- Email: [email protected]
Ion gradients and adenosine triphosphatase localization in the salt glands of Avicennia marina (Forsskål) Vierh.
S.Afr.J.Bot., 1992, 58(6): 486 - 490
486
Ion gradients and adenosine triphosphatase localization in the salt glands of A vicennia marina (Forsskal) Vierh. Philippa M. Drennan,* Patricia Berjak and N.W. Pammenter Department of Biology, University of Natal, King George V Avenue, Durban, 4001 Republic of South Africa 'Present address: UN/FRD Research Unit for Plant Growth and Development, Department of Botany, University of Natal, P.O. Box 375, Pietermaritzburg 3200, Republic of South Africa. Received 24 December 1991; revised 2 July 1992
Possible sites of iron pumping associated with glandular salt excretion from the leaf of the mangrove
Avicennia marina were investigated using energy-dispersive X-ray analysis and ATPase cytochemistry. The collecting cells of the gland had low sodium and chlorine ratios and high potassium ratios. This contrasted with the stalk/excretory cells of the gland which had very high chlorine ratios and lower potassium ratios. Considerable ATPase activity was localized at the stalk cell plasmalemma in the region of the stalk cell / collecting cell interface, and at the collecting cell plasmalemma. It is suggested that these are sites of active ion influx into the gland. The low chlorine ratio within the collecting cells by comparison to the xylem vessels and the surrounding spongy mesophyll suggests that the collecting cells are at the endpoint of a symplastic and possibly an apoplastic downhill ion gradient. ATPase activity associated with the plasmalemma of the excretory cells suggests that efflux from the gland is also an active process. Energieverspreidende X-straal-analise en ATPase-sitochemie is gebruik in die ondersoek van moontlike areas van ioonuitruiling wat geassosieer is met soutuitskeiding deur kliere in die blaar van die mangrove Avicennia marina. Die versamelselle van die klier het 'n lae natrium- en chloorverhouding en 'n hoe kaliumverhouding getoon. Dit is in teenstelling met die baie hoe chloorverhouding en laer kaliumverhouding wat in die stingeilsekreterende selle gevind is. 'n Redelik hoe aktiwiteit is gelokaliseer in die plasmalemma van die stingelselle wat gelee is in die omgewing van sowel die stingelsellversamelsel-interfase as in die versamelsel-plasmalemma. Dit mag dus wees dat dit areas van aktiewe invloei van ione na binne die klier is. Die lae chloorverhouding binne die versamelselle in vergelyking met die xileemvate en die omringende sponsagtige mesofil impliseer dat die versamelselle aan die eindpunt is van 'n simplasties en moontlik 'n apoplasties dalende ioongradient. Die ATPase-aktiwiteit wat met die plasmalemma van die sekresieselle geassosieer is, impliseer dat uitvloei vanaf die klier ook 'n aktiewe proses is.
Keyword: Avicennia marina, adenosine triphosphatase, halophyte, ion gradients, mangrove, salt excretion, salt glands.
Introduction
Glandular salt excretion in halophytes is probably an active process (Thomson et at. 1988). Studies using various metabolic inhibitors (Ariz et at. 1955; Atkinson et at. 1967; Hill & Hill 1973; LUttge 1964, 1966; LUttge & Osmond 1970) suggest that ATP drives ion transport. However, the location of the ion pumps is not known. The large number of mitochondria in salt gland cells (Atkinson et at. 1967; LUttge 1975; Faraday & Thomson 1986) suggests that at least one active process occurs in the glands. The concentration of salt (Na+ and NaCI) in the xylem sap of the salt-excreting mangrove Avicennia marina (Forsskill) Vierh. is in the range of 25 - 200 mM (Scholander et at. 1966; Drennan & Pammenter 1982; Waisel et at. 1986) whereas that of the glandular exudate is approximately 700 mM (Scholander et at. 1962), indicating that salt excretion in A. marina is energy-requiring. Shimony et at. (1973) have suggested, on the basis of a low salt content in the adaxial glands of the mature leaf, that an ion pump is located in the region of the vascular bundles with possibly a second pump in the secretory cells transferring ions from the symplast to the exterior. Drennan et at. (1987) have predicted, on the basis of the ultrastructure of functional A. marina salt glands, that at least one pump site on the
excretory route is situated in the glands at the point of symplastic loading. The present study is an investigation into possible sites of ion pumping associated with the salt glands of A. marina and involves characterization of ion gradients using energydispersive X-ray analysis of bulk frozen material and cytochemical localization of ATPase activity. Materials and Methods Material Propagules of A vicennia marina collected from the Beachwood Nature Reserve, Durban, were germinated on vermiculite and irrigated weekly with a modified Hoagland's nutrient solution (Epstein 1972). When the developing plants had at least one pair of mature leaves, they were transferred into nutrient solutions of salinities equivalent to o and 50% seawater (Drennan & Pammenter 1982). Plants were maintained in controlled environment cabinets with a 14-h/1O-h light/dark cycle and a corresponding day/night temperature cycle of 30°C/23°C at >90% r.h.
