- Email: [email protected]
Abscisic Acid in Leaf Epidermis of Commelina communis L.: Distribution and Correlation with Stomatal Closure
Abscisic Acid in Leaf Epidermis of Commelina communis L.: Distribution and Correlation with Stomatal Closure K. DORFFLING and D. TIETZI) Institut flir Allgemeine Botanik und Botanischer Garten Hamburg, OhnhorststraBe 18, D-2000 Hamburg 52 Received April 25, 1984 . Accepted July 26, 1984
Summary Changes in the abscisic acid (ABA) content of epidermis and mesophyll of isolated Comme· lina communis L. leaves were followed and compared to changes in diffusion resistance during a water stress treatment. Diffusion resistance increased within 15 min before a rise in the ABA content of the epidermis could be determined. By application of 14C-Iabelled ABA to isolated leaves via the transpiration stream it was found that the minimal amount of ABA to induce stomatal closure was in the range of 0.4 fmol per stomatal complex. Isolated epidermis was treated with tritiated ABA and subjected to microautoradiography. The stomatal complexes showed an accumulation of radioactive material. This accumulation was observable in living as well as in non-living, heat- or cold-treated epidermis. The results are discussed in relation to the hypothesis that ABA regulates stomatal closure in water-stressed leaves.
Key words: Abscisic aCId, Commelina communis L., Stomatal closure, Water stress.
Introduction The hypothesis that abscisic acid (ABA) is involved in the regulation of guard cell movements under stress conditions is based on several observations, especially the following: open stomata close rapidly, and closed stomata do not open in the presence of ABA Gones and Mansfield 1970). Water stress not only induces stomatal closure, but also causes a considerable rise in the level of ABA in leaves (Hiron and Wright 1973). Despite considerable experimental evidence for a regulatory role of ABA in stomatal movements under stress conditions several problems have remained unresolved. For example, it is still unclear in which part of the leaf ABA is being synthesized. Most authors suggest that ABA is produced in the mesophyll cells and migrates from there via the apoplast to the epidermis (Mansfield et al. 1978), which seems to be unable to synthesize ABA in response to stress (Loveys 1977; Dorffiing et al. 1980). On the other hand, Weiler et al. (1982) have suggested that the guard cells themselves produce ABA in response to stress. Experimental evidence for transport from the mesophyll to the epidermis is not yet available, although ABA has been I) Present address: National Institute of Health, Bethesda, Maryland, USA.
J. Plant Physiol.
Vol. 117. pp. 297-305 (1985)
298
K. DORFFUNG and D. TIETZ
found in the apoplastic medium (Ackerson 1982). In conflict with the suggestion that ABA is involved in stress-induced stomatal closure is the observation that stomata of water-stressed leaves close before a rise in the ABA content in the total leaf and even in the epidermis can be observed (Beardsell and Cohen 1975; Walton et al. 1977; Bengtson et al. 1977; Dorffiing et al. 1980). However, close correlation between both processes has been observed by Pierce (1981). These inconsistencies may be explained in different ways: 1) It is possible that the variations in amounts of ABA necessary for stomatal closure are so small that they cannot be measured with the available analytical methods. 2) Another possibility is that a redistribution of available ABA in the epidermis resulting in an accumulation in the guard cells (Weyers and Hillman 1979 a) is responsible for the beginning of stomatal closure. Such a redistribution would take place without a change in the total ABA content. In the present study both these possibilities have been investigated. The minimal amounts of ABA applied to induce stomatal closure have been compared with the endogenous ABA levels in epidermis and mesophyll of water-stressed Commelina leaves. Furthermore, the distribution of applied radioactive ABA in isolated epidermis has been studied.
Materials and methods Plant material
Seeds of Commelina communis L., kindly provided by Dr. T. A. Mansfield, University of Lancaster, U.K., were sown in a mixture of peat and garden soil. After cultivation for three to four weeks in a greenhouse with supplementary illumination (fluorescent lamps, Philips TLF 65 W 134; 240 JLE m- 2 sec!) the seedlings had four to six leaves. The upper, fully grown leaves were used for the experiments. Stress treatment
Leaves of well-watered plants were cut off at their base and placed on filter paper under fluorescent lamps at room temperature. At different time intervals diffusion resistance was measured with a Licor diffusion resistance meter Model Li 60. From other leaves of the same set parts of the lower epidermis and the mesophyll were isolated, weighed and immediately frozen in liquid nitrogen. With Commelina leaves it is possible to separate the epidermis from the leaf without adhering mesophyll cells. The guard cells remain intact, but part of the epidermal cells are damaged, especially when under water stress. Quantitative determination ofABA
The isolation and quantitative determination of ABA from Commelina tissue proved to be rather difficult and the method described was modified several times. Similar difficulties were reported by Singh et al. (1979). The frozen mesophyll and epidermal tissue was freezedried and For each determination about so mg dry weight subjected to extraction with ethanol at 4 epidermis and about 200 mg mesophyll were used. The ethanol was removed by vacuum evaporation and the residue dissolved in 1 % NaHC0 3-solution (PH 9). This aqueous fraction was shaken three times against ether (ether phase discarded), then acidified (PH 3) with HCl and again extracted with ether. The acid ether phase was dried, redissolved in ethyl acetate and fur-
0c.
J Plant Physiol.
