5
7
9
l i l t !
17 Bathing time
23 (h)
Fig. 2. Average potential differences (_+ SE; n = 30) recorded from pith cells of tomato stems in five distinct 2-h periods during prolonged bathing. The time area between arrows indicates the working time-span for all further experiments
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes which drove more dye into the target cell, often caused blockage of the tip, probably due to coagulation induced by local heating of the cell contents (Kater et al. 1973). Owing to the thin cytoplasmic layer, the microelectrode often punctured the tonoplast and the dye remained within the vacuole. The dye was released into the cytosol after apparent damage to the tonoplast by one or more short (1 s) high-current (up to -400 nA) pulses and subsequently transferred to the neighbouring cells.
Pressure-driven injection of liposome-encapsulated L YCH. Another approach to overcome the retention of dye in the vacuole was the injection of liposome-encapsulated LYCH. Liposomes were made by a freezing-thawing sonication procedure (Pick 1981 ; Madore and Lucas 1986). Phosphatidylserine (PS) in chloroform (40 mg in 4 ml) was taken to dryness under N2 gas in a Pyrex tube rotating in a water bath at 37~ C, resuspended in chloroform/ methanol (2:5, v/v) and dried again until a uniform layer of PS was deposited on the end of the tube. Lucifer Yellow CH (40 mg in 2 ml) was then added. The solution was vortexed vigorously under nitrogen for 10 min (20~ C) and sonicated to clarity (50 W; 3.17 mm microprobe) in ice (about 7 min). Then it was frozen by rotating the tube in liquid nitrogen and thawed at room temperature. The milky suspension was resonicated for 30 s to partial clarity. The supernatant was run in 0.2-ml fractions over 2-ml Sephadex G-50 columns (pre-equilibrated with 20 mol. m- 3 KCI) to remove non-encapsulated dye. Liposome fractions were recovered in the void volume as monitored by simultaneous blue dextran runs on other columns. Liposomes were pressure-injected (Fisher 1988) via a syringe. To avoid extreme pressures (see Chowdhury 1969) it was necessary to use microelectrodes with a large tip diameter (about 3 4 ~tm) for pressure-injection.
Metabolic activity. The metabolic activity of the various internode cells was investigated in control discs and strips of the same plants by use of thiazolyl blue, which stains dark blue upon reduction by NAD(P)H. The darkness of the cells was taken as a measure of the production of NAD(P)H, as described by Van der Schoot and Van Bel (1989b). Microscopy and photography. Injection and movement of the fluorescent dye (CF or LYCH) were observed continuously under blue light (reduced intensity) with an epifluorescence microscope (Olympus, Hamburg, FRG; BH-2) equipped with the appropriate filterset combination (Van der Schoot and Van Bel 1989c). Cell-to-cell movement of dye was drawn in detail. Observation at full intensity of blue light was restricted to short intervals (1-5 s) to minimize potential cell damage by the blue light. Photographs were made on Kodak Ektachrome 400 (colour slides) (Kodak, Rochester, N.Y., USA) or Agfa (Leverkusen, FRG) Pan 100 ISO (black-andwhite) films. The colour prints and part of the black-and-white prints were reproduced from the slides.
Results
Dye transfer from vacuole to cytoplasm after pressuredriven liposome injection. Owing to the thinness of the cytoplasmic layer, particularly in central pith cells, only a minor part of each injection was directly into the cytoplasm. To overcome this disadvantage, L Y C H - c o n t a i n ing liposomes were injected into the vacuoles. Because of their large volume (up to 4.10 -12 m 3) the cells in the central pith could easily stand the injection of small volumes (2-10-1 s_10-14 m - ~) of liposome suspension (maximally 3% cell volume increase). Small cells, like ray cells (tangential width 10-15 p.m), were often damaged too m u c h by the b r o a d tip (3-4 lam) needed for liposome injection.
11
Fluorescence of L Y C H in liposome-injected cells often faded within minutes (5-15 rain) and disappeared within 15-20 rain (Fig. 3 A - C ) . Occasionally, a very limited cell-cell dye transfer was observed (Fig. 9). Sometimes, the protoplast detached f r o m the cell wall after liposome injection. This lysis was followed by complete protoplast contraction and extinction o f the fluorescence. It was not clear if dye entered the cytoplasm before fading and lysis. In any case, no substantial dye m o v e m e n t was observed resulting f r o m liposome-tonoplast fusion.
Dye transfer from vacuole to cytoplasm after tonoplast rupture by a high current pulse. An alternative to liposome-mediated delivery of L Y C H to the cytoplasm was the application of one or m o r e relatively high current pulses (up to - 4 0 0 nA for 1 s) following dye injection by iontophoresis. The current pulse often p r o v o k e d dye release into the cytoplasm, f r o m where it was transmitted to the neighbouring cells (Fig. 4 A - C , 5A, B). The current pulse induced tonoplast leakiness, shrinkage or rupture (Fig. 2 0 A - C ) or leakage f r o m the cell (Fig. 10). The presumptive disturbance of the cytoplasmic p H did not appear to inhibit dye transfer to adjacent cells. A high current pulse was unnecessary in cells with a thick cytoplasmic layer. In multivacuolar cells the intruding microelectrode tip pushed the vacuoles aside, and had ready access to the cytoplasmic compartment. In conclusion, the use of liposomes to deliver L Y C H from vacuoles to the cytoplasm was very ineffective. The current-induced leakiness or rupture o f the tonoplast might seriously disturb the cytoplasmic p H (and thus cell physiology) but often resulted in cell-cell dye transfer. Therefore the latter method was employed whenever vacuolar injections occurred in further experiments.
