Planta
Planta (1983)159:38-45
9 Springer-Verlag 1983
Water relations of the epidermal bladder cells of O x a l i s c a r n o s a Molina E. Steudle 1, H. Ziegler 2 and U. Zimmermann 1 1Arbeitsgruppe Membranforschung am Institut ffir Medizin, Kernforschungsanlage J/jlich GmbH, Postfach 1913, D-5170 Jiilich, and z Institut f/Jr Botanik und Mikrobiologie der Technischen Universitfit, Arcisstrasse 21, D-8000 Miinchen, Federal Republic of Germany
Abstract. All of the cells of the upper (adaxial) epidermis of the leaves of Oxalis carnosa are transformed into large bladders, while in the lower epidermis the bladder cells are interrupted by "normal" cells with stomata. The epidermal bladders contain a high concentration of free oxalic acid (pH approx. 1). Water-relations parameters of these epidermal bladder cells have been determined using the pressure probe. Original cell turgor (Po) of the closely packed bladders of the upper epidermis was Po=0.7 to 2.9bar (Po=l.7_+0.5bar; mean_+ SD; N = 25 cells) and lower than that in the club-shaped bladders of the lower epidermis (P0 =1.3 to 3.7 bar; Po=2.5_+0.7 bar; N = 25 cells). Large differences in the elastic modulus (e) and the hydraulic conductivity (Lp) of the two different types of cells were observed. For the lower epidermal bladders, e = 1 8 to 166 bar and was similar to that of other higher plant cells. Also, for these cells it was found that e was increasing with both, cell turgor and cell volume. By contrast, e of the cells of the upper epidermis was by one order of magnitude smaller (e=1.9 to 17.0 bar) and no dependence of e on cell volume could be detected. The Lp values of the cell membranes were also different (lower epidermis" Lp=(2_.3_+1.6). 10- 5 cm s- 1 b a r - 1 ; upper epidermis: Lp = (3.8 _+ 2 . 4 ) . ]0 -6 c m s - 1 b a r - l ) . These differences seem to be too large to be caused by errors in determining the exchange area for water (A) between cells and adjacent tissue. The half-times of water exchange between bladders and leaf (T1/2) were, on average, somewhat longer for the upper than for the lower epidermis (lower epidermis: T1/2 = 7 to Abbreviations: P=cell turgor; Lp=hydraulic conductivity of cell membrane; T1/2 = half-time of water exchange between cell and surroundings; e=volumetric elastic modulus; A = e x change area for water of cell; V = cell volume; ~i = osmotic pressure of the cell sap
38 s; upper epidermis: T1/2=22 to 213 s), but the differences in the T1/2 values were not as distinct as for e and Lp. This is because of the compensatory effects of e, Lp and the different ratios of volume to exchange area. Since the bladders make up about 75% of the entire volume of the leaf, it is assumed that the rate of response of the leaf to changes in the water potential should be similar to that of the bladder cells. The results are discussed in terms of a possible function of the bladders in the leaf. Key words: Bladder cells - Elastic modulus - Hydraulic conductivity - Oxalic acid - Oxalis - Water transport.
Introduction
Epidermal bladder cells play an important role in both the elimination of salts from metabolically active tissue and in water storage of different plant species. For example, the large bladder cells of the halophilic Chenopodiaceae such as Atriplex spongiosa and Chenopodium album seem to be involved in the active removal of salts from the underlying tissue (for reviews see Lfittge 1975; Hill and Hill 1976), whereas the giant epidermal bladder cells of Mesembryanthemum crystallinum do not accumulate salts (Ltittge et al. 1978) but rather function as peripheral water reservoirs. They protect the plant from short-term water stress and are thus also involved in salinity adaptation (Steudle et al. 1975). Both possible functions of the bladders are related to the water-relations parameters of these cells, i.e. to the water permeability (hydraulic conductivity) of these cells and to their elastic extensibility (elastic modulus of the cell wall). A reservoir
E. Steudle et al. : Water relations of Oxalis bladder cells
39
or depot cell has to be extensible enough to provide a certain storage capacitance and a sufficiently quick transfer of the stored water or solutes across the cell membrane. In this communication we have investigated water-relations parameters of the large leaf bladder cells of the South-American species Oxalis' carnosa using the pressure-probe technique (Zimmermann and Steudle 1978, 1980). Material and methods Material. Plants of Oxalis' carnosa Molina, a species native to Chile, Peru and South Bolivia, were grown from seeds in potting soil in a climatic chamber (temperature: 22 ~ C (day); 19 ~ C 9 (night); light intensity: 170 pmol photons m 2 s-1; light/dark rhythm: 12 h/12 h; humidity: 60/85%). Plants used in the experiments were 6-9 months old and 10-20 cm high. They had already flowered once (usually 4-6 months after sowing). The plants were well watered (2-3 times per week), but were prevented from being water-logged. Scanning electron microscopy. Plant material was shock-frozen in melting freon 11 and freeze-dried on metal plates without intermediate thawing. The plates were then coated with goldpalladium and examined in a JSM 35C scanning electron microscope (Jeol, Fa. Kontron, Eching). Determination of water relations parameters. For the experiments with the pressure probe, one leaflet of the trifoliate leaf (still attached to the plant) was mounted, with either the upper or lower epidermis facing upwards, in a small chamber on a perspex block and was held by means of a cover glass which was clamped on the leaf. The cover slip had a hole (diameter: about 5 mm) through which the pressure probe could be introduced into the epidermal cells. The experiments were performed with the leaf in air. The determination of water relations parameters (half-time of water exchange between bladder cell and adjacent tissue, T1/2; volumetric elastic modulus, e; hydraulic conductivity of cell membrane, Lp) was performed as described in previous papers and reviews (e.g. Zimmermann and Steudle 1978, 1980; Tomos et al. 1981; Steudle et al. 1982; Tyerman and Steudle 1982). The upper (adaxial) epidermal cells of O. carnosa were sometimes (and mainly in the beginning of our experiments) difficult to measure, since these cells showed a tendency to become leaky. This is probably because they have a very high elastic extensibility (low e) which tends to make the seal between the cell and microcapillary leaky. Such leaks could easily be detected by a drop in cell turgor and the measurements on these ceils were discarded. In order to evaluate Lp from T1/2 and e, the osmotic pressure of the cell sap (~i) and the volume (V) to surface area (A) ratio of the cell must be known, since: V ln2 Lp = X T1/~
1 e+~r i"
(1)
~i was determined by deep freezing leaves of different plants for I h in liquid nitrogen. After thawing, cell sap was gained by centrifuging the tissue through a filter paper. For four different plants the extracted sap had an osmolarity of c~= 300 • 19 mosmol (mean i SD ; N = 4) which is equivalent to an osmotic pressure of ~i=7.32• bar (temperature=20 ~ C). The cell sap showed a very low pH of 1-1.5 caused by a high content of oxalic acid (Pkl = 1.23; Pk2 = 4.19). About one half or 153.0 • 13.6 mM of the total osmolarity of
the cell sap was oxalic acid. Oxalic acid was determined by precipitation with Ca 2 § separating the Ca-oxalate and re-dissolving it in dilute H2SO ~. Oxalic acid was titrated with 0.1 N KMnO4.