Energy-dispersive X-ray analysis Pieces (approximately 5 mm 2 ) of fresh leaf tissue from the 50% seawater treatment were placed into grooves in large
S.AfrJ.Bot.,1992,58(6)
487
brass stubs (30 mm in diameter). The stub was then submerged in liquid nitrogen and the frozen leaf fragments fractured using a razor blade. The stub with the fractured leaf was subsequently transferred as rapidly as possible to the specimen chamber of a Jeol JSM 3S SEM with attached KEVEX' microanalyser. As the microscope used was not fitted with a cold stage, specimens were maintained in the column for approximately 20 min only, after which time they were discarded. The specimens, however, apparently remained frozen during the period of analysis. The operating conditions for analysis were as follows: emission current, 17S !-LA; Voltage, IS kV; working distance, 3S mm; take-off angle, 4So; dead time, approximately 3S%; probe raster area, 2 !-Lm2. A preset time of 30 s was used for collecting counts. The results were compared on a semi-quantitative basis by expressing the area under the Kaelemental peak of each element (elemental peak integral) as a ratio to the area under the background region 4.S - S.S keY. Harvey et al. (1981, 1986) have suggested that insufficient is known about intercellular differences in ion distribution in plant tissues to determine whether the ionic content of cells is uniform within a given population, or constant with time. Thus, although averages were used in the present study to compare ion distribution data, statistical analysis of the data is not included (Harvey et al. 1986). ATPase localization ATPase activity was localized in fully expanded leaves from both SO% seawater and salt-free treatments using a lead phosphate precipitation procedure (Hall 1971). Pieces of tissue, approximately S mm 2, were fixed in 2% glutaraldehyde in a O.OSM sodium cacodylate buffer (PH 7.2) for 1 h at 4°C. Tissue segments were then washed for 1 h in O.OSM sodium cacodylate buffer (with buffer changes every IS min) followed by two IS-min rinses in O.IM Trismaleate buffer (PH 7.2). Tissue was infiltrated at 4°C with an incubation medium consisting of 2mM ATP, 6mM MgCI2 and 3.6mM Pb(N~h in O.lM Tris-maleate buffer (PH 7.2). The tissue was then incubated at 3SoC for 30 min. Following incubation, tissue was rinsed with distilled water and post-fixed in 2% osmium tetroxide in O.OSM sodium cacodylate buffer (pH 7.2) for 2 h. Tissue was dehydrated through a graded acetone series and infiltrated and embedded in Spurr's (1969) epoxy resin. Controls consisted of leaf segmens incubated in a medium from which ATP had been omitted (Hall 1971), or pieces of leaf tissue which were boiled for 20 min prior to incubation in the ATP-
containing medium. Ultrathin sections were viewed without post-staining. Results Ion distribution gradients
The spatial relationship of the various salt gland cells is shown in Figure 1. The elemental ratios within the collecting cells were distinct from those of the adjacent spongy mesophyll cells and the stalk and excretory cens (Table 1). The chlorine ratios of the collecting cens were consistently lower than those of the spongy mesophyll and markedly lower than those of the stalk and excretory cens. They were also lower than the chlorine ratios of the xylem vessels. The potassium ratios of the collecting cells were, however, much higher than those of the spongy mesophyll cells and the stalk and excretory cells. The proportional distribution of sodium was similar to that observed for chlorine. However, much lower ratios were obtained, probably as sodium is near the lower limit of elemental detection that can be achieved using energy-dispersive X-ray analysis (postek et al. 1980). ATPase localization Staining for ATPase activity, which indicated the presence of a membrane proton pump and/or a membrane-associated soluble phosphatase (Katz et al. 1988), was marked along the plasmalemma of the collecting cells in the abaxial glands of the mature leaf (Figure 2). The cytochemical technique used, although it did not distinguish between types of phosphatase, resulted in a marked polarity in distribution of lead precipitate in the stalk cell. Heavy
Figure 1 Anatomy of the abaxial salt gland of A. marina. c, collecting cell; s, stalk cell; e, excretory cell; m, spongy mesophyll cell.