Mapping of pds and of dye-coupling in internode discs. In t o m a t o internode discs (Fig. 6A) the various tissue groups, including the bicollateral vascular structure are easily discernible. The pds were systematically recorded in narrow radial bands, producing a pd m a p f r o m the central pith to the epidermis. Further, dye-coupling between the various cell types in this zone was established. Potential differences o f the large central pith cells (Fig. 6) were small ( - 2 3 m V + 3 SE). This was true for the whole of the pith inside the pith layer which contains the internal phloem (Fig. 6). Moving outwards towards the xylem cylinder, the pds tended to become more negative. The phloem-containing pith layer is a transition zone with pds, a little m o r e negative ( - 3 2 mV +__3 SE; Fig. 6) than in the central pith cells. The most peripheral pith cells possessed a pd o f - 4 3 mV +_4 SE (Fig. 6). C o n c o m i t a n t with the pd increase from the central to the peripheral pith the level of dye-coupling also increased strongly (Fig. 6). The dye-coupling here is expressed as the "dye-coupling n u m b e r " (dcn) which is defined as the n u m b e r of cells seen in the section-plane receiving dye from an injected cell (n - 1). With L Y C H , the dcn was 1.8 +_ 1.6 SE in the central pith, increasing to 4.8_+0.6 SE in the phloem-containing layer and up
12
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes
Fig. 3A-C. Pith cell in the internal phloem area of tomato stem injected with LYCH-containing liposomes and photographed after 2 min, (A), 5 rain (B) and 15 rain (C)
to two neighbouring cells (C). ia, injection area; me=microelectrode
Fig. 4A-C. A central pith cell (A), after pd measurement (pd= 23 mV) was iontophoretically injected with LYCH (B). After application of a l-s, --300-nA current pulse, the dye was transmitted
Fig. 5A, B. A central pith cell ( p d = - 2 5 mV) iontophoretically injected with CF (A), transmitted dye to six other cells (14) in the section plane after application of a -200-nA current pulse (1 s) (B)
to 14.3 + 3.1 in the outer pith (Fig. 6). Interestingly, dyec o u p l i n g with C F was similar in the outer pith ( d c n = 16.7 + 2.6 SE) but slightly higher in the m o r e inward pith layer a n d the central area, with d c n = 8 . 0 + 1.7 SE and dcn = 4.0 + 1.3 SE, respectively. This m a y be correlated
with the smaller relative molecular mass (Mr) o f C F ( M r = 4 1 8 D a ) c o m p a r e d with L Y C H ( M r = 457 Da). The d a t a on pd in the pith are summarized in Fig. 6. The pd measurements o f the internal p h l o e m (phloem p a r e n c h y m a pds - 4 7 m V to - 62 m V ; sieve
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes A
--
13
1
3 4 5 25 -45
- 40
0
LYCH
9
CF
/
2O
t3 I
B 1
2
3 4
E c
8
-35
$ -o
-30
-g
-25
15
= g r
10
III II
N
I 9
g 5
9
-20
'I III
II
I
Fig. 6A-C. Scheme of transverse view of internode disc (A), an
enlarged area of the disc (B) and average pds and dye-coupling number for LYCH and CF in the three zones discriminated in the pith (C). A, B Epidermis (1), cortex (2), external phloem (3), cambial zone (4), central pith (5), internal phloem strands (6), primary xylem areas (7), main area of secondary xylem (8). AI-a covers the main secondary xylem areas. B1_3 covers the minor
tube/companion cells pds - 8 6 mV to - 9 8 mV) conf r m e d the results of earlier work with internode discs (Van der Schoot and Van Bel 1989c). Dye-coupling between pith cells and internal phloem elements was never observed, although dye moved from the injected cell to other pith cells (e.g. Fig. 9). Movement of dye from pith cells to cells of the secondary xylem did not occur (Fig. 7, 8). Whether dye moved from the peripheral pith cells into parenchyma cells of the primary xylem was difficult to assess since the borderline between pith and primary xylem was vague. Membrane potential differences in the primary xylem were also very variable. The cells adjacent to the primary xylem vessels (contact cells) had strongly negative pds (up to - 109 mV) while the xylem ground parenchyma possessed pds similar to those o f the peripheral pith or slightly higher (up to - 7 0 mV). The secondary xylem cylinder appeared to be isolated from the enclosed tissues. Only a few of the impalements into living fibre-like axial elements o f the secondary xylem were successful, showing small pds ( - 18 mV to - 3 9 mV). Successful measurements might largely represent compartments of septate fibres (Van der Schoot and Van Bel 1989a, b), like axial parenchyma contact cells survived sectioning more often than fusiform cells (Van der Schoot and Van Bel 1989a, b). The
I'I Pith
III
zones
secondary xylem areas. B/, central pith zone; 1I, phloem-containing pith zone or sub-peripheral pith; III, peripheral pith zone. C The average pd (bar= SE) of cells in pith zones I-III (n = 35), and the average dye-coupling number (n-1 =number of cells in the section plane to which dye was transferred) with CF and LYCH for cells of the different pith zones I-III
vessel-enclosing cells (contact cells) showed quite high pds ranging from - 5 4 mV to - 1 0 9 mV (for example, see Fig. 15). Ray cells possessed pd values ( - 3 9 mV to - 9 4 mV) much more negative than the bordering axial elements (Fig. 15). The lignified walls and the small size o f the ray cells strongly hampered successful impalements. The best impalements were obtained when the electrode tip arrived at the plasma membrane after penetration through a radial pit. These thin-walled unlignified pits had diameters up to 5 gm (Van der Schoot and Van Bel 1989b). As dye injections directly into the cytoplasm were virtually impossible in ray cells, the 'one-shot currentpulse' method was applied. In contrast to pith cells, rupture of the tonoplast by a current pulse (Fig. 2 0 A - C ) often caused leakage through the plasmalemma (Fig. 10) and even lethal damage. The best impalements were apparently those which entered the cell near the nucleus and-or chloroplasts. These impalements always led to cell-cell dye transfer, either ray cell-ray cell or ray cellfibre transfer (Fig. 21 A-C). The ground parenchyma at the outside of the xylem cylinder possessed relatively small pds (Fig. 15) which ranged from - 34 mV to - 59 mV between the external phloem areas, and - 3 6 mV to - 7 0 mV in the cortex (for example, see Fig. 15). In the external phloem the
14
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes
p a r e n c h y m a cells h a d values r a n g i n g f r o m - 4 0 m V to - 62 mV, while the sieve t u b e / c o m p a n i o n cell complexes h a d values o f - 84 m V to - 111 mV. F r o m the external p h l o e m , radial transfer o f C F or L Y C H to the axial system o f the xylem was never observed. The dyes were always arrested in the cambial zone. Likewise, transfer o f C F or L Y C H f r o m the b a r k rays into the w o o d rays was very limited (only two or three cells in the radial direction) a n d in m o s t cases dye t r a n s p o r t halted in the cambial zone (Fig. 24).
Fig. 7. Iontophoretic injection of CF into a peripheral pith cell ( p d = - 4 9 mV) close to the xylem (top left) showing strong dyecoupling within 5 min. me, microelectrode; xv xylem vessel Fig. 8. Iontophoretic injection of LYCH into a peripheral pith cell ( p d = - 52 mV) at the xylem boundary. Lucifer Yellow CH was transmitted to 12 cells in the section plane within 5 min. me, microelectrode Fig. 9. Pith cell injected with liposome-encapsulated LYCH. The injected cell bordered an internal phloem strand (ip; small arrows). A faint trace of LYCH is visible in neighbour pith cells (large arrows)
Colouration by thiazolyl blue. Thiazolyl-blue staining, taken as an indicator for metabolic activity o f internodal tissue (Van der S c h o o t a n d Van Bel 1989b), s h o w e d that the g r o u n d p a r e n c h y m a in the cortex a n d pith was m e t a bolically inert in c o m p a r i s o n with ray, axial p a r e n c h y ma, c a m b i u m a n d p h l o e m cells (scheme Fig. 15). As was f o u n d in longitudinal sections, fibres were c o m m o n l y inactive a l t h o u g h a few scattered axial elements s h o w e d d a r k staining.
Fig. 10. Iontophoretic injection of LYCH into a ray cell vacuole (pd= - 6 7 mV). Application of a current pulse ( - 100 nA) caused leakage without adequate filling of the cytoplasm, v, vacuole; o, cell organelle Fig. 11. Wood fibre (1) iontophoretically filled with LYCH (pd = - 27 mV) transmitted dye to a neighbouring fiber (2) within 5 min (no current pulse) Fig. 12A, B. A Wood fibre ('fibre-tracheid'; p d = - 1 9 m V ) shocked with a 1-s current pulse (-200 nA) 5 min after it was filled with LYCH by iontophoresis. The current pulse split the vacuole in two parts, c, cytoplasm; me, microelectrode. B Fibre of A after another 5 min; no dye-coupling was observed. Some leakage is visible. Ray cells are visible, c, cytoplasm; v, vacuole
Experiments with torn internode strips. The axial orientation o f the vascular tissue p r o m p t e d us to use longitudinal tissue strips for better electrode impalement. Strips were radially t o r n and thus fractured along the rays (Fig. 1). C o m p a r e d with transverse sections, m a n y m o r e axial elements remained alive a n d the success rate o f impalements, pd m e a s u r e m e n t s (fibres: - 24 to - 4 4 mV) and dye injections increased spectacularly. A n additional a d v a n t a g e o f such xylem strips was that only one ' h a l f ' o f the lignified cell wall h a d to be p u n c t u r e d as the other ' h a l f ' was t o r n o f f along the middle lamella. Inevitably, however, the p h o t o g r a p h s lost quality as a result o f the irregularity o f the t o r n surface.