Results
Cell geometry and exchange area f o r water flow. Cell geometry and cell dimensions were determined with the aid of the microscope (by looking at surface views and at cross sections of the leaves) as well as by the use of scanning electron micrographs (Fig. 1). In most cases, the cells of the upper (adaxial) epidermis were approximated either as a prism with a hexagonal cross section or as a cylinder. Sometimes (mainly for cells of small volumes) the shape was similar to that of a rectangular prism. The dimensions of the cross section were determined for each individual cell from the surface view (see Fig. 1 a), whereas for the height (h), an average value, which was h=163.5 • gm (mean + SD; N = 39 cells), was determined from cross sections of different leaves. Since the bladder cells of the upper epidermis are closely packed, the exchange area for water of these cells was calculated from the sum of the area of the anticlinal cell walls plus the area of the base of the cells. For the cells of the upper epidermis, the exchange area, A, ranged between 4.5 and 2 3 . 0 . 1 0 - 4 c m 2 and the cell volume, V, between 0.8 and 11.9 nl. In the lower epidermis, the bladder cells are not closely packed since there has to be space for stomata complexes (Fig. i b). The shape of the bladders was approximated by spherical segments or by a truncated cone with a half of a sphere attached to the larger (end) base of the cone. The diameter of the sphere (2 R) was determined from the surface view of the leaf using a stereomicroscope. The diameter of the circular base of the epidermal bladders (2r) as well as the height of the cells (h) could be determined by cutting the leaf transversely after the measurement, turning it by 90 ~ and looking at the cells from the side. With these data, the volume and the exchange area of the cell for water (i.e. the area of the circular base) could be calculated. For the cells investigated in this paper 2 R = 1 2 0 - 3 2 5 ~m, 2 r = 7 0 - 2 8 5 gm, which resulted in values of the cell volume of V = 0.9 to 18.1 nl and of the exchange area of A = 0 . 4 to 2.8- 10-4 cm 2. Thus, the V/A ratio of the bladder cells in the lower epidermis was larger than that of the ceils in the upper epidermis. Water relations. Cell turgor (Po) in the bladder cells of both the lower and upper epidermis was similar,
40
E. Steudle et al. : Water relations of Oxalis bladder cells
Fig. 1 a-c. Scanning electron micrographs of the upper (a) and
lower (b) surface and of a fractured segment (c) of an carnosa leaf
although there was a trend for Po to be slightly lower in the upper epidermis and this is perhaps related to the fact that e in the upper epidermis is much lower than that of the lower epidermis (see below). In the lower epidermis Po ranged between 1.3 and 3.7 bar (Po=2.5_+0.7bar; mean _+SD ( N = 25 cells)), whereas in the upper epidermis P 0 = 0 . 7 to 2 . 9 b a r ( P 0 = l . 7 _ + 0 . 5 b a r ; N = 25 cells). The osmotic pressures of lower and upper epidermal cells could not be determined separately (as already mentioned in Material and methods), but only an average value for the whole leaf (including the mesophyll) could be obtained. This value of ~zi = 7.3 bar should represent a reasonable estimate of the mean osmotic pressure of the bladder cells, because the bladders make up about 75% of the total leaf volume. However, an asymmetry in rci which may be indicated by the differences in Po and other water-relations parameters (see below), may have been overlooked by using this simple method. At present, no reliable method seems to be available for determining the osmotic pressure of amounts of cell sap of the order of picoliters. This would be necessary to determine rc~ of
Oxalis
the two different epidermal layers or even of individual cells. The water-relations parameters of the bladder cells of the lower and upper epidermis showed significant differences (Table 1). On average, the halftime of water exchange between bladder cells and their surroundings (T1/2) was larger in the upper than in the lower epidermis (Table 1, Fig. 2). According to equation (1) this difference may have various causes. Since e of the bladders of the lower epidermis is about one order of magnitude larger than that of the upper epidermis, considerable differences in T~/2 are expected. However, the effect of the elastic modulus on TI/2 should be largely offset by the fact that the ratio of volume to exchange area (V/A) for the lower epidermis is much larger than that for the upper epidermis (V/A = 2 to 1 5 " 1 0 - 2 c m for the lower and 1.5 to 5.2.10-3 cm for the upper bladders). Furthermore, the hydraulic conductivity of the bladders of both epidermal layers appears to be significantly different. The Lp-values for the lower bladders ranged from 0.5 to 6.7.10 -5 cm s -1 bar -1 (L-p=(2.3_+ 1.6)'10 -5 cm s -1 b a r - l ; N = 2 2 cells), whereas Lp for the upper bladders was L p = 0 . 5 to
E. Steudle et al. : Water relations of Oxalis bladder cells
41
Table l. Typical values of the volumetric elastic modulus (e), half-time of water exchange (Tin) and hydraulic conductivity (Lp) of epidermal bladder cells of Oxalis carnosa leaves. The cells shown cover the total ranges of cell volumes investigated. In the lower epidermis, e is a function of cell turgor and volume, whereas in the upper epidermis no dependence on cell volume could be detected. Note the differences in T1/2 and Lp between upper and lower epidermis. Mean values are given _+ SD with the number of experiments in brackets Cell No.
Lower epidermis
Upper epidermis
Cell V/A x 103 volume, V (nl) (cm)
Pressure range for e (bar)
e
2
0.89
23
0.6-2.6 1.8-4.0
29 _+33 (2) 34 -+
11
5.6
18
1.6-3.5 2.1~4.4
50 89
35.0_+ 5.6(2)
i6
8.1
33
2.0-2.5 2.2 3.2
32 89
28.0_+11.5(2)
23
21
10.5
38
0.2-1.5 1.1 2.7
30 104
47.6
15
25
18.1
150
1.3-5.3
158
26.3-+ 9.1 (2)
26 _+12 (2)
2
1.5
1.5
1.5 2.4
1.9_+ 1.1 (2)
65
_+ 1 (2)
1.7_+ 0.0(2)
3
2.4
2.5
0.7-1.8
9.0_+ 5.3 (2)
54 _+15 (2)
1.3_+ 0.3 (2)
10
4.1
2.9
0.8 1.8
6.4_+ 0.9 (2)
30 _+13 (2)
5.4_+ 2.2 (2)
15
6.0
4.1
1.2-2.0
15.1
29 _+ 4 (2)
4.4_+ 0.6 (2)
16
6.1
4.1
1.7-2.9
3.8
119
20
7.4
4.4
2.9-3.5
5.3
213
25
11.9
5.2
0.0-2.3
8 . 0 . 1 0 - 6 c m s-1 bar-1 ( L p = 3 . 8 + 2 . 4 ) . 1 0 - 6 c m s -1 bar 1; N=18cells), i.e. the average values were almost one order of magnitude different. The Lp of the upper epidermal cells was similar to most of the values reported in the literature for higherplant cells (see reviews by Zimmermann and Steudle 1978, 1980; Dainty 1976), but the Lp for the lower epidermis was larger. An asymmetry of water-relations parameters between bladders in the two epidermal layers has been also found with respect to the elastic modulus of the cells. For the closely packed cells in the upper epidermis e ranged between 2 and 17 bar and is much smaller than the e values found for other higher plant cells which are of the order of 10-100 bar at normal cell turgor (Zimmermann and Steudle 1978, 1980; Dainty 1976). The absolute values of e of the upper epidermal bladders were of the same order as the values given for the giant-celled alga Halicystis parvula which were 0.5 to 2 bar under stationary (long-term) conditions and 1 to 16 bar under short-term conditions (Zimmermann and H/]sken 1980). In contrast to the upper epidermis, the values of e in the sepa-
(bar)
_+20 (4)
15.7_+ 0.5 (2)
Half-time of water Hydraulic conducexchange, T i n tivity, Lp- 10 6 (s) (cm .s- 1. bar- 1) 8.7+ - 2.1 (4)
_+35 (2)
48
-+12 (2)
5 _+ I (3) _+ 9(2)
2.3_+ 0.6 (2) 1.1
-
rated bladder cells of the lower epidermis were 18 to 166 bar and, therefore, similar to that of other higher plant cells, e.g., of the epidermal bladder cells of Mesembryanthemum crystallinum (Steudle et al. 1977). The considerable difference in e between the two types of cells cannot be accounted for by uncertainties in determining the cell volume which should be 50% at most (see error discussion in Steudle et al. 1980; Tomos et al. 1981), but should reflect structural differences in the cell wall (see Discussion). Like the bladder cells of Mesembryanthemum (Steudle et al. 1977) and other higher plant cells, the cells from the lower epidermis displayed a dependence of e on cell turgor, as demonstrated for some cells for which e could be measured in two different pressure ranges (Table 1). Furthermore, it was found that e of these cells depended on cell volume or cell diameter (Table 1, Fig. 3). No dependence of e on cell volume was detectable for the cells in the upper epidermis. Cell turgor could not be varied substantially in these cells with the aid of the pressure probe, since the cells tended to become leaky upon large variations (see above)
42
E. Steudle et al. : Water relations of Oxalis bladder cells
~
.