Table 1 The ratio of the elemental peak integral to the integral of the background region 4.5 - 5.5 keV for the elemental analysis of ce lls of the abaxial salt glands of leaves of A. marina grown in 50% S.W.a Cell type
n
Xylem vessel Spongy mesophyll Collecting cell Stalk/excretory cell b
4 6
3 3
Na 0.13 0.43 0.09 0.89
(0.02 (0.12 (0.07 (0.72
Cl -
0.34) 0.64) 0.12) 1.13)
1.74 4.27 0.90 7.02
(1.06 (2.56 (0.75 (5.86
• Number of cells is indicated (n), and range is given in parentheses. b Stalk cell could not always be distinguished from excretory cell.
K
-
2.85) 5.94) 0.95) 8.47)
1.17 1.51 2.12 0.78
(0.88 (0.73 (1.98 (0.50
-
1.64) 1.95) 2.32) 0.82)
488
staining occurred along the plasmalemma of the stalk cells at the stalk cell/collecting cell interface (Figure 2) and at the points of adhesion of the stalk cell plasmalemma to the cutinized lateral walls of the stalk cells (Figure 2). Weak staining was found along the plasmalemma of the excretory cells (Figure 3). The staining at this surface was much less
S.-Afr.Tydskr.Plantk., 1992, 58(6)
intense than that found at the stalk cell plasmalemma (cf. Figures 2 and 3). No staining of endomembranes in the stalk or excretory cells was apparent. The pattern and intensity of staining was similar in glands from 0 and 50% SW treatments (cf. Figures 2 and 4). Both controls showed no membrane reaction (Figure 5).
Figure 2 Portion of an abaxial gland from a mature leaf of a 50% SW plant. Dense staining for ATPase activity (-+) is associated with the collecting cell (c) plasmalemma and the stalk cell (s) plasmalemma at the collecting cell/stalk cell interface and at the plasmalemma! lateral wall adhesion (pa). Figure 3 ATPase activity (-+) localized at the surface of the excretory cells (e) of a gland from a 50% SW treatment. Figure 4 Portion of an abaxial gland from a mature leaf of a 0% SW plant. The staining pattern for ATPase activity (-+) is similar to that found in 50% SW material. c, collecting cell; s, stalk cell; e, excretory cell. Figure 5 Control tissue (spongy mesophyll) in which the ATPase has been deactivated by boiling, showing an absence of staining (cf. Figure 4). Similar results were obtained in controls when substrate was omitted.