Vol. 117. pp. 297-305 (1985)
Abscisic acid and stomatal closure
299
ther purified by thin layer chromatography on silica gel Merck HF 254 with two different solvent systems (toluene-ethyl acetate-acetic acid 50 : 30 : 4 and hexane-ethyl acetate 1: 1). The second system was used after the purified ABA fraction had been methylated with diazomethane. Quantitative determination of the methylated ABA, dissolved in isooctane, was performed by gas chromatography on glass columns packed with SE 30 10 % on Chromosorb WAW. In most cases a Packard 429 gas chromatograph equipped with a 63Ni electron capture detector was used. The conditions were: column temperature 190°C, detector temperature 235°C, injector temperature 220°, nitrogen flow rate 50 ml min-I. Losses during the purification p,rocedure were determined by means of 2-'4C-ABA as an internal standard. Low amounts of 2- 4C-ABA which were beyond the limit of gas chromatographic detection, were added to the fresh extract and determined at the end of the procedure by scintillation counting.
Application of radioactive ABA a) Uptake studies Freshly cut leaves were placed upright with their cut ends into distilled water. Four cm from the leaf base a cylindrical HMP humidity sensor (0.5 cm diameter) of a Vaisala hygrometer HMJ 14 was attached to the lower epidermis and the humidity of the epidermal surface registrated continuously. The humidity in the growth chamber was 60 %, the temperature 24°C and the light intensity 250ILE m- 2 sec-I. When steady state conditions had been reached, the solution was changed against buffer solution containing 10- 5 moll-I 2_ 14C_ABA, specific activity 348 MBq mmol- ' , obtained from Amersham. At different time intervals after the beginning of stomatal closure about 1 cm2 epidermis from the leaf area where the stomatal transpiration had been measured was isolated. The radioactivity of the tissue was measured in a mixture of 0.5 ml tissue solubilizer and 3 ml scintillation cocktail (4.3 g diphenyloxazol and 0.43 g phenylenbimethylphenyloxazol in 11 toluene) in a Philips scintillation counter. The mesophyll of this leaf area was scraped off and measured for radioactivity in the same manner. The values obtained were corrected for quenching. b) Distribution studies Pieces of the lower epidermis from freshly cut leaves were isolated and incubated on glass slides in 10 mmoll- ' phosphate buffer pH 6.7 + 100 mmoll- ' NaN0 3 containing 0.32 BMq ml- ' 3H-ABA (Amersham, specific activity 832.5 GBq mmol- ' ) or 3H-indole-acetic acid (Amersham, specific activity 777 GBq mmol- ' ). After 260 min the epidermal pieces were washed twice with phosphate buffer, each time for several seconds. After this the epidermal pieces were frozen in liquid nitrogen, freezedried on glass slides and exposed for two months to an Ilford 18 DIN Pan F film. Other epidermal pieces were treated for five min with high (65°C) or low (-25°C) temperatures or for one min with ethanol at 65°C before being frozen and exposed to radioactive ABA.
Results and discussion Correlation between stress ABA production and stomatal closure
When well-watered leaves are subjected to water stress by cutting them from the plant and placing them under fluorescent tubes, the diffusion resistance increases suddenly after 15 min. This increase indicates stomatal closure (Fig. 1). The ABA level in the epidermis, on the other hand, remains constant for at least 30 min before a remarkable rise takes places and the ABA level reaches fifteen times the original content. On the basis of fresh weight or leaf area the ABA level in the mesophyll of
J. Plant Physiol.
Vol. 117. pp. 297-305 (1985)
K.
300 rs ( reI.)
DORFFLING
and D.
TIETZ
Diffusion Resistance rs
120 80
40 Or-~~--~--~--~~
800
ABA
300
!m\
ABA
J!!.
mg FW
600
0--0
200 400
200
0
J 0
100
2
4h
Fig. 1: Above: Changes in diffusion resistance of the abaxial leaf side of isolated Commelina leaves subjected to water stress. Below: Changes in ABA content in the lower leaf epidermis and mesophyll.
unstressed leaves is higher than in the epidermis. During the stress treatment the ABA level in the mesophyll seems to increase earlier than in the epidermis. The maximal level in this tissue after four hours is lower than in the stressed epidermis. The ABA level measured in unstressed epidermis is in the same order of magnitude as found by Singh et al. (1979) and by Pierce (1981) in Commelina epidermis. Isolated epidermis of Commelina and Vicia does not produce ABA in response to stress (Loveys 1977; D6rffiing et al. 1980). However, stomata close before a rise of ABA in the epidermis can be observed. Therefore, attempts were undertaken to determine the minimal concentration of ABA for the induction of stomatal closure. Minimal amounts of I4CABA for the induction of stomatal closure
Isolated leaves were treated at the base with an aqueous solution containing 14elabelled ABA, and the transpiration of a leaf area (lower side) four cm from the base was measured continuously. Fig. 2 shows that about 16 min (15.7± 1.7 min) after ap-