O.13_O_O- 13.0- 0-0-0-13__0_0-0O~U~
o~o~'* ~.,. ~kO"8 " "NR~' " ~N-'*~~176174 LOr',. u3 ~ ~ u'~r,4-4- t--- eOk~
Upper epidermis
_
.4. ,d3 -.~t I
'
/1/
~
.
f
I
I
I
I
I
I
I
Tll2=/.4.7 E = /,.2bar Lp = 5.8.10 .6 c m . s - l . b e r -1
.13
E = 5.15bor Lp = 9.7.10-6cm.s-l.bar -1
rl
g
I
I
I
I
I
l
I
I
0
60
120
180
2/,0
300
360
/.20
Time, t [s]
if} if} (3_
Tl/2 = 13.9s
O I:J~6
Lower
epidermis
Q. O. O. 13_ 0.13.
(3- O-
1:3- O- Q- t'~ s
r~CL
ed ui~Sm oie5 ,-~ ~6 ...t'c4Gdt-.-u5 ~ ~
. .lO-Scm.s-l.bor -1
r--t-.m~o~m-e
eqm-.~-.t.~
/,
2 1
0
E = 93.7 bor Lp= 2.8-10-Scm.s -1. bar -1 I.
I
0
60
I
120
/
180
I
I
I
I
2/,0
300
360
/,20
- -
Time, t [s] Fig. 2. Pressure relaxation curves of two bladder cells of the upper and lower epidermis of an Oxalis carnosa leaf. Turgor was changed nearly instantaneously with the aid of the pressure probe to produce inwardly or outwardly directed water flows. From the rate constant (half-time) of the relaxations Lp is calculated. Nearly instantaneous changes of cell turgor (AP) in response to changes of cell volume (AV in picoliters) yield the volumetric elastic modulus ( e = V . A P / A V ) of the cells (right-hand part of the traces)
so that it was not possible to look for changes of e with turgor. The missing volume (size) dependence of e of the cells may be a consequence of the different structural properties of the cell walls resulting in only small differences in e at normal cell turgor. Furthermore, differences with the lower bladder cells may occur because the upper epidermal cells are closely packed and e may not only reflect the mechanical properties of individual cells but also those of the surrounding tissue. Thus, a volume effect, as found for the widely spaced bladders in the lower epidermis, could have been evened out. It should be stressed that differences in the elastic properties of the cells become significant when measuring the responses of cell turgor to changes in cell volume (see Fig. 2). In order to produce the same AP, in the upper epidermis, a much larger AV was necessary as compared with the lower epidermis. In most cases e was smaller than rci ( = 7 . 3 bar) or similar in the upper epidermis and, according to eqn. (1) changes in e should have had a much smaller influence on the half-time of water exchange between bladder cell and adjacent tissue
as compared with the bladders in the lower epidermis. Thus, a significant asymmetry in 7ri (see above) between both epidermal layers may have some influence on the evaluation of Lp, but this should not change the general picture of the water relations of the epidermal bladder cells of O. c a r n o s a described here. Discussion
The water-relations parameters of the bladder cells of the O. c a r n o s a leaf show significant differences when comparing the upper and lower epidermis and these differences are most obvious for s and Lp. Some caution is needed with respect to the Lp values because of the uncertainties in determining the exchange area for water flow. It has been assumed that in the lower epidermis water is exchanged only via the (circular) base of the cells, whereas in the upper epidermis it may not only flow across the base but also be exchanged laterally between bladder cells. This latter assumption seems to be reasonable, but, in principle, there may be a difference in the Lp values for the two path-
E. Steudle et al. : W a t e r r e l a t i o n s o f Oxalis b l a d d e r cells
_Up_perepidermis
20
X
L
o
.I3
x
l.lJ
x
x x
'~ 10 "13
X
X
X
0
X x
E
x
X
._
4-' I/]
X
Xx X x
X
LLJ
xx
X
X
I
i
5
10
Cell volume, V [nil
Lower epidermis
200 0 .13 w
x X
-~ 100
X
0
E
X
X
XX
X
X xx x
U ..,,.=,
u~
X
LU ,x 0
I
I
10
I
I
~__
20
Cell volume, V[nl] Fig. 3. Effect of cell volume (V) on the elastic modulus (e) for bladder cells of the lower and upper epidermis of Oxalis carnosa. For the upper epidermis there is no trend for e to change with cell volume, whereas for the lower epidermis e increases with increasing V. Since e of the bladder cells of the lower epidermis is a function of cell turgor, in the lower part of the figure e values are given only for P> 1 bar
ways. At present, we cannot exclude this possibility and are uncertain of using the correct values of the exchange areas, but nevertheless, we can state that the bladders on the twe epidermal layers should be different with respect to their hydraulic conductivity (water permeability). If it turns out that there are big differences in Lp, this would point also to differences in the structure and-or composition of the cell membranes. It should be possible that the exchange area (A) may be more accurately measured using the charge-pulse technique (Zimmermann et al. 1981). The method has been used to determine the membrane area of iso-
43
lated algal cells and has resulted in a good agreement between geometrically and electrically determined surface areas for Valonia utricularis. The physiological importance of the asymmetry in e and Lp is not known. The large differences in e should have consequences for both the mechanics of the entire leaf and its water and solute storage properties. Since the upper epidermis is rather thick (as compared with the leaf thickness; Fig. 1) and closely packed with cells, its mechanical rigidity would be rather high if e of the cells were high. In this case the upper epidermis would govern the mechanics of the leaf and this could result in shear stresses between the two epidermal layers. However, since e of the upper epidermal cells is low, the extensibility of the upper layer may be similar to that of the lower epidermis which contains not only the big bladders (having high e), but also cells which are much smaller. The result that e of the bladder cells in the lower epidermis is a function of cell turgor as well as of cell volume is similar to the pressure and volume dependence of e found for some giantcelled algae (Zimmermann and Steudle 1978, 1980) and also for the bladder cells of M e s e m b r y a n t h e m u m . The dependence of e on the volume (size) of these cells was more pronounced at high rather than at low turgor. This effect has been explained (Zimmermann and Steudle 1978, 1980) in terms of a strong dependence of the elastic moduli of the Cell wall (Young's moduli, Poisson ratios) on the tension in the cell wall, which is proportional to cell diameter in cylindrical and spherical cells. Another explanation could be that plant cells may consist of areas of different elastic extensibility and that the relative amount of the less extensible parts of a cell increases as the cell becomes larger (Zimmermann and Steudle 1975; Steudle et al. 1977). The storage properties of the bladders of the epidermal layers should be quite different. For water, the storage capacitance of a cell (Co) is given by the amount of water which is taken up by the cell from the surroundings at a change of the water dV V potential (gt) of one bar, i.e. C c - d v '
E -}-TCi
(Dainty 1976; Zimmermann and Steudle 1978). It should be noted that this storage capacitance is different from the term "succulence" which is normally (and not correctly) taken as a quantitative measure for the ability of cells to store water (Steudle et al. 1980). If the bladder cells were mainly used as water stores, then the upper epidermis would be much more efficient. However, it does not seem to be reasonable that the main function of the bladders of O. carnosa is that of a water
44