S.AfrJ.Bot., 1992,58(6)
Discussion Salt excreted via the glands is probably derived from both the apoplast and the symplast. The low chlorine and sodium ratios of the collecting cells relative to the xylem vessels and surrounding spongy mesophyll cells suggest that the collecting cells are at the endpoint of a symplastic and possibly an apoplastic downhill ion gradient towards the glands. Shimony et al. (1973) suggested that there was a downhill gradient from the vascular bundles to the excretory surface (as opposed to the collecting cells). These authors made the assumption that the adaxial glands on the mature leaf excrete salt. However, as the leaf matures, the glands on the adaxial surface undergo autolysis (Drennan & Berjak 1982). The high ratio for potassium in the collecting cell in the absence of high concentrations of sodium and chlorine, suggests that the accumulation of potassium maintains a low osmotic potential in the collecting cells. The collecting cells probably provide the immediate source of water for movement through the gland cells. The reservoir cell of the digestive glands of Pinguicula ionantha which would be the collecting cell equivalent, is suggested to provide the immediate source of water of the excreted fluid in that species (Heslop-Harrison & Heslop-Harrison 1980). As the collecting cells in A. marina and the equivalents in many species are distinguishable from the adjacent spongy mesophyll cells only by their spatial relationship to the gland and a lack of well-developed chloroplasts (Thomson 1975), their possible role in the excretory mechanism has generally been overlooked. However, the current study suggests that the collecting cell may provide a crucial link between mesophyll salt regulation and salt export. In only one other species, Pinguicula ionantha, has a definite role been ascribed to the collecting cell equivalent (reservoir cell) which is suggested to act as a site of preferential chlorine accumulation, the chlorine subsequently being pumped through the gland on stimulation by prey capture (Heslop-Harrison & Heslop-Harrison 1980). The implied high chlorine content of the reservoir cell of P. ionantha by contrast to the measured low chlorine ratio of the A. marina collecting cell, may be a reflection of inducible ion pumps in the former as opposed to continuous-functioning ion pumps in the latter. Similarity in staining patterns for ATPase activity in non-excreting and excreting A. marina suggests that pumps in A. marina are constitutive. The mechanisms by which the characteristic ion ratios of the collecting cells are regulated are, however, not known. It would be of particular interest to ascertain whether the elemental ratios of the collecting cell vary with changes in salt loading of the surrounding symplast and apoplast, or whether they provide a 'set point' in the excretory system. The maintenance of a constant level of cr and Na+ in the collecting cell with all incoming Na+ and cr over and above that amount moving into the stalk cell, could explain how an increase in salt loading of the leaf results in an increased excretion rate (Drennan & Pammenter 1982). Furthermore, a constant K+ level in the collecting cell might result in a more concentrated saline solution being excreted under conditions of water stress. The possible significance of the ionic ratios of the collecting cell with respect to the salt and water
489
budget of the leaf underscores the need for further research in this area. The marked ultrastructural differences between the collecting and stalk cells of A. marina as well as the cutinized lateral thickenings of the stalk cell, suggest that apoplastic salt must move into the gland symplasm at least before reaching the collecting/stalk cell interface (Drennan et al. 1987). These authors suggested that loading of the stalk cell symplasm involves transport of salt ions across both the stalk cell plasmalemma and the tubular endoplasmic reticulum (tER) membranes in series. The high chlorine ratios of the stalk/excretory cells, as well as the large amount of ATPase activity localized at the stalk cell plasmalemma at the stalk cell/collecting cell interface, suggest that symplastic loading is active and possibly represents concentration against an electrochemical gradient. This would enable the formation of a downhill electrochemical gradient between the point of loading at the stalk cell and the plasmalemma of the excretory cells, as well as a gradient between the collecting cells and mesophyll cells and apoplast (xylem). The localization of some ATPase activity at the excretory cell surface suggests that the unloading of the gland is also an active process. Unloading does not necessarily occur against an electrochemical gradient. However, it is possible that energy is required for the eccrine transfer of ions across the membranes (tER and plasmalemma) at the gland surface. Faraday and Thomson (1986) have suggested that the high mitochondrial volume and elaborated surface area of the excretory cells of the salt glands of Limonium perezzi are indicative of the fact that ion efflux rather than influx is the active step in the excretory process. Similar suggestions have been made, on an ultrastructural basis, by several other researchers (Hill & Hill 1973, 1976; Shimony et al. 1973; LUuge 1975). However, trans-tissue microelectrode insertion through the salt glands of the mangrove Aegicerus corniculatum (Bostrom & Field 1973; Billard & Field 1974) have shown that a major change in potential occurs between the mesophyll and the basal gland cell (stalk cell equivalent), which led these authors to conclude that the inner basal cell membrane is of importance in the electrophysiology of the excretory process. The present study strongly suggests that although efflux from the salt glands of A . marina may be an active process, influx of ions into the stalk cell symplasm is a primary site of ion pumping. Acknowledgements We gratefully acknowledge the financial assistance of the Foundation for Research Development, Pretoria, and the technical assistance of P. Evers (EDX) of the University of Natal, Durban, Electron Microscope Unit, and J. Lawton (A TPase localization) of the University of Durban-Westville Electron Microscope Unit. C. Snyman is thanked for the Afrikaans translation of the abstract. References ARIZ, W.H., CAMPHUIS, I.J., HEIKENS, H. & VAN TOOREN, AJ . 1955. The secretion of the salt glands of Limonium lalifolium Ktze. Acla Bal. Neerl. 4: 322 - 338 .