J. Plant Physiol.
Vol. 117. pp. 297-305 (1985)
Abscisic acid and stomatal closure
o
ISO
301
min
Fig. 2: Changes in transpiration of Commelina leaves after application of ABA 10- 5 mol I-I via the transpiration stream (time zero). Transpiration was measured with a Vaisala hygrometer. The sensor cup (0.5 cm2 i.d.) was applied at the lower half of the leaf four cm from the leaf base. The curves represent four different experiments. plication of 10- 5 mol I-I ABA the humidity outside the lower epidermis declines significantly. This indicates the beginning of stomatal closure. Higher concentrations induced earlier closure: 0.5 10- 4 mol I-I ABA within 12 min (12.2± 1.2 min) and 10- 4 moll- I within 10 min (10.4± 1.3 min). The reduction of transpiration was not continuous, but occurred in oscillations. When 14C_ABA was applied, radioactivity accumulated preferably in the mesophyll and to a much smaller extent in the epidermis (Fig. 3). 19.7 min after application of 10- 5 mol I-I 2)4C-ABA which closely corresponds to the beginning of stomatal closure, 455 pg ABA ( = 0.6 Bq) were present per cm2 epidermis. This is equivalent to 0.39 fmol per stomatal complex. After 30 min the amount of radioactive ABA had increased to 816 pg cm- 2 epidermis or 0.7 fmol per stomatal complex. These amounts are comparable to data found by other authors. Raschke (1975) found that two to four fmol ABA per stomatal complex were necessary to induce 5 % closure in Commelina. Weyers and Hillman (1979 a) calculated that no more than six fmol ABA were present per stomatal complex at the time of closure. In a later paper Weyers and Hillman (1979 b) reported that between 12.6 and 45.4 amol radioactive ABA per complex were present when statistically significant narrowing had occurred. This is less than one tenth of the amount found in the present investigation. In comparision to these data the amounts of endogenous ABA in
J. Plant Physiol.
Vol. 117. pp. 297-305 (1985)
302
K. DORFFLING and D. TIETZ
30
20
10
o L. ...- - -
o
30
eo
10
120 ",in
Fig. 3: Increase of radioactivity in lower epidermis and mesophyll after application of 2_ 14C_ ABA 10- 5 moll-I via the transpiration stream. The area measured was the same as that in Fig. 2. The curves represent the mean of six experiments. Vertical bars: standard deviation. the unstressed epidermis are much lower (about 50 pg cm- 2 or 43 amol per stomatal complex (Fig. 1). An increase of the endogenous epidermal ABA content to the amount of 455 pg cm- 2 which was present after application of ABA at the beginning of stomatal closure (Fig.3) would have been measurable. This calculation is, of course, based on the assumption that applied synthetic ABA is as effective as the endogenous hormone, an assumption which may not be tenable. Moreover, the calculation does not take into consideration that probably only the + - isomer of ABA is active on stomata (Milborrow 1980). Since synthetic ABA is a racemate of + and - forms, the values for applied ABA must be halved. It may be concluded, that either very small, unmeasurable changes in the concentration of the endogenous ABA are responsible for the induction of stomatal closure, or that a redistribution of available ABA within the epidermis occurs and is necessary for stomatal closure.
Distribution of 3H-ABA in isolated epidermis strips Whether unequal distribution of ABA in the Commelina epidermis is possible and whether it is an essential property of the living tissue has been tested by application of tritiated ABA to isolated epidermis and subsequent autoradiography. Fig. 4 shows microautoradiographs from Commelina epidermis which had been incubated in 3H_ ]. Plant Physiol. Vol. 117. pp. 297-305 (1985)
Abscisic acid and stomatal closure
303
Fig. 4: Microautoradiography of epidermal strips incubated in 3H-ABA (right). The accumulation of radioactivity corresponds to the location and number of stomatal complexes (left). Upper half: living epidermis. Lower half: frozen (-25 °C)-thawed epidermis. Incubation medium: P-buffer 10 mmoll- 1, pH 6.7, + 100 mmoll- 1 NaN0 3 + 0.32 MBq 3H-ABA ml- 1• Incubation time 260 min.
ABA. It is evident that radioactive ABA accumulates preferably in the stomatal complexes. This confirms results obtained by Itai et al. (1978) and by Weyers and Hillman (1979 a, b), who suggested that this accumulation of ABA at or near the guard cells "is further evidence that ABA travels during stress from the mesophyll (the site of synthesis) to the guard cell region (the site of action) ... ». However, the accumulation of radioactive material occurs also in non-living epidermis which has been treated either with heat (65°), or cold (-25°C). Even ethanol-treated (65°C) stomatal complexes showed this effect. The accumulation of radioactive material is, moreover, not restricted to ABA. We have observed that 3H-indoleacetic acid accumulates in a similar
J. Plant P}rysiol. Vol. 117. pp. 297-305 (1985)
304
K. DORFFLING and D. TIETZ
manner in stomatal complexes of living and non-living epidermis. The concentration used (3.5 .10- 5 moll-I) had no effect on stomatal pore width. This is in disagreement with observations of Pemadesa (1982). Conclusion The fact that stomata of water-stressed Commelina leaves close before a rise in the ABA content of the epidermis can be observed needs further explanation. It confirms the observations of D6rffiing et al. (1980), but is in contrast to recent data obtained by Pierce (1981). Raschke (1982) has calculated from the data of Pierce (1981) that the epidermal ABA content rose by less than 26 pg ABA cm- 2 when the stomata closed. Such a rise should have been measurable in the present investigation. A rise in the ABA content before the stomata begin to close should also be expected if it is assumed that the stress-ABA is produced in the guard cells themselves, as Weiler et al. (1982) suggested. The data of Weiler et al. (1982), however, also showed that the ABA content in broad bean epidermis rose after the beginning of stomatal closure. An explanation for the inconsistency of the kinetics of ABA increase and stomatal closure would be the redistribution of available ABA in the epidermis resulting in an accumulation of the hormone at or near the guard cells. This could be a first step in the chain of events leading to stomatal closure. Supporting this possibility is the accumulation of radioactive ABA in stomatal complexes. However, non-living epidermis also accumulates applied ABA. The relevance of this process should be investigated further. Theoretically, two other mechanisms may also provide explanations for the inconsistency between stomatal closure and ABA levels: an intracellular rather than an intercellular redistribution, and!or a change in the sensitivity of the storr.ata to ABA. Guard cells contain ABA (Weiler et al. 1982), and it is possible that a stress-induced redistribution of ABA within the guard cells, brought about by the mechanisms proposed by Heilmann et al. (1980), leads to an accumulation of ABA at the sites of action, and thereby to stomatal closure. A stress-induced increase in the sensitivity of the guard cells to their own ABA should also be taken into account. Davies (1978), Ackerson (1980) and Eamus and Wilson (1983) have provided experimental evidence for this possibility. Acknowledgements The financial support of the Deutsche Forschungsgemeinschaft is gratefully acknowledged. We thank Mrs. Gabriele Reichau for excellent technical assistance.