store enabling the plant to survive under conditions of water stress, because in its natural habitat the plant is hardly subjected to a shortage in water. The situation for Oxalis, a genus with the C3-type of photosynthetic CO z fixation (data not shown), seems to be quite different from that of the potential Crassulacean acid metabolism-plant Mesembryanthemum crystallinum for which it has been postulated that the epidermal bladders function as a peripheral water reservoir protecting the plant from short-term stress (Steudle et al. 1975, 1977). The fact that the bladders contain large amounts of oxalic acid indicates that this solute is stored in the bladders either to protect the plant against being eaten by animals or to get rid of the acid produced metabolically. If this were true, the bladders would have a function comparable to the salt glands of Limonium (Ziegler and Liittge 1966, 1967; Hill and Hill 1976) or to the bladders of Atriplex and Chenopodium (Lfittge 1975; Hill and Hill 1976). However, this suggestion is at present purely speculative, because nothing is known about the distribution of oxalic acid in the leaf except that the pH of Oxalis leaf phloem sap, collected as aphid exudate and measured using pHindicator paper, is about 7 pH units less acid than the bladder (leaf) sap. It cannot be concluded from the experiments that there is an accumulation of certain solutes in the bladders, but because of the low e of the upper epidermis these cells would allow large amounts of solutes to be stored without changing the water potential and cell turgor of the bladders too much. If this hypothesis is true, the reason why the bladders of the lower epidermis have to have a much higher e is evident. In the lower epidermis, a very close packing of bladder cells would reduce the gas exchange between stomata and atmosphere and this is prevented by the low extensibility of the cell walls, an extensibility which decreases with increasing cell volume. Considering the fact that the bladders constitute about 75% of the leaf volume, the water relations of the leaf should be mainly determined by the water relations of the bladders. If we assume that the mesophyll cells (which have not been measured until now) have a similar Lp, the half-time of water exchange of these cells (because of their small V/A ratio) would be much smaller, so that changes in the water potential should be quickly propagated across the mesophyll. We may expect that the response of the entire leaf to changes in water potential (soil, atmosphere) should be similar to that of the bladders. In fact, preliminary results show that a half-time of this order is observed for the drop in turgor in lower epidermal
E. Steudle et al. : Water relations of Oxalis bladder cells
bladders when the leaf is cut off the plant. Further experiments are needed to show whether the bladders of both layers react in the same way. These experiments should include measurements of the water-relations parameters of mesophyll cells which are somewhat difficult to determine because the microcapillary has to be inserted into rather small cells across the big bladder cells. However, it should be worthwhile to attempt these measurements because the leaf of O. carnosa provides a system in which the macroscopic water status of a leaf can be changed and also the response at different sites of the leaf can be measured in order to obtain more information about water transport in leaves. We are indepted to Dr. Margit Altendorfer and Dr. J. Froh, Technische Universit/it Mfinchen, for their help in scanning electron microscopy. We thank J. Zillikens, G. B61ing and H. J/ickel for skillful technical assistance and Drs. M. Arnold and S.D. Tyerman, KFA Jfilich, for reading the manuscript. This work was supported by the Deutsche Forschungsgemeinschaft, Zi 99/8 and Zi 23/48.
References Dainty, J. (1976) Water relations of plant cells. In: Encyclopedia of plant physiology, N. S., vol. 2: Transport in plants II, pt. A: Cells, pp. 12-35, Lfittge, U., Pitman, M.G., eds. Springer, Berlin Heidelberg New York Hill, A.E., Hill, B.S. (1976) Elimination processes by glands: mineral ions. In: Encyclopedia of plant physiology, N. S., vol. 2: Transport in plants II, pt. B: Tissues and organs, pp. 225-243, Lfittge, U., Pitman, M.G., eds. Springer, Berlin Heidelberg New York Liittge, U. (1975) Salt glands. In: Ion transport in plant cells and tissues, pp. 335-376, Baker, D.A., Hall, J.L., eds. North-Holland Publ. Co., Amsterdam Lfittge, U., Fischer, E., Steudle, E. (1978) Membrane potentials and salt distribution in epidermal bladders and photosynthetic tissue of Mesembryanthemum crystallinum L. Plant Cell Environ. 1, 121-129 Steudle, E., Liittge, U., Zimmermann, U. (1975) Water relations of the epidermal bladder cells of the halophytic species Mesembryanthemum crystallinum: direct measurement of hydrostatic pressure and hydraulic conductivity. Planta 126, 229-246 Steudle, E., Smith, J.A.C., Liittge, U. (1980) Water relation parameters of individual mesophylI cells of the CAM plant Kalancho6 daigremontiana. Plant Physiol. 66, 1155-1163 Steudle, E., Zimmermann, U., Lfittge, U. (1977) Effect of turgor pressure and cell size on the wall elasticity of plant cells. Plant Physiol. 59, 285-289 Steudle, E., Zimmermann, U. Zillikens, J. (1982) Effect of cell turgor on hydraulic conductivity and elastic modulus of Elodea leaf cells. Planta 154, 371-380 Tomos, A.D., Steudle, E., Zimmermann, U., Schulze, E.-D. (1981) Water relations of leaf epidermal cells of Tradescantia virginiana. Plant Physiol. 68, 1135-1143 Tyerman, S.D., Steudle, E. (1982) Comparison between osmotic and hydrostatic water flows in a higher plant cell: determi-
E. Steudle et al. : Water relations of Oxalis bladder cells nation of hydraulic conductivities and reflection coefficients in isolated epidermis of Tradescamia virginiana. Aust. J. Plant Physiol. 9, 461-480 Ziegler, H., Liittge, U. (1966) Die Salzdr/isen von Limonium vulgare. I. Die Feinstruktur. Planta 70, 193 206 Ziegler, H., Lfittge, U. (1967) Die Salzdr/isen von Limonium vulgare. II. Die Lokalisierung des Chlorids. Planta 74, I 17 Zimmermann, U., Benz, R., Koch, H. (1981) A new electrical method for the determination of the cell membrane area in plant cells. Planta 152, 352-355 Zimmermann, U., Hiisken, D. (1980) Turgor pressure and cell volume relaxation in Halicystis parvula. J. Membr. Biol. 56, 55-64
45 Zimmermann, U., Steudle, E. (I 975) The hydraulic conductivity and volumetric elastic modulus of cells and isolated cell walls of Nitella and Chara spp. : pressure and volume effects. Aust. J. Plant Physiol. 2, 1-12 Zimmermann, U., Steudle, E. (1978) Physical aspects of water relations of plant cells. Adv. Bot. Res. 6, 45-117 Zimmermann, U., Steudle, E. (1980) Fundamental water relations parameters. In: Plant membrane transport: current conceptual issues, pp. 113-127, Spanswick, R.M., Lucas, W.J., Dainty, J., eds. Elsevier/North-Holland Biomedical Press, Amsterdam Received 26 February; accepted 18 April 1983
Note added in proof: It should be mentioned that while this work was in progress, large differences in Lp (as for the two different epidermal layers of Oxalis) have also been found for the fruit of Capsicum annuum. For the giant subepidermal bladder cells at the inner pericarp of this species Rygol and Lfittge (1983) reported an Lp value of 6"10 -s cm s -1 bar -1, whereas for the small parenchyma cells the hydraulic conductivity was more than one order of magnitude smaller (Plant, Cell, and Environment 6, in press).