Fig. 13. Wood fibre (1, fibre-tracheid, pd= - 2 4 mV) iontophoretically injected with LYCH into the cytoplasm displays dye-coupling with two other wood fibres (2 and 3, libriform wood fibres) but not with ray cells, xv, xylem vessel, v, vacuole Fig. 14. Libriform wood fibre (pd = - 2 5 mV) iontophoretically injected with LYCH did not show dye-coupling within 10 min, although the dye was in the cytoplasm as indicated by the fluorescent fibre tips (t). r, ray cell
potential difference mapping (-mV) colouratioe by thiazolyl blue infernal phloem
contact cells
41~9 contact cells
!
90 517 95
106 94 68
wood ray
wood ray 62 70 68 62
60 6O 56
23 2~
cambial zone
72
15
external phloem 191 a
Fig. 15. Mapping of membrane potentials ( - m V , all measured values given in the scheme) and of the metabolic activity as indicated by the staining with thiazolyl blue. The scheme is a composition of three seriessections through a main vascular area of a basal tomato internode. The metabolically active tissues which constitute the frame for long-distance translocation and interchange with storage sites are marked with a name. The scheme shows from the left to the fight: (i) pith with embedded internal phloem, (ii) xylem with wood rays, fibres, and xylem vessels surrounded by axial parenchyma (contact cells), (iii) cambial zone, (iv) external phloem, and (v) cortex cells
16
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes
Fig. 16. Control picture (simultaneous blue and white light) of Fig. 11 to show the position of the fibre-tracheid in the wood. v, vacuole; c, cytoplasm; r, ray cell; xv, xylem vessel Fig. 17. Control picture (simultaneous blue and white light) of Fig. 12, showing the position of the fibres in the wood. 1, injected fibre; 2 and 3, position of dye-coupled fibres; v, vacuole; r, ray cell; xv, xylem vessel Fig. 18A, B. Iontophoretic injection of LYCH into a cambium cell (pd = - 54 mY) and immediate rapid transfer in the axial direction (arrows), followed by a spreading sidewards, eventually to 14 cells, me, microelectrode (A). B Cambium cells of A, showing dye displacement in two adjacent axial cell rows (arrows). me, microelectrode Fig. 19. Iontophoretically injected LYCH (target cell pd = - 62 mV) showed immediate and rapid spread in fusiform cambium cells in longitudinal and tangential direction (arrows)
Fig. 20A-C. A Iontophoretically filled ray cell (LYCH; pd = --53 mV) with a shrinking vacuole (v) upon application of a one-shot current pulse of - 100 nA. B 1 min later the same ray cell does not show any dye-coupling after dye release from the vacuole (v) to the cytoplasm (c). C Control picture for A, B. r, ray cell; arrow, injected ray cell
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes
17
Fig. 21A-C. Ray cell in an internode strip of a tomato stem injected iontophoretically with LYCH. A Control, photographed in simultaneous blue and white light, showing ray cells (r) and injection area (ia). B Injected ray cell (r) filled with LYCH (pd = - 7 5 mV). me, microelectrode. C Within l0 min L Y C H has spread from the injected ray cell to the next in line (r) and to several fibre-like adjacent wood elements (arrows) Fig. 22. Strong dye-coupling between axial contact cells. W h e n LYCH was iontophoretically injected into a strand (pd = - 8 8 mV) the dye immediately spread (arrows) to seven or eight other strands (each composed of two to three cells), ia, injection area Fig. 23A, B. Libriform wood fibre injected with LYCH after pd measurement (pd = - 2 7 mV). A Control, photographed in simultaneous blue and white light, r, ray cells: ia, injected area. B Application of a current pulse (1 s, --100 nA) filled the fibre tip (small arrow) but did not result in dye transfer to neighbouring cells. The plasmalemma has become detached laterally (dp). ia, injection area
18
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes
a
Fig. 24. Schematic representation showing types of dye-coupling with LYCH (Mr=457) occurring in the secondary xylem of basal tomato internodes. (a) strong dye-coupling of axial vessel contact cells. (b) dye captured in vacuole without dye-coupling. (c) radial dye transfer to a ray cell next in file; transfer cambium-directed. (d) radial dye transfer in both directions to ray cells in a radial file. (e) dye-coupling between fibre-like elements (or fibres); no movement to ray cells, if) movement of dye from ray cell to neighbouring fibres or fibrous elements. (g) absence of dye-coupling after fibre-injection. (h) strong axial dye-coupling in cambial zone. (i) radial dye transfer from ray cells to cambial cells. (,9 radial dye-transfer from bark ray cells through cambial zone to wood ray cells. *, Injected cells; ,~, cell-cell dye transfer
In the secondary xylem virtually all axial elements were connected with others, as judged from intercellular transfer of CF and LYCH. Dye transfer (both CF and L Y C H ) occurred between all different living fibre-like cell types ('axial parenchyma', 'libriform fibres' and 'fibre-tracheids'; see Van der Schoot and Van Bel 1989 a) (Figs. 11, 13, 17). But vessel contact cells of the axial system appeared to be isolated from the rest of the axial system. Injection of L Y C H or CF into axial vessel contact cells, both in secondary and primary xylem, was often followed by substantial dye transfer exclusively between the contact cells (Fig. 24). In most cases the dye was transferred to several overlapping strands, the latter commonly consisting of two or three cells each. In a few cases many axial elements (up to seven) were dyecoupled (Fig. 22, 24). The reflection coefficient (a) of L Y C H is much higher than that of either 6-CF (Goodwin and Erwee 1985) or 5,6-CF (this paper). Cells injected with L Y C H were therefore almost perfectly outlined, with all the peculiarities of their shape standing out clearly. The cytoplasm, for instance was seen to extend into the far tips of the cells (Fig. 11-14, 16, 17, 23 B). Occasionally, the plasmalemma was detached laterally (Fig. 22 B), probably due to the high current pulse.
No dye transfer was observed from axial elements to ray cells. Interestingly, however, dye-injected ray cells often delivered dye to axial elements (Fig. 21 A-C). Radial dye transfer to other cells was often followed by longitudinal transfer. In the cambium, dye transfer to other cells was very rapid, particularly in the axial direction. The movement of dye (Fig. 18A, B) in the cambium was studied in transverse, radial and tangential sections. Dye-coupling between cambial cells was the most intense of all intercellular movements. It was often difficult to discriminate the different types of cambial cells, as the thin-walled cambium cells were not silhouetted clearly like the lignifled wood cells. In the few clear-cut cases, however, the ray cell initials were observed to be dye-coupled longitudinally (Fig. 18A, B) without side contacts to fusiform cambium initials (tangential section not shown). In other cases, the fusiform initials were observed to be dye-coupled tangentially as well as longitudinally without side contacts to ray initials (Fig. 19). Intimate dye-coupling has been observed between sieve tubes and companion cells (Van der Schoot and Van Bel 1989c). No evidence was found for connections between the external phloem cells and the cortical parenchyma. The transverse and axial dye transport between cortical parenchyma cells was variable ( d c n = 3-7 and 1-3, respectively). Discussion
Liposome injection and vacuole-to-cytoplasm dye transfer. Dye transfer from liposome-injected pith cells to neighbouring cells was virtually absent. It is likely therefore that the liposome/tonoplast fusion rate is very low. Exposure to ultraviolet light might result in fading of the fluorescence (e.g. Goodwin and Erwee 1985; Fisher 1988). Despite the use of blue light (Van der Schoot and Van Bel 1989c) at a low intensity in liposome experiments, we frequently observed a strong quenching or complete fading of fluorescence which also occurred after intermittent exposure to blue light and even during darkness. Such a rapid fading of fluorescence was observed much less frequently after iontophoresis of LYCH. We therefore assume that L Y C H is degraded when the encapsulated form (liposomes) is injected into the cells. Containment of L Y C H by cytoplasmic components of animal cells (De Laat et al. 1980) and plant cells (Oparka et al. 1989) does not result in a decrease of its fluorescence (De Laat et al. 1980; Goodwin and Erwee 1985). Lucifer Yellow CH may form complexes with proteins (Fisher 1988; Oparka et al. 1989). Since LYCH, unlike Lucifer Yellow VS, does not react directly with proteins, but reacts instead with aliphatic aldehydes (Stewart 1981), it might bind during the fusion process to aldehyde groups in the tonoplast prior to protein binding and enzymatic degradation.