References
ACKERSON, R. c.: Stomatal response of cotton to water stress and abscisic acid as affected by water stress history. Plant Physiol. 65, 455-459 (1980). - Synthesis and movement of abscisic acid in water-stressed cotton leaves. Plant Physiol. 69, 609-613 (1982). BEARDSELL, M. F. and D. COHEN: Relationship between leaf water status, abscisic acid levels and stomatal resistance in maize and Sorghum. Plant Physiol. 56, 207-212 (1975).
J Plant Pbysiol.
Vol. 117. pp. 297-305 (1985)
Abscisic acid and stomatal closure
305
BENGTSON, c., S. O. FALK, and S. LARSSON: The after-effect of water stress on transpiration rate and changes in abscisic acid content of young wheat plants. Physiol. Plant. 41, 149-154 (1977). DAVIES, W. J.: Some effects of abscisic acid and water stress on stomata of Vicia faba L. J. expo Bot. 29, 175-182 (1978). DORFFLING, K., D. TIETZ, J. STREICH, and M. LUDEWIG: Studies on the role of abscisic acid in stomatal movements. In: Plant growth substances 1979, pp. 274-285, SKOOG, F. (eds.). Springer, Berlin, Heidelberg, New York, 1980. EAMUS, D. and J. M. Wn.sON: ABA levels and effects in chilled and hardened Phaseolus vulgaris. J. expo Bot. 34, 1000-1006 (1983). HEILMANN, B., W. HAKTUNG, and H. GIMMLER: The distribution of abscisic acid between chloroplasts and cytoplasm of leaf cells and the permeability of the chloroplast envelope for abscisic acid. Z. Pflanzenphysiol. 97, 67 -78 (1980). HIRON, R. W. P. and S. T. C. WRIGHT: The role of endogenous abscisic acid in the response of plants to stress. J. expo Bot. 24, 769-781 (1973). ITA!, c., J. D. B. WEYERS, J. R. HILLMAN, H. MEIDNER, and C. WILLMER: Abscisic acid and guard cells of Commelina communis L. Nature (London) 271, 652-653 (1978). JONES, R. J. and T. A. MANSFIELD: Suppression of stomatal opening in leaves treated with abscisic acid.J. expo Bot. 21, 714-719 (1970). LOVEYS, B. R.: The intracellular location of abscisic acid in stressed and non-stressed leaf tissue. Physiol. Plant. 40, 6-10 (1977). MANSFIELD, T. A., A. R. WELLBURN, and T. J. S. MOREIRA: The role of abscisic acid and farnesol in the alleviation of water stress. Philos. Trans. R. Soc. London Ser. B. 284, 471-482 (1978). MILBORROW, B. Y.: A Distinction between the fast and slow responses to abscisic acid. Aust. J. Plant Physiol. 7, 749-754 (1980). PEMADASA, M. A.: Differential abaxial and adaxial stomatal responses to indole-3-acetic acid in Commelina communis. New Phytol. 90, 209-220 (1982). PIERCE, M. L.: Turgor dependence of biosynthesis and metabolism of abscisic acid. Ph. D. thesis, Michigan State University, East Lansing, Michigan, 1981. RASCHKE, K.: Simultaneous requirement of carbon dioxide and abscisic acid for stomatal closure in Xanthium strumarium. L. Planta 125, 243-259 (1975). - Involvement of abscisic acid in the regulation of gas exchange: evidence and inconsistencies. In: Plant growth substances 1982, pp. 581-590, WAREING, P. F., ed. (1982). SINGH, B. N., E. GALSON, W. DASHER, and D. C. WALTON: Abscisic acid levels and metabolism in the leaf epidermal tissue of Tulipa gesneriana L. and Commelina communis L. Planta 146, 135-138 (1979). WALTON, D. c., E. GALSON, and M. A. HARRISON: The relationship between stomatal resistance and abscisic acid levels in leaves of water-stressed bean plants. Planta 133, 145-148 (1977). WEILER, E. W., H. SCHNABL, and C. HORNBERG: Stress-related levels of abscisic acid in guard cell protoplasts of Vida faba L. Planta 154, 24-28 (1982). WEYERS, J. D. B. and J. R. HILLMAN: Uptake and distribution of abscisic acid in Commelina leaf epidermis. Planta 144, 167 -172 (1979 a). - - Sensitivity of Commelina stomata to abscisic acid. Planta 146,623-628 (1979 b).