Planta (1983)159:38-45
9 Springer-Verlag 1983
Water relations of the epidermal bladder cells of O x a l i s c a r n o s a Molina E. Steudle 1, H. Ziegler 2 and U. Zimmermann 1 1Arbeitsgruppe Membranforschung am Institut ffir Medizin, Kernforschungsanlage J/jlich GmbH, Postfach 1913, D-5170 Jiilich, and z Institut f/Jr Botanik und Mikrobiologie der Technischen Universitfit, Arcisstrasse 21, D-8000 Miinchen, Federal Republic of Germany
Abstract. All of the cells of the upper (adaxial) epidermis of the leaves of Oxalis carnosa are transformed into large bladders, while in the lower epidermis the bladder cells are interrupted by "normal" cells with stomata. The epidermal bladders contain a high concentration of free oxalic acid (pH approx. 1). Water-relations parameters of these epidermal bladder cells have been determined using the pressure probe. Original cell turgor (Po) of the closely packed bladders of the upper epidermis was Po=0.7 to 2.9bar (Po=l.7_+0.5bar; mean_+ SD; N = 25 cells) and lower than that in the club-shaped bladders of the lower epidermis (P0 =1.3 to 3.7 bar; Po=2.5_+0.7 bar; N = 25 cells). Large differences in the elastic modulus (e) and the hydraulic conductivity (Lp) of the two different types of cells were observed. For the lower epidermal bladders, e = 1 8 to 166 bar and was similar to that of other higher plant cells. Also, for these cells it was found that e was increasing with both, cell turgor and cell volume. By contrast, e of the cells of the upper epidermis was by one order of magnitude smaller (e=1.9 to 17.0 bar) and no dependence of e on cell volume could be detected. The Lp values of the cell membranes were also different (lower epidermis" Lp=(2_.3_+1.6). 10- 5 cm s- 1 b a r - 1 ; upper epidermis: Lp = (3.8 _+ 2 . 4 ) . ]0 -6 c m s - 1 b a r - l ) . These differences seem to be too large to be caused by errors in determining the exchange area for water (A) between cells and adjacent tissue. The half-times of water exchange between bladders and leaf (T1/2) were, on average, somewhat longer for the upper than for the lower epidermis (lower epidermis: T1/2 = 7 to Abbreviations: P=cell turgor; Lp=hydraulic conductivity of cell membrane; T1/2 = half-time of water exchange between cell and surroundings; e=volumetric elastic modulus; A = e x change area for water of cell; V = cell volume; ~i = osmotic pressure of the cell sap
38 s; upper epidermis: T1/2=22 to 213 s), but the differences in the T1/2 values were not as distinct as for e and Lp. This is because of the compensatory effects of e, Lp and the different ratios of volume to exchange area. Since the bladders make up about 75% of the entire volume of the leaf, it is assumed that the rate of response of the leaf to changes in the water potential should be similar to that of the bladder cells. The results are discussed in terms of a possible function of the bladders in the leaf. Key words: Bladder cells - Elastic modulus - Hydraulic conductivity - Oxalic acid - Oxalis - Water transport.
Introduction
Epidermal bladder cells play an important role in both the elimination of salts from metabolically active tissue and in water storage of different plant species. For example, the large bladder cells of the halophilic Chenopodiaceae such as Atriplex spongiosa and Chenopodium album seem to be involved in the active removal of salts from the underlying tissue (for reviews see Lfittge 1975; Hill and Hill 1976), whereas the giant epidermal bladder cells of Mesembryanthemum crystallinum do not accumulate salts (Ltittge et al. 1978) but rather function as peripheral water reservoirs. They protect the plant from short-term water stress and are thus also involved in salinity adaptation (Steudle et al. 1975). Both possible functions of the bladders are related to the water-relations parameters of these cells, i.e. to the water permeability (hydraulic conductivity) of these cells and to their elastic extensibility (elastic modulus of the cell wall). A reservoir
E. Steudle et al. : Water relations of Oxalis bladder cells
39
or depot cell has to be extensible enough to provide a certain storage capacitance and a sufficiently quick transfer of the stored water or solutes across the cell membrane. In this communication we have investigated water-relations parameters of the large leaf bladder cells of the South-American species Oxalis' carnosa using the pressure-probe technique (Zimmermann and Steudle 1978, 1980). Material and methods Material. Plants of Oxalis' carnosa Molina, a species native to Chile, Peru and South Bolivia, were grown from seeds in potting soil in a climatic chamber (temperature: 22 ~ C (day); 19 ~ C 9 (night); light intensity: 170 pmol photons m 2 s-1; light/dark rhythm: 12 h/12 h; humidity: 60/85%). Plants used in the experiments were 6-9 months old and 10-20 cm high. They had already flowered once (usually 4-6 months after sowing). The plants were well watered (2-3 times per week), but were prevented from being water-logged. Scanning electron microscopy. Plant material was shock-frozen in melting freon 11 and freeze-dried on metal plates without intermediate thawing. The plates were then coated with goldpalladium and examined in a JSM 35C scanning electron microscope (Jeol, Fa. Kontron, Eching). Determination of water relations parameters. For the experiments with the pressure probe, one leaflet of the trifoliate leaf (still attached to the plant) was mounted, with either the upper or lower epidermis facing upwards, in a small chamber on a perspex block and was held by means of a cover glass which was clamped on the leaf. The cover slip had a hole (diameter: about 5 mm) through which the pressure probe could be introduced into the epidermal cells. The experiments were performed with the leaf in air. The determination of water relations parameters (half-time of water exchange between bladder cell and adjacent tissue, T1/2; volumetric elastic modulus, e; hydraulic conductivity of cell membrane, Lp) was performed as described in previous papers and reviews (e.g. Zimmermann and Steudle 1978, 1980; Tomos et al. 1981; Steudle et al. 1982; Tyerman and Steudle 1982). The upper (adaxial) epidermal cells of O. carnosa were sometimes (and mainly in the beginning of our experiments) difficult to measure, since these cells showed a tendency to become leaky. This is probably because they have a very high elastic extensibility (low e) which tends to make the seal between the cell and microcapillary leaky. Such leaks could easily be detected by a drop in cell turgor and the measurements on these ceils were discarded. In order to evaluate Lp from T1/2 and e, the osmotic pressure of the cell sap (~i) and the volume (V) to surface area (A) ratio of the cell must be known, since: V ln2 Lp = X T1/~
1 e+~r i"
(1)
~i was determined by deep freezing leaves of different plants for I h in liquid nitrogen. After thawing, cell sap was gained by centrifuging the tissue through a filter paper. For four different plants the extracted sap had an osmolarity of c~= 300 • 19 mosmol (mean i SD ; N = 4) which is equivalent to an osmotic pressure of ~i=7.32• bar (temperature=20 ~ C). The cell sap showed a very low pH of 1-1.5 caused by a high content of oxalic acid (Pkl = 1.23; Pk2 = 4.19). About one half or 153.0 • 13.6 mM of the total osmolarity of
the cell sap was oxalic acid. Oxalic acid was determined by precipitation with Ca 2 § separating the Ca-oxalate and re-dissolving it in dilute H2SO ~. Oxalic acid was titrated with 0.1 N KMnO4.