Tonoplast rupture and dye transfer from vacuole to cytoplasm. A high current pulse was effective in provoking dye delivery from vacuole to cytoplasm. This was appar-
C. van der Schoot and A.J.E. van Bel: Symplastpermeabilityin tomato internodes ently due to increased tonoplast leakiness in some cases and to tonoplast rupture in others. Mechanical rupture of the tonoplast by a relatively broad-tipped microelectrode also enabled transfer of LYCH between mesophyll cells (e.g. Fisher 1988). A massive disruption of the vacuole will lower the pH of the cytoplasm (pH 7.2; Felle 1987), presumably near to the level of the vacuole (pH 6.0; e.g. Strack et al. 1987). Although the fall in pI-I may have seriously disturbed the cell physiology, cell-to-cell transfer often remained unaffected. Temporary vacuole leakiness (without bursting) might induce smaller shifts in the cytoplasmic pH than vacuolar bursting does. Felle (1988) found evidence for a relation between cytoplasmic pH and free Ca 2 + : acidification led to an increase in cytoplasmic free Ca 2+, and vice versa. A small pH-shift in the cytoplasm might lead to a change in the level of free Ca 2 + via Ca 2 +binding proteins with pH-sensitive binding sites (Felle 1988). The relevance to cell-cell communication is that calcium in turn may reduce dye-coupling. Calcium is known to induce the appearance of electron-dense granules around the plasmodesmatal neck regions, which may contain a Ca-ATPase (Belitser et al. 1982), and acts presumably directly on the plasmodesmata (Goodwin and Erwee 1985). Calcium has been reported to reduce cell-cell dye transfer, probably by closing the plasmodesmata (Erwee and Goodwin 1983). If so, the extent of dye transfer is underestimated in part of our measurements whenever temporary vacuole leakiness followed a current pulse.
Symplast structure in the pith. The increasing dye-coupling towards the peripheral pith probably arises from both an increase in the diameter of the plasmodesmata (Fig. 14; dcn C F > d c n LYCH in central pith and dcn CF
Planta 0990)182:9-21
9 Springer-Verlag1990
Mapping membrane potential differences and dye-coupling in internodal tissues of tomato (Solarium lycopersicum L.) C. van der S c h o o t * and A.J.E. van Bel
Transport PhysiologyResearch Group, Botanical Laboratory, University of Utrecht, Lange Nieuwstraat 106, NL-3512 PN Utrecht, The Netherlands Received 6 November 1989; accepted 2 January 1990 Abstract. Symplastic continuity in internodal tissues of
tomato was investigated by electrophysiological and micro-injection techniques. Recordings of potential differences (pd) were combined with iontophoretic injection of 5,6-carboxyfluorescein (CF) and Lucifer Yellow CH (LYCH). Mapping of pds and dye-coupling respectively labeled the potential sites of major solute uptake and the spatial differentiation in symplast permeability. The cells of the central pith had low negative pds ( - 23 mV) and limited dye-coupling. In the pith periphery, pds ( 43 mV) and dye-coupling were more pronounced. The pith was symplastically isolated from the secondary xylem. Axial contact cells (pd - 5 4 to - 1 0 9 mV) showed a strong dye-coupling within the vessel-enclosing sheath. No dye transfer was observed from the vessel sheaths to other cells. Fibre-tracheids and libriform wood fibres were dye-coupled. Ray cells (pd - 3 9 to - 9 4 mV) transported dye to adjacent fibres (pd - 1 8 to - 4 1 mV). The reverse was not observed. Except for a few cases of limited radial cell-cell transport, no dye movement through rays was observed. Cambium cells were connected mainly axially (ray cell initials) or both axially and tangentially (fusiform initials). Radial dye-coupling of cambium cells was rare. The limitations of the present approach and the significance of the differential symplast permeability for xylem-to-phloem transfer are discussed. Key words: Dye-coupling - Lycopersicum (membrane potential, dye transport) - Membrane potential difference - Transport, symplastic - Xylem, secondary
* Present address and address for correspondence: ATO Agrotechnologie, Postbus 17, NL-6700 AA Wageningen,The Netherlands Abbreviations: LYCH= Lucifer Yellow CH ; CF = 5,6-carboxyfluorescein; dcn = dye-couplingnumber; pd = membrane potential difference; Mr = relative molecularmass
Introduction
The stem is the physiological intermediate between the roots and leaves. A major function of the stem is partitioning of materials by long-distance transport, combined with storage and mobilization. This requires an efficient exchange between the long-distance transport channels and metabolic compartments which, in turn, is facilitated by a well-tuned coherence between the hydrosystem and the living elements. The importance of the vascular architecture for longitudinal and lateral transfer has been described recently for internodal tomato xylem (Van der Schoot and Van Bel 1989b). In internodal xylem, the plasma membrane of the contact cells (or vessel-associated cells) appears to present the main barrier between the bulk of the apoplast fluid (the vessel content) and the adjacent symplast elements. The contact pits at the vessel-contact cell interface are highly permeable, in contrast to the remaining wall areas (Wisniewski et al. 1987a, b). It has been suggested that the vessel-associated cells control the vessel contents and act as a kind of 'vessel-companion cell' (Czaninski 1977). These metabolically active cells (Van der Schoot and Van Bel 1989b) execute carrier-mediated processes energized, at least partly, by the proton-motive force (Van Bel and Van Erven 1979a, b; Van Bel et al. 1979; Sauter 1981, 1982). Solute accumulated by ray contact cells may be readily transferred to other ray cells for storage or for transport to the phloem. It is widely accepted that ray transport occurs via plasmodesmata which constitute a symplastic continuity (e.g. Ziegler 1964, 1965; H611 1975; Sauter 1982). It has not been unequivocally demonstrated, however, that this potential symplastic continuum indeed acts as such. Neither is it clear if the ray is a restricted symplast unit. If this is so, the question arises whether other symplastic units exist in the stem, either fixed or variable in size.