490 ATKINSON, M.R., FINDLAY, c.P., HOPE, A.B., PITMAN, M.G., SADDLER, H.D.W. & WEST, K.R. 1967. Salt regulation in the mangroves Rhizophora mucronata Lam. and Aegialitis annulata R. Br. Aust. 1. Bioi. Sci. 20: 589 - 599. BILLARD, B. & FIELD, C.D. 1974. Electrical properties of the salt gland of Aegiceras. Planta 115: 285 - 296. BOSTROM, T.E. & FIELD, C.D. 1973. Electrical potentials in the salt gland of Aegiceras. In: Ion Transport in Plants, ed. W.P. Anderson, pp. 385 - 392. Academic Press, London. DRENNAN, P.M. & BERJAK, P. 1982. Degeneration of the salt glands accompanying foliar maturation in Avicennia marina (Forssk!l.l) Vierh. New Phytol. 90: 165 - 176. DRENNAN, P.M., BERJAK, P., LAWTON, J.R. & PAMMENTER, N.W. 1987. Ultrastructure of the salt glands of the mangrove, Avicennia marina (Forssk.) Vierh., as indicated by the use of selective membrane staining. Planta 172: 176 - 183. DRENNAN, P.M. & PAMMENTER, N.W. 1982. Physiology of salt excretion in the mangrove Avicennia marina (Forssk.) Vierh. New Phytol. 91: 597 - 606. EPSTEIN, E.L. 1972. Mineral Nutrition in Plants: Principles and Perspectives. Wiley, New York. FARADA Y, C.D. & THOMSON, W.W. 1986. Morphometric analysis of Limonium salt glands in relation to ion flux. 1. Exp. Bot. 37: 471 - 481. HALL, J.L. 1971. Cytochemical localization of ATPase activity in plant root cells. 1. Microsc. 93: 219 - 225. HARVEY, D.M.R., HALL, J.L., FLOWERS, T.J. & KENT, B. 1981. Quantitative ion localization within Suaeda maritima leaf mesophyll cells. Planta 151: 555 - 560. HARVEY, D.M.R., STELZER, R., BRANDTNER, R. & KRAMER, D. 1986. Effects of salinity on ultrastructure and ion distributions in roots of Plantago coronopus. Physiol. Pl. 66: 328 - 338. HESLOP-HARRISON, Y. & HESLOP-HARRISON, J. 1980. Chloride ion movement and enzyme secretion from the digestive glands of Pinguicula. Ann. Bot. 45: 729 - 731. HILL, A.E. & HILL, B.S. 1973. The electrogenic chloride pump of the Limonium salt gland. 1. Membr. Bioi. 12: 129 - 144. KATZ, D.B., SUSSMAN, M.R., MIERZWA, R.J. & EVERT, R.F. 1988. Cytochemical localization of ATPase activity in oat
S.-Afr.Tydskr.Plantk., 1992,58(6) roots localizes a plasma membrane-associated soluble phosphatase, not the proton pump. Pl. Physiol. 86: 841 - 847. LUTTGE, U. 1964. Untersuchungen zur Physiologie der Carnivoren-Driisen. Ber. dt. bot. Ges. 63 (Supp!.): 181 - 187. LUTTGE, U. 1966. Untersuchungen zur Physiologie der Carnivoren-Driisen. V. Mikroautoradiographische Untersuchung der Chloridsekretion durch das Driisengewebe von Nepenthes. Planta 68: 269 - 285. LUTTGE, U. 1975. Salt Glands. In: Ion Transport in Plant Cells and Tissues, eds. D.A. Baker & J.L. Hall, pp. 335 - 336. North Holland Publishing Company, Amsterdam. LUTTGE, U. & OSMOND, C.B. 1970. Ion absorption in Atriplex leaf tissue. III. Site of metabolic control of light-dependent chloride secretion to epidermal bladders. Aust. 1. bioi. Sci. 23: 17 - 25. POSTEK, T.P., HOWARD, K.S., JOHNSON, A.H. & McMICHAEL, K.L. 1980. Scanning Electron Microscopy. A Student's Handbook. Ladd Research Industries, Inc. SCHOLANDER, P.F., BRADSTREET, E.D., HAMMEL, H.T. & HEMMINGSON, E.A. 1966. Sap concentrations in halophytes and some other plants . Pl. Physiol. 41: 529 - 532. SCHOLANDER, P.F., HAMMEL, H.T., HEMMINGSON, E. & GAREY, W. 1962. Salt balance in mangroves. Pl. Physiol. 