Results
Cell geometry and exchange area f o r water flow. Cell geometry and cell dimensions were determined with the aid of the microscope (by looking at surface views and at cross sections of the leaves) as well as by the use of scanning electron micrographs (Fig. 1). In most cases, the cells of the upper (adaxial) epidermis were approximated either as a prism with a hexagonal cross section or as a cylinder. Sometimes (mainly for cells of small volumes) the shape was similar to that of a rectangular prism. The dimensions of the cross section were determined for each individual cell from the surface view (see Fig. 1 a), whereas for the height (h), an average value, which was h=163.5 • gm (mean + SD; N = 39 cells), was determined from cross sections of different leaves. Since the bladder cells of the upper epidermis are closely packed, the exchange area for water of these cells was calculated from the sum of the area of the anticlinal cell walls plus the area of the base of the cells. For the cells of the upper epidermis, the exchange area, A, ranged between 4.5 and 2 3 . 0 . 1 0 - 4 c m 2 and the cell volume, V, between 0.8 and 11.9 nl. In the lower epidermis, the bladder cells are not closely packed since there has to be space for stomata complexes (Fig. i b). The shape of the bladders was approximated by spherical segments or by a truncated cone with a half of a sphere attached to the larger (end) base of the cone. The diameter of the sphere (2 R) was determined from the surface view of the leaf using a stereomicroscope. The diameter of the circular base of the epidermal bladders (2r) as well as the height of the cells (h) could be determined by cutting the leaf transversely after the measurement, turning it by 90 ~ and looking at the cells from the side. With these data, the volume and the exchange area of the cell for water (i.e. the area of the circular base) could be calculated. For the cells investigated in this paper 2 R = 1 2 0 - 3 2 5 ~m, 2 r = 7 0 - 2 8 5 gm, which resulted in values of the cell volume of V = 0.9 to 18.1 nl and of the exchange area of A = 0 . 4 to 2.8- 10-4 cm 2. Thus, the V/A ratio of the bladder cells in the lower epidermis was larger than that of the ceils in the upper epidermis. Water relations. Cell turgor (Po) in the bladder cells of both the lower and upper epidermis was similar,
40
E. Steudle et al. : Water relations of Oxalis bladder cells
Fig. 1 a-c. Scanning electron micrographs of the upper (a) and
lower (b) surface and of a fractured segment (c) of an carnosa leaf
although there was a trend for Po to be slightly lower in the upper epidermis and this is perhaps related to the fact that e in the upper epidermis is much lower than that of the lower epidermis (see below). In the lower epidermis Po ranged between 1.3 and 3.7 bar (Po=2.5_+0.7bar; mean _+SD ( N = 25 cells)), whereas in the upper epidermis P 0 = 0 . 7 to 2 . 9 b a r ( P 0 = l . 7 _ + 0 . 5 b a r ; N = 25 cells). The osmotic pressures of lower and upper epidermal cells could not be determined separately (as already mentioned in Material and methods), but only an average value for the whole leaf (including the mesophyll) could be obtained. This value of ~zi = 7.3 bar should represent a reasonable estimate of the mean osmotic pressure of the bladder cells, because the bladders make up about 75% of the total leaf volume. However, an asymmetry in rci which may be indicated by the differences in Po and other water-relations parameters (see below), may have been overlooked by using this simple method. At present, no reliable method seems to be available for determining the osmotic pressure of amounts of cell sap of the order of picoliters. This would be necessary to determine rc~ of
Oxalis
the two different epidermal layers or even of individual cells. The water-relations parameters of the bladder cells of the lower and upper epidermis showed significant differences (Table 1). On average, the halftime of water exchange between bladder cells and their surroundings (T1/2) was larger in the upper than in the lower epidermis (Table 1, Fig. 2). According to equation (1) this difference may have various causes. Since e of the bladders of the lower epidermis is about one order of magnitude larger than that of the upper epidermis, considerable differences in T~/2 are expected. However, the effect of the elastic modulus on TI/2 should be largely offset by the fact that the ratio of volume to exchange area (V/A) for the lower epidermis is much larger than that for the upper epidermis (V/A = 2 to 1 5 " 1 0 - 2 c m for the lower and 1.5 to 5.2.10-3 cm for the upper bladders). Furthermore, the hydraulic conductivity of the bladders of both epidermal layers appears to be significantly different. The Lp-values for the lower bladders ranged from 0.5 to 6.7.10 -5 cm s -1 bar -1 (L-p=(2.3_+ 1.6)'10 -5 cm s -1 b a r - l ; N = 2 2 cells), whereas Lp for the upper bladders was L p = 0 . 5 to
E. Steudle et al. : Water relations of Oxalis bladder cells
41
Table l. Typical values of the volumetric elastic modulus (e), half-time of water exchange (Tin) and hydraulic conductivity (Lp) of epidermal bladder cells of Oxalis carnosa leaves. The cells shown cover the total ranges of cell volumes investigated. In the lower epidermis, e is a function of cell turgor and volume, whereas in the upper epidermis no dependence on cell volume could be detected. Note the differences in T1/2 and Lp between upper and lower epidermis. Mean values are given _+ SD with the number of experiments in brackets Cell No.
Lower epidermis
Upper epidermis
Cell V/A x 103 volume, V (nl) (cm)
Pressure range for e (bar)
e
2
0.89
23
0.6-2.6 1.8-4.0
29 _+33 (2) 34 -+
11
5.6
18
1.6-3.5 2.1~4.4
50 89
35.0_+ 5.6(2)
i6
8.1
33
2.0-2.5 2.2 3.2
32 89
28.0_+11.5(2)
23
21
10.5
38
0.2-1.5 1.1 2.7
30 104
47.6
15
25
18.1
150
1.3-5.3
158
26.3-+ 9.1 (2)
26 _+12 (2)
2
1.5
1.5
1.5 2.4
1.9_+ 1.1 (2)
65
_+ 1 (2)
1.7_+ 0.0(2)
3
2.4
2.5
0.7-1.8
9.0_+ 5.3 (2)
54 _+15 (2)
1.3_+ 0.3 (2)
10
4.1
2.9
0.8 1.8
6.4_+ 0.9 (2)
30 _+13 (2)
5.4_+ 2.2 (2)
15
6.0
4.1
1.2-2.0
15.1
29 _+ 4 (2)
4.4_+ 0.6 (2)
16
6.1
4.1
1.7-2.9
3.8
119
20
7.4
4.4
2.9-3.5
5.3
213
25
11.9
5.2
0.0-2.3
8 . 0 . 1 0 - 6 c m s-1 bar-1 ( L p = 3 . 8 + 2 . 4 ) . 1 0 - 6 c m s -1 bar 1; N=18cells), i.e. the average values were almost one order of magnitude different. The Lp of the upper epidermal cells was similar to most of the values reported in the literature for higherplant cells (see reviews by Zimmermann and Steudle 1978, 1980; Dainty 1976), but the Lp for the lower epidermis was larger. An asymmetry of water-relations parameters between bladders in the two epidermal layers has been also found with respect to the elastic modulus of the cells. For the closely packed cells in the upper epidermis e ranged between 2 and 17 bar and is much smaller than the e values found for other higher plant cells which are of the order of 10-100 bar at normal cell turgor (Zimmermann and Steudle 1978, 1980; Dainty 1976). The absolute values of e of the upper epidermal bladders were of the same order as the values given for the giant-celled alga Halicystis parvula which were 0.5 to 2 bar under stationary (long-term) conditions and 1 to 16 bar under short-term conditions (Zimmermann and H/]sken 1980). In contrast to the upper epidermis, the values of e in the sepa-
(bar)
_+20 (4)
15.7_+ 0.5 (2)
Half-time of water Hydraulic conducexchange, T i n tivity, Lp- 10 6 (s) (cm .s- 1. bar- 1) 8.7+ - 2.1 (4)
_+35 (2)
48
-+12 (2)
5 _+ I (3) _+ 9(2)
2.3_+ 0.6 (2) 1.1
-
rated bladder cells of the lower epidermis were 18 to 166 bar and, therefore, similar to that of other higher plant cells, e.g., of the epidermal bladder cells of Mesembryanthemum crystallinum (Steudle et al. 1977). The considerable difference in e between the two types of cells cannot be accounted for by uncertainties in determining the cell volume which should be 50% at most (see error discussion in Steudle et al. 1980; Tomos et al. 1981), but should reflect structural differences in the cell wall (see Discussion). Like the bladder cells of Mesembryanthemum (Steudle et al. 1977) and other higher plant cells, the cells from the lower epidermis displayed a dependence of e on cell turgor, as demonstrated for some cells for which e could be measured in two different pressure ranges (Table 1). Furthermore, it was found that e of these cells depended on cell volume or cell diameter (Table 1, Fig. 3). No dependence of e on cell volume was detectable for the cells in the upper epidermis. Cell turgor could not be varied substantially in these cells with the aid of the pressure probe, since the cells tended to become leaky upon large variations (see above)
42
E. Steudle et al. : Water relations of Oxalis bladder cells
~
.