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes
10
~....,, ~
m.e./i.e,
Fig. 1. Preparation of plant material. A part of the third internode was excised (a) and a tangential incision was made (al) along which the internode was split (a2), slices with a fracture face were excised with a razor blade (a3), and the slices kept in a bathing medium (e). Alternatively, discs were cut from the internode (b) and kept in the bathing medium (c). Bathing time varied between 1 and 5 h. Tissues were fixed in a perspex bathing chamber and the target cells were impaled under the microscope, r.e., reference electrode; m.e./i,e., measuring/injection electrode
t l_ s
The subdivision of the symplast into 'domains' (Erw e e a n d G o o d w i n 1985) w o u l d g r e a t l y c o n t r i b u t e to a n efficient and selective interchange of materials between s e v e r a l c o m p a r t m e n t s in t h e s t e m . T h e p r e s e n t s t u d y d e s c r i b e s m e m b r a n e - p o t e n t i a l rec o r d i n g s a n d d y e - c o u p l i n g m e a s u r e m e n t s in t h e p i t h - t o cortex region of tomato internodes. The experiments w e r e d e s i g n e d t o i d e n t i f y s y m p l a s t d o m a i n s a n d to d e m o n s t r a t e t h e i r c a p a c i t y f o r a c t i v e u p t a k e . Less a t t e n t i o n is d e v o t e d to t h e sieve t u b e - c o m p a n i o n cell c o m p l e x e s , the electrophysiology and plasmodesmatal coupling of w h i c h h a v e b e e n d e a l t w i t h e l s e w h e r e (Van d e r S c h o o t a n d V a n Bel 1989c).
Material and methods
Preparation of plant material. Tomato plants (Solanum lycopersieum L. cv. Moneymaker) were cultivated as described before (Van der Schoot and Van Bel 1989a). Pot-grown plants with 9 or 10 leaves were used. Tissue slices were excised from the third internode, counted from the cotyledons (Fig. 1). Internode discs of 2-3 mm in thickness were transversely excised and kept in a standard bathing medium (125 m o l . m -3 mannitol, 10 m o l . m 3 NaOH-2-(Nmorpholino)ethane sulfonic acid (Mes) buffer, 0.5 mol-m 3 KC1, 0.5 m o l . m -3 MgC12 and 0.5 mol-m -3 CaC12; pH 5.7) until use (Fig. 1 b, c). Before experimental manipulation, the discs were clamped in a perspex bathing chamber filled with the standard medium. Alternatively, strips were torn along a tangential incision from 3-cm-long internode segments (Fig. 1, al-3). This caused the cells in the fracture face to break along the middle lamella and yielded a fairly regular radial surface as the wood split up along the ray-fibre interface. The convex ' b a c k ' of such split internodes (peripheral stem part) was removed with a razor-blade (Fig. 1, a3). The internode strips were fixed with plastic tape onto microscope slides and fastened in a perspex bathing chamber filled with the standard medium. The wood cells at the fracture face remained viable and thus passage through only one wall was required for microelectrode impalement.
Bathing period and effects of cutting and ageing. As in a previous study (Van der Schoot and Van Bel 1989 c), a minimum acclimatization period of I h was given after cutting. The influence of wounding and ageing on the membrane potential difference (pd) of pith cells (e.g. Mertz and Higinbotham 1976; Pierce and Hendrix 1981)
was determined to check the reproducibility of the measurements in the working time-span (Fig. 2). Tissues were kept in the standard medium up to 24 h. In five subsequent 2-h periods ((~2 h, 3-5 h, 6-8 h, 16-18 h, 22-24 h) pds were measured in pith cells of the phloem-containing pith zone and the inner pith (Fig. 2). The pith cells, selected on grounds of their good accessibility for electrodes, did not show a significant increase of pd during the first 5-7 h but there was a strong increase in the subsequent 15 h (Fig. 2). All electrophysiological and dye-coupling experiments were, therefore, carried out between I and 6 h after excision.
Electrophysiology and iontophoretic injection. Measurement of pds was carried out with conventional electrophysiological techniques as described elsewhere (Van der Schoot and Van Bel 1989c). Glass microelectrodes with an inner filament were fabricated using a vertical pipette-puller. Outer tip diameters were less than I ~tm. For pd measurements, the microelectrodes were backfilled with 100 m o l . m -3 KC1 (3000 mol.mo1-3 KCI in the holder). For dye injection the shanks of the microelectrodes were backfilled with a 0.5% solution of 5,6-carboxyfluorescein (CF) in 0.1% LiC1, or a 0.5% solution of the dilithium salt of Lucifer Yellow CH (LYCH); in either case, filling was completed with LiC1 (3000 mol. m-3). Prior to dye injection, pds were measured to ascertain the viability of the impaled cells. Each measurement was corrected for the electrode potential. Dye injection was carried out by pulsing an intermittent current (usually - 2 to - 2 0 nA) through the injection electrode for 5-60s. Higher currents (--50 to --100nA),
E (~
pith ceils
-150-125-
o
-100 ~D
-~
"[3
-
75-
-
50-
-25[1.
!
i l l t r
0
1
3
5
7
9
l i l t !
17 Bathing time
23 (h)
Fig. 2. Average potential differences (_+ SE; n = 30) recorded from pith cells of tomato stems in five distinct 2-h periods during prolonged bathing. The time area between arrows indicates the working time-span for all further experiments
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes which drove more dye into the target cell, often caused blockage of the tip, probably due to coagulation induced by local heating of the cell contents (Kater et al. 1973). Owing to the thin cytoplasmic layer, the microelectrode often punctured the tonoplast and the dye remained within the vacuole. The dye was released into the cytosol after apparent damage to the tonoplast by one or more short (1 s) high-current (up to -400 nA) pulses and subsequently transferred to the neighbouring cells.
Pressure-driven injection of liposome-encapsulated L YCH. Another approach to overcome the retention of dye in the vacuole was the injection of liposome-encapsulated LYCH. Liposomes were made by a freezing-thawing sonication procedure (Pick 1981 ; Madore and Lucas 1986). Phosphatidylserine (PS) in chloroform (40 mg in 4 ml) was taken to dryness under N2 gas in a Pyrex tube rotating in a water bath at 37~ C, resuspended in chloroform/ methanol (2:5, v/v) and dried again until a uniform layer of PS was deposited on the end of the tube. Lucifer Yellow CH (40 mg in 2 ml) was then added. The solution was vortexed vigorously under nitrogen for 10 min (20~ C) and sonicated to clarity (50 W; 3.17 mm microprobe) in ice (about 7 min). Then it was frozen by rotating the tube in liquid nitrogen and thawed at room temperature. The milky suspension was resonicated for 30 s to partial clarity. The supernatant was run in 0.2-ml fractions over 2-ml Sephadex G-50 columns (pre-equilibrated with 20 mol. m- 3 KCI) to remove non-encapsulated dye. Liposome fractions were recovered in the void volume as monitored by simultaneous blue dextran runs on other columns. Liposomes were pressure-injected (Fisher 1988) via a syringe. To avoid extreme pressures (see Chowdhury 1969) it was necessary to use microelectrodes with a large tip diameter (about 3 4 ~tm) for pressure-injection.
Metabolic activity. The metabolic activity of the various internode cells was investigated in control discs and strips of the same plants by use of thiazolyl blue, which stains dark blue upon reduction by NAD(P)H. The darkness of the cells was taken as a measure of the production of NAD(P)H, as described by Van der Schoot and Van Bel (1989b). Microscopy and photography. Injection and movement of the fluorescent dye (CF or LYCH) were observed continuously under blue light (reduced intensity) with an epifluorescence microscope (Olympus, Hamburg, FRG; BH-2) equipped with the appropriate filterset combination (Van der Schoot and Van Bel 1989c). Cell-to-cell movement of dye was drawn in detail. Observation at full intensity of blue light was restricted to short intervals (1-5 s) to minimize potential cell damage by the blue light. Photographs were made on Kodak Ektachrome 400 (colour slides) (Kodak, Rochester, N.Y., USA) or Agfa (Leverkusen, FRG) Pan 100 ISO (black-andwhite) films. The colour prints and part of the black-and-white prints were reproduced from the slides.