J. Plant Physiol.
Vol. 117. pp. 297-305 (1985)
Summary Changes in the abscisic acid (ABA) content of epidermis and mesophyll of isolated Comme· lina communis L. leaves were followed and compared to changes in diffusion resistance during a water stress treatment. Diffusion resistance increased within 15 min before a rise in the ABA content of the epidermis could be determined. By application of 14C-Iabelled ABA to isolated leaves via the transpiration stream it was found that the minimal amount of ABA to induce stomatal closure was in the range of 0.4 fmol per stomatal complex. Isolated epidermis was treated with tritiated ABA and subjected to microautoradiography. The stomatal complexes showed an accumulation of radioactive material. This accumulation was observable in living as well as in non-living, heat- or cold-treated epidermis. The results are discussed in relation to the hypothesis that ABA regulates stomatal closure in water-stressed leaves.
Key words: Abscisic aCId, Commelina communis L., Stomatal closure, Water stress.
Introduction The hypothesis that abscisic acid (ABA) is involved in the regulation of guard cell movements under stress conditions is based on several observations, especially the following: open stomata close rapidly, and closed stomata do not open in the presence of ABA Gones and Mansfield 1970). Water stress not only induces stomatal closure, but also causes a considerable rise in the level of ABA in leaves (Hiron and Wright 1973). Despite considerable experimental evidence for a regulatory role of ABA in stomatal movements under stress conditions several problems have remained unresolved. For example, it is still unclear in which part of the leaf ABA is being synthesized. Most authors suggest that ABA is produced in the mesophyll cells and migrates from there via the apoplast to the epidermis (Mansfield et al. 1978), which seems to be unable to synthesize ABA in response to stress (Loveys 1977; Dorffiing et al. 1980). On the other hand, Weiler et al. (1982) have suggested that the guard cells themselves produce ABA in response to stress. Experimental evidence for transport from the mesophyll to the epidermis is not yet available, although ABA has been I) Present address: National Institute of Health, Bethesda, Maryland, USA.
J. Plant Physiol.
Vol. 117. pp. 297-305 (1985)
298
K. DORFFUNG and D. TIETZ
found in the apoplastic medium (Ackerson 1982). In conflict with the suggestion that ABA is involved in stress-induced stomatal closure is the observation that stomata of water-stressed leaves close before a rise in the ABA content in the total leaf and even in the epidermis can be observed (Beardsell and Cohen 1975; Walton et al. 1977; Bengtson et al. 1977; Dorffiing et al. 1980). However, close correlation between both processes has been observed by Pierce (1981). These inconsistencies may be explained in different ways: 1) It is possible that the variations in amounts of ABA necessary for stomatal closure are so small that they cannot be measured with the available analytical methods. 2) Another possibility is that a redistribution of available ABA in the epidermis resulting in an accumulation in the guard cells (Weyers and Hillman 1979 a) is responsible for the beginning of stomatal closure. Such a redistribution would take place without a change in the total ABA content. In the present study both these possibilities have been investigated. The minimal amounts of ABA applied to induce stomatal closure have been compared with the endogenous ABA levels in epidermis and mesophyll of water-stressed Commelina leaves. Furthermore, the distribution of applied radioactive ABA in isolated epidermis has been studied.
Materials and methods Plant material
Seeds of Commelina communis L., kindly provided by Dr. T. A. Mansfield, University of Lancaster, U.K., were sown in a mixture of peat and garden soil. After cultivation for three to four weeks in a greenhouse with supplementary illumination (fluorescent lamps, Philips TLF 65 W 134; 240 JLE m- 2 sec!) the seedlings had four to six leaves. The upper, fully grown leaves were used for the experiments. Stress treatment
Leaves of well-watered plants were cut off at their base and placed on filter paper under fluorescent lamps at room temperature. At different time intervals diffusion resistance was measured with a Licor diffusion resistance meter Model Li 60. From other leaves of the same set parts of the lower epidermis and the mesophyll were isolated, weighed and immediately frozen in liquid nitrogen. With Commelina leaves it is possible to separate the epidermis from the leaf without adhering mesophyll cells. The guard cells remain intact, but part of the epidermal cells are damaged, especially when under water stress. Quantitative determination ofABA
The isolation and quantitative determination of ABA from Commelina tissue proved to be rather difficult and the method described was modified several times. Similar difficulties were reported by Singh et al. (1979). The frozen mesophyll and epidermal tissue was freezedried and For each determination about so mg dry weight subjected to extraction with ethanol at 4 epidermis and about 200 mg mesophyll were used. The ethanol was removed by vacuum evaporation and the residue dissolved in 1 % NaHC0 3-solution (PH 9). This aqueous fraction was shaken three times against ether (ether phase discarded), then acidified (PH 3) with HCl and again extracted with ether. The acid ether phase was dried, redissolved in ethyl acetate and fur-
0c.
J Plant Physiol.