O.13_O_O- 13.0- 0-0-0-13__0_0-0O~U~
o~o~'* ~.,. ~kO"8 " "NR~' " ~N-'*~~176174 LOr',. u3 ~ ~ u'~r,4-4- t--- eOk~
Upper epidermis
_
.4. ,d3 -.~t I
'
/1/
~
.
f
I
I
I
I
I
I
I
Tll2=/.4.7 E = /,.2bar Lp = 5.8.10 .6 c m . s - l . b e r -1
.13
E = 5.15bor Lp = 9.7.10-6cm.s-l.bar -1
rl
g
I
I
I
I
I
l
I
I
0
60
120
180
2/,0
300
360
/.20
Time, t [s]
if} if} (3_
Tl/2 = 13.9s
O I:J~6
Lower
epidermis
Q. O. O. 13_ 0.13.
(3- O-
1:3- O- Q- t'~ s
r~CL
ed ui~Sm oie5 ,-~ ~6 ...t'c4Gdt-.-u5 ~ ~
. .lO-Scm.s-l.bor -1
r--t-.m~o~m-e
eqm-.~-.t.~
/,
2 1
0
E = 93.7 bor Lp= 2.8-10-Scm.s -1. bar -1 I.
I
0
60
I
120
/
180
I
I
I
I
2/,0
300
360
/,20
- -
Time, t [s] Fig. 2. Pressure relaxation curves of two bladder cells of the upper and lower epidermis of an Oxalis carnosa leaf. Turgor was changed nearly instantaneously with the aid of the pressure probe to produce inwardly or outwardly directed water flows. From the rate constant (half-time) of the relaxations Lp is calculated. Nearly instantaneous changes of cell turgor (AP) in response to changes of cell volume (AV in picoliters) yield the volumetric elastic modulus ( e = V . A P / A V ) of the cells (right-hand part of the traces)
so that it was not possible to look for changes of e with turgor. The missing volume (size) dependence of e of the cells may be a consequence of the different structural properties of the cell walls resulting in only small differences in e at normal cell turgor. Furthermore, differences with the lower bladder cells may occur because the upper epidermal cells are closely packed and e may not only reflect the mechanical properties of individual cells but also those of the surrounding tissue. Thus, a volume effect, as found for the widely spaced bladders in the lower epidermis, could have been evened out. It should be stressed that differences in the elastic properties of the cells become significant when measuring the responses of cell turgor to changes in cell volume (see Fig. 2). In order to produce the same AP, in the upper epidermis, a much larger AV was necessary as compared with the lower epidermis. In most cases e was smaller than rci ( = 7 . 3 bar) or similar in the upper epidermis and, according to eqn. (1) changes in e should have had a much smaller influence on the half-time of water exchange between bladder cell and adjacent tissue
as compared with the bladders in the lower epidermis. Thus, a significant asymmetry in 7ri (see above) between both epidermal layers may have some influence on the evaluation of Lp, but this should not change the general picture of the water relations of the epidermal bladder cells of O. c a r n o s a described here. Discussion
The water-relations parameters of the bladder cells of the O. c a r n o s a leaf show significant differences when comparing the upper and lower epidermis and these differences are most obvious for s and Lp. Some caution is needed with respect to the Lp values because of the uncertainties in determining the exchange area for water flow. It has been assumed that in the lower epidermis water is exchanged only via the (circular) base of the cells, whereas in the upper epidermis it may not only flow across the base but also be exchanged laterally between bladder cells. This latter assumption seems to be reasonable, but, in principle, there may be a difference in the Lp values for the two path-
E. Steudle et al. : W a t e r r e l a t i o n s o f Oxalis b l a d d e r cells
_Up_perepidermis
20
X
L
o
.I3
x
l.lJ
x
x x
'~ 10 "13
X
X
X
0
X x
E
x
X
._
4-' I/]
X
Xx X x
X
LLJ
xx
X
X
I
i
5
10
Cell volume, V [nil
Lower epidermis
200 0 .13 w
x X
-~ 100
X
0
E
X
X
XX
X
X xx x
U ..,,.=,
u~
X
LU ,x 0
I
I
10
I
I
~__
20
Cell volume, V[nl] Fig. 3. Effect of cell volume (V) on the elastic modulus (e) for bladder cells of the lower and upper epidermis of Oxalis carnosa. For the upper epidermis there is no trend for e to change with cell volume, whereas for the lower epidermis e increases with increasing V. Since e of the bladder cells of the lower epidermis is a function of cell turgor, in the lower part of the figure e values are given only for P> 1 bar
ways. At present, we cannot exclude this possibility and are uncertain of using the correct values of the exchange areas, but nevertheless, we can state that the bladders on the twe epidermal layers should be different with respect to their hydraulic conductivity (water permeability). If it turns out that there are big differences in Lp, this would point also to differences in the structure and-or composition of the cell membranes. It should be possible that the exchange area (A) may be more accurately measured using the charge-pulse technique (Zimmermann et al. 1981). The method has been used to determine the membrane area of iso-
43
lated algal cells and has resulted in a good agreement between geometrically and electrically determined surface areas for Valonia utricularis. The physiological importance of the asymmetry in e and Lp is not known. The large differences in e should have consequences for both the mechanics of the entire leaf and its water and solute storage properties. Since the upper epidermis is rather thick (as compared with the leaf thickness; Fig. 1) and closely packed with cells, its mechanical rigidity would be rather high if e of the cells were high. In this case the upper epidermis would govern the mechanics of the leaf and this could result in shear stresses between the two epidermal layers. However, since e of the upper epidermal cells is low, the extensibility of the upper layer may be similar to that of the lower epidermis which contains not only the big bladders (having high e), but also cells which are much smaller. The result that e of the bladder cells in the lower epidermis is a function of cell turgor as well as of cell volume is similar to the pressure and volume dependence of e found for some giantcelled algae (Zimmermann and Steudle 1978, 1980) and also for the bladder cells of M e s e m b r y a n t h e m u m . The dependence of e on the volume (size) of these cells was more pronounced at high rather than at low turgor. This effect has been explained (Zimmermann and Steudle 1978, 1980) in terms of a strong dependence of the elastic moduli of the Cell wall (Young's moduli, Poisson ratios) on the tension in the cell wall, which is proportional to cell diameter in cylindrical and spherical cells. Another explanation could be that plant cells may consist of areas of different elastic extensibility and that the relative amount of the less extensible parts of a cell increases as the cell becomes larger (Zimmermann and Steudle 1975; Steudle et al. 1977). The storage properties of the bladders of the epidermal layers should be quite different. For water, the storage capacitance of a cell (Co) is given by the amount of water which is taken up by the cell from the surroundings at a change of the water dV V potential (gt) of one bar, i.e. C c - d v '
E -}-TCi
(Dainty 1976; Zimmermann and Steudle 1978). It should be noted that this storage capacitance is different from the term "succulence" which is normally (and not correctly) taken as a quantitative measure for the ability of cells to store water (Steudle et al. 1980). If the bladder cells were mainly used as water stores, then the upper epidermis would be much more efficient. However, it does not seem to be reasonable that the main function of the bladders of O. carnosa is that of a water
44