Results
Dye transfer from vacuole to cytoplasm after pressuredriven liposome injection. Owing to the thinness of the cytoplasmic layer, particularly in central pith cells, only a minor part of each injection was directly into the cytoplasm. To overcome this disadvantage, L Y C H - c o n t a i n ing liposomes were injected into the vacuoles. Because of their large volume (up to 4.10 -12 m 3) the cells in the central pith could easily stand the injection of small volumes (2-10-1 s_10-14 m - ~) of liposome suspension (maximally 3% cell volume increase). Small cells, like ray cells (tangential width 10-15 p.m), were often damaged too m u c h by the b r o a d tip (3-4 lam) needed for liposome injection.
11
Fluorescence of L Y C H in liposome-injected cells often faded within minutes (5-15 rain) and disappeared within 15-20 rain (Fig. 3 A - C ) . Occasionally, a very limited cell-cell dye transfer was observed (Fig. 9). Sometimes, the protoplast detached f r o m the cell wall after liposome injection. This lysis was followed by complete protoplast contraction and extinction o f the fluorescence. It was not clear if dye entered the cytoplasm before fading and lysis. In any case, no substantial dye m o v e m e n t was observed resulting f r o m liposome-tonoplast fusion.
Dye transfer from vacuole to cytoplasm after tonoplast rupture by a high current pulse. An alternative to liposome-mediated delivery of L Y C H to the cytoplasm was the application of one or m o r e relatively high current pulses (up to - 4 0 0 nA for 1 s) following dye injection by iontophoresis. The current pulse often p r o v o k e d dye release into the cytoplasm, f r o m where it was transmitted to the neighbouring cells (Fig. 4 A - C , 5A, B). The current pulse induced tonoplast leakiness, shrinkage or rupture (Fig. 2 0 A - C ) or leakage f r o m the cell (Fig. 10). The presumptive disturbance of the cytoplasmic p H did not appear to inhibit dye transfer to adjacent cells. A high current pulse was unnecessary in cells with a thick cytoplasmic layer. In multivacuolar cells the intruding microelectrode tip pushed the vacuoles aside, and had ready access to the cytoplasmic compartment. In conclusion, the use of liposomes to deliver L Y C H from vacuoles to the cytoplasm was very ineffective. The current-induced leakiness or rupture o f the tonoplast might seriously disturb the cytoplasmic p H (and thus cell physiology) but often resulted in cell-cell dye transfer. Therefore the latter method was employed whenever vacuolar injections occurred in further experiments.
Mapping of pds and of dye-coupling in internode discs. In t o m a t o internode discs (Fig. 6A) the various tissue groups, including the bicollateral vascular structure are easily discernible. The pds were systematically recorded in narrow radial bands, producing a pd m a p f r o m the central pith to the epidermis. Further, dye-coupling between the various cell types in this zone was established. Potential differences o f the large central pith cells (Fig. 6) were small ( - 2 3 m V + 3 SE). This was true for the whole of the pith inside the pith layer which contains the internal phloem (Fig. 6). Moving outwards towards the xylem cylinder, the pds tended to become more negative. The phloem-containing pith layer is a transition zone with pds, a little m o r e negative ( - 3 2 mV +__3 SE; Fig. 6) than in the central pith cells. The most peripheral pith cells possessed a pd o f - 4 3 mV +_4 SE (Fig. 6). C o n c o m i t a n t with the pd increase from the central to the peripheral pith the level of dye-coupling also increased strongly (Fig. 6). The dye-coupling here is expressed as the "dye-coupling n u m b e r " (dcn) which is defined as the n u m b e r of cells seen in the section-plane receiving dye from an injected cell (n - 1). With L Y C H , the dcn was 1.8 +_ 1.6 SE in the central pith, increasing to 4.8_+0.6 SE in the phloem-containing layer and up
12
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes
Fig. 3A-C. Pith cell in the internal phloem area of tomato stem injected with LYCH-containing liposomes and photographed after 2 min, (A), 5 rain (B) and 15 rain (C)
to two neighbouring cells (C). ia, injection area; me=microelectrode
Fig. 4A-C. A central pith cell (A), after pd measurement (pd= 23 mV) was iontophoretically injected with LYCH (B). After application of a l-s, --300-nA current pulse, the dye was transmitted
Fig. 5A, B. A central pith cell ( p d = - 2 5 mV) iontophoretically injected with CF (A), transmitted dye to six other cells (14) in the section plane after application of a -200-nA current pulse (1 s) (B)
to 14.3 + 3.1 in the outer pith (Fig. 6). Interestingly, dyec o u p l i n g with C F was similar in the outer pith ( d c n = 16.7 + 2.6 SE) but slightly higher in the m o r e inward pith layer a n d the central area, with d c n = 8 . 0 + 1.7 SE and dcn = 4.0 + 1.3 SE, respectively. This m a y be correlated
with the smaller relative molecular mass (Mr) o f C F ( M r = 4 1 8 D a ) c o m p a r e d with L Y C H ( M r = 457 Da). The d a t a on pd in the pith are summarized in Fig. 6. The pd measurements o f the internal p h l o e m (phloem p a r e n c h y m a pds - 4 7 m V to - 62 m V ; sieve
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes A
--
13
1
3 4 5 25 -45
- 40
0
LYCH
9
CF
/
2O
t3 I
B 1
2
3 4
E c
8
-35
$ -o
-30
-g
-25
15
= g r
10
III II
N
I 9
g 5
9
-20
'I III
II
I
Fig. 6A-C. Scheme of transverse view of internode disc (A), an
enlarged area of the disc (B) and average pds and dye-coupling number for LYCH and CF in the three zones discriminated in the pith (C). A, B Epidermis (1), cortex (2), external phloem (3), cambial zone (4), central pith (5), internal phloem strands (6), primary xylem areas (7), main area of secondary xylem (8). AI-a covers the main secondary xylem areas. B1_3 covers the minor
tube/companion cells pds - 8 6 mV to - 9 8 mV) conf r m e d the results of earlier work with internode discs (Van der Schoot and Van Bel 1989c). Dye-coupling between pith cells and internal phloem elements was never observed, although dye moved from the injected cell to other pith cells (e.g. Fig. 9). Movement of dye from pith cells to cells of the secondary xylem did not occur (Fig. 7, 8). Whether dye moved from the peripheral pith cells into parenchyma cells of the primary xylem was difficult to assess since the borderline between pith and primary xylem was vague. Membrane potential differences in the primary xylem were also very variable. The cells adjacent to the primary xylem vessels (contact cells) had strongly negative pds (up to - 109 mV) while the xylem ground parenchyma possessed pds similar to those o f the peripheral pith or slightly higher (up to - 7 0 mV). The secondary xylem cylinder appeared to be isolated from the enclosed tissues. Only a few of the impalements into living fibre-like axial elements o f the secondary xylem were successful, showing small pds ( - 18 mV to - 3 9 mV). Successful measurements might largely represent compartments of septate fibres (Van der Schoot and Van Bel 1989a, b), like axial parenchyma contact cells survived sectioning more often than fusiform cells (Van der Schoot and Van Bel 1989a, b). The
I'I Pith
III
zones
secondary xylem areas. B/, central pith zone; 1I, phloem-containing pith zone or sub-peripheral pith; III, peripheral pith zone. C The average pd (bar= SE) of cells in pith zones I-III (n = 35), and the average dye-coupling number (n-1 =number of cells in the section plane to which dye was transferred) with CF and LYCH for cells of the different pith zones I-III
vessel-enclosing cells (contact cells) showed quite high pds ranging from - 5 4 mV to - 1 0 9 mV (for example, see Fig. 15). Ray cells possessed pd values ( - 3 9 mV to - 9 4 mV) much more negative than the bordering axial elements (Fig. 15). The lignified walls and the small size o f the ray cells strongly hampered successful impalements. The best impalements were obtained when the electrode tip arrived at the plasma membrane after penetration through a radial pit. These thin-walled unlignified pits had diameters up to 5 gm (Van der Schoot and Van Bel 1989b). As dye injections directly into the cytoplasm were virtually impossible in ray cells, the 'one-shot currentpulse' method was applied. In contrast to pith cells, rupture of the tonoplast by a current pulse (Fig. 2 0 A - C ) often caused leakage through the plasmalemma (Fig. 10) and even lethal damage. The best impalements were apparently those which entered the cell near the nucleus and-or chloroplasts. These impalements always led to cell-cell dye transfer, either ray cell-ray cell or ray cellfibre transfer (Fig. 21 A-C). The ground parenchyma at the outside of the xylem cylinder possessed relatively small pds (Fig. 15) which ranged from - 34 mV to - 59 mV between the external phloem areas, and - 3 6 mV to - 7 0 mV in the cortex (for example, see Fig. 15). In the external phloem the
14
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes
p a r e n c h y m a cells h a d values r a n g i n g f r o m - 4 0 m V to - 62 mV, while the sieve t u b e / c o m p a n i o n cell complexes h a d values o f - 84 m V to - 111 mV. F r o m the external p h l o e m , radial transfer o f C F or L Y C H to the axial system o f the xylem was never observed. The dyes were always arrested in the cambial zone. Likewise, transfer o f C F or L Y C H f r o m the b a r k rays into the w o o d rays was very limited (only two or three cells in the radial direction) a n d in m o s t cases dye t r a n s p o r t halted in the cambial zone (Fig. 24).