Vol. 117. pp. 297-305 (1985)
Abscisic acid and stomatal closure
299
ther purified by thin layer chromatography on silica gel Merck HF 254 with two different solvent systems (toluene-ethyl acetate-acetic acid 50 : 30 : 4 and hexane-ethyl acetate 1: 1). The second system was used after the purified ABA fraction had been methylated with diazomethane. Quantitative determination of the methylated ABA, dissolved in isooctane, was performed by gas chromatography on glass columns packed with SE 30 10 % on Chromosorb WAW. In most cases a Packard 429 gas chromatograph equipped with a 63Ni electron capture detector was used. The conditions were: column temperature 190°C, detector temperature 235°C, injector temperature 220°, nitrogen flow rate 50 ml min-I. Losses during the purification p,rocedure were determined by means of 2-'4C-ABA as an internal standard. Low amounts of 2- 4C-ABA which were beyond the limit of gas chromatographic detection, were added to the fresh extract and determined at the end of the procedure by scintillation counting.
Application of radioactive ABA a) Uptake studies Freshly cut leaves were placed upright with their cut ends into distilled water. Four cm from the leaf base a cylindrical HMP humidity sensor (0.5 cm diameter) of a Vaisala hygrometer HMJ 14 was attached to the lower epidermis and the humidity of the epidermal surface registrated continuously. The humidity in the growth chamber was 60 %, the temperature 24°C and the light intensity 250ILE m- 2 sec-I. When steady state conditions had been reached, the solution was changed against buffer solution containing 10- 5 moll-I 2_ 14C_ABA, specific activity 348 MBq mmol- ' , obtained from Amersham. At different time intervals after the beginning of stomatal closure about 1 cm2 epidermis from the leaf area where the stomatal transpiration had been measured was isolated. The radioactivity of the tissue was measured in a mixture of 0.5 ml tissue solubilizer and 3 ml scintillation cocktail (4.3 g diphenyloxazol and 0.43 g phenylenbimethylphenyloxazol in 11 toluene) in a Philips scintillation counter. The mesophyll of this leaf area was scraped off and measured for radioactivity in the same manner. The values obtained were corrected for quenching. b) Distribution studies Pieces of the lower epidermis from freshly cut leaves were isolated and incubated on glass slides in 10 mmoll- ' phosphate buffer pH 6.7 + 100 mmoll- ' NaN0 3 containing 0.32 BMq ml- ' 3H-ABA (Amersham, specific activity 832.5 GBq mmol- ' ) or 3H-indole-acetic acid (Amersham, specific activity 777 GBq mmol- ' ). After 260 min the epidermal pieces were washed twice with phosphate buffer, each time for several seconds. After this the epidermal pieces were frozen in liquid nitrogen, freezedried on glass slides and exposed for two months to an Ilford 18 DIN Pan F film. Other epidermal pieces were treated for five min with high (65°C) or low (-25°C) temperatures or for one min with ethanol at 65°C before being frozen and exposed to radioactive ABA.
Results and discussion Correlation between stress ABA production and stomatal closure
When well-watered leaves are subjected to water stress by cutting them from the plant and placing them under fluorescent tubes, the diffusion resistance increases suddenly after 15 min. This increase indicates stomatal closure (Fig. 1). The ABA level in the epidermis, on the other hand, remains constant for at least 30 min before a remarkable rise takes places and the ABA level reaches fifteen times the original content. On the basis of fresh weight or leaf area the ABA level in the mesophyll of
J. Plant Physiol.
Vol. 117. pp. 297-305 (1985)
K.
300 rs ( reI.)
DORFFLING
and D.
TIETZ
Diffusion Resistance rs
120 80
40 Or-~~--~--~--~~
800
ABA
300
!m\
ABA
J!!.
mg FW
600
0--0
200 400
200
0
J 0
100
2
4h
Fig. 1: Above: Changes in diffusion resistance of the abaxial leaf side of isolated Commelina leaves subjected to water stress. Below: Changes in ABA content in the lower leaf epidermis and mesophyll.
unstressed leaves is higher than in the epidermis. During the stress treatment the ABA level in the mesophyll seems to increase earlier than in the epidermis. The maximal level in this tissue after four hours is lower than in the stressed epidermis. The ABA level measured in unstressed epidermis is in the same order of magnitude as found by Singh et al. (1979) and by Pierce (1981) in Commelina epidermis. Isolated epidermis of Commelina and Vicia does not produce ABA in response to stress (Loveys 1977; D6rffiing et al. 1980). However, stomata close before a rise of ABA in the epidermis can be observed. Therefore, attempts were undertaken to determine the minimal concentration of ABA for the induction of stomatal closure. Minimal amounts of I4CABA for the induction of stomatal closure
Isolated leaves were treated at the base with an aqueous solution containing 14elabelled ABA, and the transpiration of a leaf area (lower side) four cm from the base was measured continuously. Fig. 2 shows that about 16 min (15.7± 1.7 min) after ap-