store enabling the plant to survive under conditions of water stress, because in its natural habitat the plant is hardly subjected to a shortage in water. The situation for Oxalis, a genus with the C3-type of photosynthetic CO z fixation (data not shown), seems to be quite different from that of the potential Crassulacean acid metabolism-plant Mesembryanthemum crystallinum for which it has been postulated that the epidermal bladders function as a peripheral water reservoir protecting the plant from short-term stress (Steudle et al. 1975, 1977). The fact that the bladders contain large amounts of oxalic acid indicates that this solute is stored in the bladders either to protect the plant against being eaten by animals or to get rid of the acid produced metabolically. If this were true, the bladders would have a function comparable to the salt glands of Limonium (Ziegler and Liittge 1966, 1967; Hill and Hill 1976) or to the bladders of Atriplex and Chenopodium (Lfittge 1975; Hill and Hill 1976). However, this suggestion is at present purely speculative, because nothing is known about the distribution of oxalic acid in the leaf except that the pH of Oxalis leaf phloem sap, collected as aphid exudate and measured using pHindicator paper, is about 7 pH units less acid than the bladder (leaf) sap. It cannot be concluded from the experiments that there is an accumulation of certain solutes in the bladders, but because of the low e of the upper epidermis these cells would allow large amounts of solutes to be stored without changing the water potential and cell turgor of the bladders too much. If this hypothesis is true, the reason why the bladders of the lower epidermis have to have a much higher e is evident. In the lower epidermis, a very close packing of bladder cells would reduce the gas exchange between stomata and atmosphere and this is prevented by the low extensibility of the cell walls, an extensibility which decreases with increasing cell volume. Considering the fact that the bladders constitute about 75% of the leaf volume, the water relations of the leaf should be mainly determined by the water relations of the bladders. If we assume that the mesophyll cells (which have not been measured until now) have a similar Lp, the half-time of water exchange of these cells (because of their small V/A ratio) would be much smaller, so that changes in the water potential should be quickly propagated across the mesophyll. We may expect that the response of the entire leaf to changes in water potential (soil, atmosphere) should be similar to that of the bladders. In fact, preliminary results show that a half-time of this order is observed for the drop in turgor in lower epidermal
E. Steudle et al. : Water relations of Oxalis bladder cells
bladders when the leaf is cut off the plant. Further experiments are needed to show whether the bladders of both layers react in the same way. These experiments should include measurements of the water-relations parameters of mesophyll cells which are somewhat difficult to determine because the microcapillary has to be inserted into rather small cells across the big bladder cells. However, it should be worthwhile to attempt these measurements because the leaf of O. carnosa provides a system in which the macroscopic water status of a leaf can be changed and also the response at different sites of the leaf can be measured in order to obtain more information about water transport in leaves. We are indepted to Dr. Margit Altendorfer and Dr. J. Froh, Technische Universit/it Mfinchen, for their help in scanning electron microscopy. We thank J. Zillikens, G. B61ing and H. J/ickel for skillful technical assistance and Drs. M. Arnold and S.D. Tyerman, KFA Jfilich, for reading the manuscript. This work was supported by the Deutsche Forschungsgemeinschaft, Zi 99/8 and Zi 23/48.
References Dainty, J. (1976) Water relations of plant cells. In: Encyclopedia of plant physiology, N. S., vol. 2: Transport in plants II, pt. A: Cells, pp. 12-35, Lfittge, U., Pitman, M.G., eds. Springer, Berlin Heidelberg New York Hill, A.E., Hill, B.S. (1976) Elimination processes by glands: mineral ions. In: Encyclopedia of plant physiology, N. S., vol. 2: Transport in plants II, pt. B: Tissues and organs, pp. 225-243, Lfittge, U., Pitman, M.G., eds. Springer, Berlin Heidelberg New York Liittge, U. (1975) Salt glands. In: Ion transport in plant cells and tissues, pp. 335-376, Baker, D.A., Hall, J.L., eds. North-Holland Publ. Co., Amsterdam Lfittge, U., Fischer, E., Steudle, E. (1978) Membrane potentials and salt distribution in epidermal bladders and photosynthetic tissue of Mesembryanthemum crystallinum L. Plant Cell Environ. 1, 121-129 Steudle, E., Liittge, U., Zimmermann, U. (1975) Water relations of the epidermal bladder cells of the halophytic species Mesembryanthemum crystallinum: direct measurement of hydrostatic pressure and hydraulic conductivity. Planta 126, 229-246 Steudle, E., Smith, J.A.C., Liittge, U. (1980) Water relation parameters of individual mesophylI cells of the CAM plant Kalancho6 daigremontiana. Plant Physiol. 66, 1155-1163 Steudle, E., Zimmermann, U., Lfittge, U. (1977) Effect of turgor pressure and cell size on the wall elasticity of plant cells. Plant Physiol. 59, 285-289 Steudle, E., Zimmermann, U. Zillikens, J. (1982) Effect of cell turgor on hydraulic conductivity and elastic modulus of Elodea leaf cells. Planta 154, 371-380 Tomos, A.D., Steudle, E., Zimmermann, U., Schulze, E.-D. (1981) Water relations of leaf epidermal cells of Tradescantia virginiana. Plant Physiol. 68, 1135-1143 Tyerman, S.D., Steudle, E. (1982) Comparison between osmotic and hydrostatic water flows in a higher plant cell: determi-
E. Steudle et al. : Water relations of Oxalis bladder cells nation of hydraulic conductivities and reflection coefficients in isolated epidermis of Tradescamia virginiana. Aust. J. Plant Physiol. 9, 461-480 Ziegler, H., Liittge, U. (1966) Die Salzdr/isen von Limonium vulgare. I. Die Feinstruktur. Planta 70, 193 206 Ziegler, H., Lfittge, U. (1967) Die Salzdr/isen von Limonium vulgare. II. Die Lokalisierung des Chlorids. Planta 74, I 17 Zimmermann, U., Benz, R., Koch, H. (1981) A new electrical method for the determination of the cell membrane area in plant cells. Planta 152, 352-355 Zimmermann, U., Hiisken, D. (1980) Turgor pressure and cell volume relaxation in Halicystis parvula. J. Membr. Biol. 56, 55-64
45 Zimmermann, U., Steudle, E. (I 975) The hydraulic conductivity and volumetric elastic modulus of cells and isolated cell walls of Nitella and Chara spp. : pressure and volume effects. Aust. J. Plant Physiol. 2, 1-12 Zimmermann, U., Steudle, E. (1978) Physical aspects of water relations of plant cells. Adv. Bot. Res. 6, 45-117 Zimmermann, U., Steudle, E. (1980) Fundamental water relations parameters. In: Plant membrane transport: current conceptual issues, pp. 113-127, Spanswick, R.M., Lucas, W.J., Dainty, J., eds. Elsevier/North-Holland Biomedical Press, Amsterdam Received 26 February; accepted 18 April 1983
Note added in proof: It should be mentioned that while this work was in progress, large differences in Lp (as for the two different epidermal layers of Oxalis) have also been found for the fruit of Capsicum annuum. For the giant subepidermal bladder cells at the inner pericarp of this species Rygol and Lfittge (1983) reported an Lp value of 6"10 -s cm s -1 bar -1, whereas for the small parenchyma cells the hydraulic conductivity was more than one order of magnitude smaller (Plant, Cell, and Environment 6, in press).