Fig. 7. Iontophoretic injection of CF into a peripheral pith cell ( p d = - 4 9 mV) close to the xylem (top left) showing strong dyecoupling within 5 min. me, microelectrode; xv xylem vessel Fig. 8. Iontophoretic injection of LYCH into a peripheral pith cell ( p d = - 52 mV) at the xylem boundary. Lucifer Yellow CH was transmitted to 12 cells in the section plane within 5 min. me, microelectrode Fig. 9. Pith cell injected with liposome-encapsulated LYCH. The injected cell bordered an internal phloem strand (ip; small arrows). A faint trace of LYCH is visible in neighbour pith cells (large arrows)
Colouration by thiazolyl blue. Thiazolyl-blue staining, taken as an indicator for metabolic activity o f internodal tissue (Van der S c h o o t a n d Van Bel 1989b), s h o w e d that the g r o u n d p a r e n c h y m a in the cortex a n d pith was m e t a bolically inert in c o m p a r i s o n with ray, axial p a r e n c h y ma, c a m b i u m a n d p h l o e m cells (scheme Fig. 15). As was f o u n d in longitudinal sections, fibres were c o m m o n l y inactive a l t h o u g h a few scattered axial elements s h o w e d d a r k staining.
Fig. 10. Iontophoretic injection of LYCH into a ray cell vacuole (pd= - 6 7 mV). Application of a current pulse ( - 100 nA) caused leakage without adequate filling of the cytoplasm, v, vacuole; o, cell organelle Fig. 11. Wood fibre (1) iontophoretically filled with LYCH (pd = - 27 mV) transmitted dye to a neighbouring fiber (2) within 5 min (no current pulse) Fig. 12A, B. A Wood fibre ('fibre-tracheid'; p d = - 1 9 m V ) shocked with a 1-s current pulse (-200 nA) 5 min after it was filled with LYCH by iontophoresis. The current pulse split the vacuole in two parts, c, cytoplasm; me, microelectrode. B Fibre of A after another 5 min; no dye-coupling was observed. Some leakage is visible. Ray cells are visible, c, cytoplasm; v, vacuole
Experiments with torn internode strips. The axial orientation o f the vascular tissue p r o m p t e d us to use longitudinal tissue strips for better electrode impalement. Strips were radially t o r n and thus fractured along the rays (Fig. 1). C o m p a r e d with transverse sections, m a n y m o r e axial elements remained alive a n d the success rate o f impalements, pd m e a s u r e m e n t s (fibres: - 24 to - 4 4 mV) and dye injections increased spectacularly. A n additional a d v a n t a g e o f such xylem strips was that only one ' h a l f ' o f the lignified cell wall h a d to be p u n c t u r e d as the other ' h a l f ' was t o r n o f f along the middle lamella. Inevitably, however, the p h o t o g r a p h s lost quality as a result o f the irregularity o f the t o r n surface.
Fig. 13. Wood fibre (1, fibre-tracheid, pd= - 2 4 mV) iontophoretically injected with LYCH into the cytoplasm displays dye-coupling with two other wood fibres (2 and 3, libriform wood fibres) but not with ray cells, xv, xylem vessel, v, vacuole Fig. 14. Libriform wood fibre (pd = - 2 5 mV) iontophoretically injected with LYCH did not show dye-coupling within 10 min, although the dye was in the cytoplasm as indicated by the fluorescent fibre tips (t). r, ray cell
potential difference mapping (-mV) colouratioe by thiazolyl blue infernal phloem
contact cells
41~9 contact cells
!