J. Plant Physiol.
Vol. 117. pp. 297-305 (1985)
Abscisic acid and stomatal closure
o
ISO
301
min
Fig. 2: Changes in transpiration of Commelina leaves after application of ABA 10- 5 mol I-I via the transpiration stream (time zero). Transpiration was measured with a Vaisala hygrometer. The sensor cup (0.5 cm2 i.d.) was applied at the lower half of the leaf four cm from the leaf base. The curves represent four different experiments. plication of 10- 5 mol I-I ABA the humidity outside the lower epidermis declines significantly. This indicates the beginning of stomatal closure. Higher concentrations induced earlier closure: 0.5 10- 4 mol I-I ABA within 12 min (12.2± 1.2 min) and 10- 4 moll- I within 10 min (10.4± 1.3 min). The reduction of transpiration was not continuous, but occurred in oscillations. When 14C_ABA was applied, radioactivity accumulated preferably in the mesophyll and to a much smaller extent in the epidermis (Fig. 3). 19.7 min after application of 10- 5 mol I-I 2)4C-ABA which closely corresponds to the beginning of stomatal closure, 455 pg ABA ( = 0.6 Bq) were present per cm2 epidermis. This is equivalent to 0.39 fmol per stomatal complex. After 30 min the amount of radioactive ABA had increased to 816 pg cm- 2 epidermis or 0.7 fmol per stomatal complex. These amounts are comparable to data found by other authors. Raschke (1975) found that two to four fmol ABA per stomatal complex were necessary to induce 5 % closure in Commelina. Weyers and Hillman (1979 a) calculated that no more than six fmol ABA were present per stomatal complex at the time of closure. In a later paper Weyers and Hillman (1979 b) reported that between 12.6 and 45.4 amol radioactive ABA per complex were present when statistically significant narrowing had occurred. This is less than one tenth of the amount found in the present investigation. In comparision to these data the amounts of endogenous ABA in
J. Plant Physiol.
Vol. 117. pp. 297-305 (1985)
302
K. DORFFLING and D. TIETZ
30
20
10
o L. ...- - -
o
30
eo
10
120 ",in
Fig. 3: Increase of radioactivity in lower epidermis and mesophyll after application of 2_ 14C_ ABA 10- 5 moll-I via the transpiration stream. The area measured was the same as that in Fig. 2. The curves represent the mean of six experiments. Vertical bars: standard deviation. the unstressed epidermis are much lower (about 50 pg cm- 2 or 43 amol per stomatal complex (Fig. 1). An increase of the endogenous epidermal ABA content to the amount of 455 pg cm- 2 which was present after application of ABA at the beginning of stomatal closure (Fig.3) would have been measurable. This calculation is, of course, based on the assumption that applied synthetic ABA is as effective as the endogenous hormone, an assumption which may not be tenable. Moreover, the calculation does not take into consideration that probably only the + - isomer of ABA is active on stomata (Milborrow 1980). Since synthetic ABA is a racemate of + and - forms, the values for applied ABA must be halved. It may be concluded, that either very small, unmeasurable changes in the concentration of the endogenous ABA are responsible for the induction of stomatal closure, or that a redistribution of available ABA within the epidermis occurs and is necessary for stomatal closure.
Distribution of 3H-ABA in isolated epidermis strips Whether unequal distribution of ABA in the Commelina epidermis is possible and whether it is an essential property of the living tissue has been tested by application of tritiated ABA to isolated epidermis and subsequent autoradiography. Fig. 4 shows microautoradiographs from Commelina epidermis which had been incubated in 3H_ ]. Plant Physiol. Vol. 117. pp. 297-305 (1985)
Abscisic acid and stomatal closure
303
Fig. 4: Microautoradiography of epidermal strips incubated in 3H-ABA (right). The accumulation of radioactivity corresponds to the location and number of stomatal complexes (left). Upper half: living epidermis. Lower half: frozen (-25 °C)-thawed epidermis. Incubation medium: P-buffer 10 mmoll- 1, pH 6.7, + 100 mmoll- 1 NaN0 3 + 0.32 MBq 3H-ABA ml- 1• Incubation time 260 min.
ABA. It is evident that radioactive ABA accumulates preferably in the stomatal complexes. This confirms results obtained by Itai et al. (1978) and by Weyers and Hillman (1979 a, b), who suggested that this accumulation of ABA at or near the guard cells "is further evidence that ABA travels during stress from the mesophyll (the site of synthesis) to the guard cell region (the site of action) ... ». However, the accumulation of radioactive material occurs also in non-living epidermis which has been treated either with heat (65°), or cold (-25°C). Even ethanol-treated (65°C) stomatal complexes showed this effect. The accumulation of radioactive material is, moreover, not restricted to ABA. We have observed that 3H-indoleacetic acid accumulates in a similar
J. Plant P}rysiol. Vol. 117. pp. 297-305 (1985)
304
K. DORFFLING and D. TIETZ
manner in stomatal complexes of living and non-living epidermis. The concentration used (3.5 .10- 5 moll-I) had no effect on stomatal pore width. This is in disagreement with observations of Pemadesa (1982). Conclusion The fact that stomata of water-stressed Commelina leaves close before a rise in the ABA content of the epidermis can be observed needs further explanation. It confirms the observations of D6rffiing et al. (1980), but is in contrast to recent data obtained by Pierce (1981). Raschke (1982) has calculated from the data of Pierce (1981) that the epidermal ABA content rose by less than 26 pg ABA cm- 2 when the stomata closed. Such a rise should have been measurable in the present investigation. A rise in the ABA content before the stomata begin to close should also be expected if it is assumed that the stress-ABA is produced in the guard cells themselves, as Weiler et al. (1982) suggested. The data of Weiler et al. (1982), however, also showed that the ABA content in broad bean epidermis rose after the beginning of stomatal closure. An explanation for the inconsistency of the kinetics of ABA increase and stomatal closure would be the redistribution of available ABA in the epidermis resulting in an accumulation of the hormone at or near the guard cells. This could be a first step in the chain of events leading to stomatal closure. Supporting this possibility is the accumulation of radioactive ABA in stomatal complexes. However, non-living epidermis also accumulates applied ABA. The relevance of this process should be investigated further. Theoretically, two other mechanisms may also provide explanations for the inconsistency between stomatal closure and ABA levels: an intracellular rather than an intercellular redistribution, and!or a change in the sensitivity of the storr.ata to ABA. Guard cells contain ABA (Weiler et al. 1982), and it is possible that a stress-induced redistribution of ABA within the guard cells, brought about by the mechanisms proposed by Heilmann et al. (1980), leads to an accumulation of ABA at the sites of action, and thereby to stomatal closure. A stress-induced increase in the sensitivity of the guard cells to their own ABA should also be taken into account. Davies (1978), Ackerson (1980) and Eamus and Wilson (1983) have provided experimental evidence for this possibility. Acknowledgements The financial support of the Deutsche Forschungsgemeinschaft is gratefully acknowledged. We thank Mrs. Gabriele Reichau for excellent technical assistance.