90 517 95
106 94 68
wood ray
wood ray 62 70 68 62
60 6O 56
23 2~
cambial zone
72
15
external phloem 191 a
Fig. 15. Mapping of membrane potentials ( - m V , all measured values given in the scheme) and of the metabolic activity as indicated by the staining with thiazolyl blue. The scheme is a composition of three seriessections through a main vascular area of a basal tomato internode. The metabolically active tissues which constitute the frame for long-distance translocation and interchange with storage sites are marked with a name. The scheme shows from the left to the fight: (i) pith with embedded internal phloem, (ii) xylem with wood rays, fibres, and xylem vessels surrounded by axial parenchyma (contact cells), (iii) cambial zone, (iv) external phloem, and (v) cortex cells
16
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes
Fig. 16. Control picture (simultaneous blue and white light) of Fig. 11 to show the position of the fibre-tracheid in the wood. v, vacuole; c, cytoplasm; r, ray cell; xv, xylem vessel Fig. 17. Control picture (simultaneous blue and white light) of Fig. 12, showing the position of the fibres in the wood. 1, injected fibre; 2 and 3, position of dye-coupled fibres; v, vacuole; r, ray cell; xv, xylem vessel Fig. 18A, B. Iontophoretic injection of LYCH into a cambium cell (pd = - 54 mY) and immediate rapid transfer in the axial direction (arrows), followed by a spreading sidewards, eventually to 14 cells, me, microelectrode (A). B Cambium cells of A, showing dye displacement in two adjacent axial cell rows (arrows). me, microelectrode Fig. 19. Iontophoretically injected LYCH (target cell pd = - 62 mV) showed immediate and rapid spread in fusiform cambium cells in longitudinal and tangential direction (arrows)
Fig. 20A-C. A Iontophoretically filled ray cell (LYCH; pd = --53 mV) with a shrinking vacuole (v) upon application of a one-shot current pulse of - 100 nA. B 1 min later the same ray cell does not show any dye-coupling after dye release from the vacuole (v) to the cytoplasm (c). C Control picture for A, B. r, ray cell; arrow, injected ray cell
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes
17
Fig. 21A-C. Ray cell in an internode strip of a tomato stem injected iontophoretically with LYCH. A Control, photographed in simultaneous blue and white light, showing ray cells (r) and injection area (ia). B Injected ray cell (r) filled with LYCH (pd = - 7 5 mV). me, microelectrode. C Within l0 min L Y C H has spread from the injected ray cell to the next in line (r) and to several fibre-like adjacent wood elements (arrows) Fig. 22. Strong dye-coupling between axial contact cells. W h e n LYCH was iontophoretically injected into a strand (pd = - 8 8 mV) the dye immediately spread (arrows) to seven or eight other strands (each composed of two to three cells), ia, injection area Fig. 23A, B. Libriform wood fibre injected with LYCH after pd measurement (pd = - 2 7 mV). A Control, photographed in simultaneous blue and white light, r, ray cells: ia, injected area. B Application of a current pulse (1 s, --100 nA) filled the fibre tip (small arrow) but did not result in dye transfer to neighbouring cells. The plasmalemma has become detached laterally (dp). ia, injection area
18
C. van der Schoot and A.J.E. van Bel: Symplast permeability in tomato internodes
a
Fig. 24. Schematic representation showing types of dye-coupling with LYCH (Mr=457) occurring in the secondary xylem of basal tomato internodes. (a) strong dye-coupling of axial vessel contact cells. (b) dye captured in vacuole without dye-coupling. (c) radial dye transfer to a ray cell next in file; transfer cambium-directed. (d) radial dye transfer in both directions to ray cells in a radial file. (e) dye-coupling between fibre-like elements (or fibres); no movement to ray cells, if) movement of dye from ray cell to neighbouring fibres or fibrous elements. (g) absence of dye-coupling after fibre-injection. (h) strong axial dye-coupling in cambial zone. (i) radial dye transfer from ray cells to cambial cells. (,9 radial dye-transfer from bark ray cells through cambial zone to wood ray cells. *, Injected cells; ,~, cell-cell dye transfer
In the secondary xylem virtually all axial elements were connected with others, as judged from intercellular transfer of CF and LYCH. Dye transfer (both CF and L Y C H ) occurred between all different living fibre-like cell types ('axial parenchyma', 'libriform fibres' and 'fibre-tracheids'; see Van der Schoot and Van Bel 1989 a) (Figs. 11, 13, 17). But vessel contact cells of the axial system appeared to be isolated from the rest of the axial system. Injection of L Y C H or CF into axial vessel contact cells, both in secondary and primary xylem, was often followed by substantial dye transfer exclusively between the contact cells (Fig. 24). In most cases the dye was transferred to several overlapping strands, the latter commonly consisting of two or three cells each. In a few cases many axial elements (up to seven) were dyecoupled (Fig. 22, 24). The reflection coefficient (a) of L Y C H is much higher than that of either 6-CF (Goodwin and Erwee 1985) or 5,6-CF (this paper). Cells injected with L Y C H were therefore almost perfectly outlined, with all the peculiarities of their shape standing out clearly. The cytoplasm, for instance was seen to extend into the far tips of the cells (Fig. 11-14, 16, 17, 23 B). Occasionally, the plasmalemma was detached laterally (Fig. 22 B), probably due to the high current pulse.
No dye transfer was observed from axial elements to ray cells. Interestingly, however, dye-injected ray cells often delivered dye to axial elements (Fig. 21 A-C). Radial dye transfer to other cells was often followed by longitudinal transfer. In the cambium, dye transfer to other cells was very rapid, particularly in the axial direction. The movement of dye (Fig. 18A, B) in the cambium was studied in transverse, radial and tangential sections. Dye-coupling between cambial cells was the most intense of all intercellular movements. It was often difficult to discriminate the different types of cambial cells, as the thin-walled cambium cells were not silhouetted clearly like the lignifled wood cells. In the few clear-cut cases, however, the ray cell initials were observed to be dye-coupled longitudinally (Fig. 18A, B) without side contacts to fusiform cambium initials (tangential section not shown). In other cases, the fusiform initials were observed to be dye-coupled tangentially as well as longitudinally without side contacts to ray initials (Fig. 19). Intimate dye-coupling has been observed between sieve tubes and companion cells (Van der Schoot and Van Bel 1989c). No evidence was found for connections between the external phloem cells and the cortical parenchyma. The transverse and axial dye transport between cortical parenchyma cells was variable ( d c n = 3-7 and 1-3, respectively). Discussion
Liposome injection and vacuole-to-cytoplasm dye transfer. Dye transfer from liposome-injected pith cells to neighbouring cells was virtually absent. It is likely therefore that the liposome/tonoplast fusion rate is very low. Exposure to ultraviolet light might result in fading of the fluorescence (e.g. Goodwin and Erwee 1985; Fisher 1988). Despite the use of blue light (Van der Schoot and Van Bel 1989c) at a low intensity in liposome experiments, we frequently observed a strong quenching or complete fading of fluorescence which also occurred after intermittent exposure to blue light and even during darkness. Such a rapid fading of fluorescence was observed much less frequently after iontophoresis of LYCH. We therefore assume that L Y C H is degraded when the encapsulated form (liposomes) is injected into the cells. Containment of L Y C H by cytoplasmic components of animal cells (De Laat et al. 1980) and plant cells (Oparka et al. 1989) does not result in a decrease of its fluorescence (De Laat et al. 1980; Goodwin and Erwee 1985). Lucifer Yellow CH may form complexes with proteins (Fisher 1988; Oparka et al. 1989). Since LYCH, unlike Lucifer Yellow VS, does not react directly with proteins, but reacts instead with aliphatic aldehydes (Stewart 1981), it might bind during the fusion process to aldehyde groups in the tonoplast prior to protein binding and enzymatic degradation.
Tonoplast rupture and dye transfer from vacuole to cytoplasm. A high current pulse was effective in provoking dye delivery from vacuole to cytoplasm. This was appar-
C. van der Schoot and A.J.E. van Bel: Symplastpermeabilityin tomato internodes ently due to increased tonoplast leakiness in some cases and to tonoplast rupture in others. Mechanical rupture of the tonoplast by a relatively broad-tipped microelectrode also enabled transfer of LYCH between mesophyll cells (e.g. Fisher 1988). A massive disruption of the vacuole will lower the pH of the cytoplasm (pH 7.2; Felle 1987), presumably near to the level of the vacuole (pH 6.0; e.g. Strack et al. 1987). Although the fall in pI-I may have seriously disturbed the cell physiology, cell-to-cell transfer often remained unaffected. Temporary vacuole leakiness (without bursting) might induce smaller shifts in the cytoplasmic pH than vacuolar bursting does. Felle (1988) found evidence for a relation between cytoplasmic pH and free Ca 2 + : acidification led to an increase in cytoplasmic free Ca 2+, and vice versa. A small pH-shift in the cytoplasm might lead to a change in the level of free Ca 2 + via Ca 2 +binding proteins with pH-sensitive binding sites (Felle 1988). The relevance to cell-cell communication is that calcium in turn may reduce dye-coupling. Calcium is known to induce the appearance of electron-dense granules around the plasmodesmatal neck regions, which may contain a Ca-ATPase (Belitser et al. 1982), and acts presumably directly on the plasmodesmata (Goodwin and Erwee 1985). Calcium has been reported to reduce cell-cell dye transfer, probably by closing the plasmodesmata (Erwee and Goodwin 1983). If so, the extent of dye transfer is underestimated in part of our measurements whenever temporary vacuole leakiness followed a current pulse.
Symplast structure in the pith. The increasing dye-coupling towards the peripheral pith probably arises from both an increase in the diameter of the plasmodesmata (Fig. 14; dcn C F > d c n LYCH in central pith and dcn CF