References
ACKERSON, R. c.: Stomatal response of cotton to water stress and abscisic acid as affected by water stress history. Plant Physiol. 65, 455-459 (1980). - Synthesis and movement of abscisic acid in water-stressed cotton leaves. Plant Physiol. 69, 609-613 (1982). BEARDSELL, M. F. and D. COHEN: Relationship between leaf water status, abscisic acid levels and stomatal resistance in maize and Sorghum. Plant Physiol. 56, 207-212 (1975).
J Plant Pbysiol.
Vol. 117. pp. 297-305 (1985)
Abscisic acid and stomatal closure
305
BENGTSON, c., S. O. FALK, and S. LARSSON: The after-effect of water stress on transpiration rate and changes in abscisic acid content of young wheat plants. Physiol. Plant. 41, 149-154 (1977). DAVIES, W. J.: Some effects of abscisic acid and water stress on stomata of Vicia faba L. J. expo Bot. 29, 175-182 (1978). DORFFLING, K., D. TIETZ, J. STREICH, and M. LUDEWIG: Studies on the role of abscisic acid in stomatal movements. In: Plant growth substances 1979, pp. 274-285, SKOOG, F. (eds.). Springer, Berlin, Heidelberg, New York, 1980. EAMUS, D. and J. M. Wn.sON: ABA levels and effects in chilled and hardened Phaseolus vulgaris. J. expo Bot. 34, 1000-1006 (1983). HEILMANN, B., W. HAKTUNG, and H. GIMMLER: The distribution of abscisic acid between chloroplasts and cytoplasm of leaf cells and the permeability of the chloroplast envelope for abscisic acid. Z. Pflanzenphysiol. 97, 67 -78 (1980). HIRON, R. W. P. and S. T. C. WRIGHT: The role of endogenous abscisic acid in the response of plants to stress. J. expo Bot. 24, 769-781 (1973). ITA!, c., J. D. B. WEYERS, J. R. HILLMAN, H. MEIDNER, and C. WILLMER: Abscisic acid and guard cells of Commelina communis L. Nature (London) 271, 652-653 (1978). JONES, R. J. and T. A. MANSFIELD: Suppression of stomatal opening in leaves treated with abscisic acid.J. expo Bot. 21, 714-719 (1970). LOVEYS, B. R.: The intracellular location of abscisic acid in stressed and non-stressed leaf tissue. Physiol. Plant. 40, 6-10 (1977). MANSFIELD, T. A., A. R. WELLBURN, and T. J. S. MOREIRA: The role of abscisic acid and farnesol in the alleviation of water stress. Philos. Trans. R. Soc. London Ser. B. 284, 471-482 (1978). MILBORROW, B. Y.: A Distinction between the fast and slow responses to abscisic acid. Aust. J. Plant Physiol. 7, 749-754 (1980). PEMADASA, M. A.: Differential abaxial and adaxial stomatal responses to indole-3-acetic acid in Commelina communis. New Phytol. 90, 209-220 (1982). PIERCE, M. L.: Turgor dependence of biosynthesis and metabolism of abscisic acid. Ph. D. thesis, Michigan State University, East Lansing, Michigan, 1981. RASCHKE, K.: Simultaneous requirement of carbon dioxide and abscisic acid for stomatal closure in Xanthium strumarium. L. Planta 125, 243-259 (1975). - Involvement of abscisic acid in the regulation of gas exchange: evidence and inconsistencies. In: Plant growth substances 1982, pp. 581-590, WAREING, P. F., ed. (1982). SINGH, B. N., E. GALSON, W. DASHER, and D. C. WALTON: Abscisic acid levels and metabolism in the leaf epidermal tissue of Tulipa gesneriana L. and Commelina communis L. Planta 146, 135-138 (1979). WALTON, D. c., E. GALSON, and M. A. HARRISON: The relationship between stomatal resistance and abscisic acid levels in leaves of water-stressed bean plants. Planta 133, 145-148 (1977). WEILER, E. W., H. SCHNABL, and C. HORNBERG: Stress-related levels of abscisic acid in guard cell protoplasts of Vida faba L. Planta 154, 24-28 (1982). WEYERS, J. D. B. and J. R. HILLMAN: Uptake and distribution of abscisic acid in Commelina leaf epidermis. Planta 144, 167 -172 (1979 a). - - Sensitivity of Commelina stomata to abscisic acid. Planta 146,623-628 (1979 b).
J. Plant Physiol.
Vol. 117. pp. 297-305 (1985)