3: 722 - 729. SHIMONY, c., FAHN, A. & RHEIN HOLD, L. 1973. Ultrastructure and ion gradients in the salt glands of A vicennia marina (Forssk.) Vierh. New Phytol. 72: 27 - 36. SPURR, A.R. 1969. A low viscosity epoxy resin embedding medium for electron microscopy. 1. Ultrastruet. Res. 26: 31 - 43. THOMSON, W.W. 1975. The structure and function of salt . glands. In: Plants in Saline Environments, Ecological Studies 15, eds . A. Poljakoff-Mayber & J. Gale, pp. 118 - 146. Springer Verlag, Berlin, Heidelberg. THOMSON, W.W., FARADAY, C.D. & OROS, J.w. 1988. Salt glands. In: Solute Transport in Plant Cells and Tissues, eds. D.A. Baker & J.L. Hall, ch. 13. Longman Scientific and Technical, Essex. WAISEL, Y., ESHEL, A. & AGAMI, M. 1986. Salt balance of the leaves of the mangrove Avicennia marina. Physiol. Pl. 67: 67 - 72.
486
Ion gradients and adenosine triphosphatase localization in the salt glands of A vicennia marina (Forsskal) Vierh. Philippa M. Drennan,* Patricia Berjak and N.W. Pammenter Department of Biology, University of Natal, King George V Avenue, Durban, 4001 Republic of South Africa 'Present address: UN/FRD Research Unit for Plant Growth and Development, Department of Botany, University of Natal, P.O. Box 375, Pietermaritzburg 3200, Republic of South Africa. Received 24 December 1991; revised 2 July 1992
Possible sites of iron pumping associated with glandular salt excretion from the leaf of the mangrove
Avicennia marina were investigated using energy-dispersive X-ray analysis and ATPase cytochemistry. The collecting cells of the gland had low sodium and chlorine ratios and high potassium ratios. This contrasted with the stalk/excretory cells of the gland which had very high chlorine ratios and lower potassium ratios. Considerable ATPase activity was localized at the stalk cell plasmalemma in the region of the stalk cell / collecting cell interface, and at the collecting cell plasmalemma. It is suggested that these are sites of active ion influx into the gland. The low chlorine ratio within the collecting cells by comparison to the xylem vessels and the surrounding spongy mesophyll suggests that the collecting cells are at the endpoint of a symplastic and possibly an apoplastic downhill ion gradient. ATPase activity associated with the plasmalemma of the excretory cells suggests that efflux from the gland is also an active process. Energieverspreidende X-straal-analise en ATPase-sitochemie is gebruik in die ondersoek van moontlike areas van ioonuitruiling wat geassosieer is met soutuitskeiding deur kliere in die blaar van die mangrove Avicennia marina. Die versamelselle van die klier het 'n lae natrium- en chloorverhouding en 'n hoe kaliumverhouding getoon. Dit is in teenstelling met die baie hoe chloorverhouding en laer kaliumverhouding wat in die stingeilsekreterende selle gevind is. 'n Redelik hoe aktiwiteit is gelokaliseer in die plasmalemma van die stingelselle wat gelee is in die omgewing van sowel die stingelsellversamelsel-interfase as in die versamelsel-plasmalemma. Dit mag dus wees dat dit areas van aktiewe invloei van ione na binne die klier is. Die lae chloorverhouding binne die versamelselle in vergelyking met die xileemvate en die omringende sponsagtige mesofil impliseer dat die versamelselle aan die eindpunt is van 'n simplasties en moontlik 'n apoplasties dalende ioongradient. Die ATPase-aktiwiteit wat met die plasmalemma van die sekresieselle geassosieer is, impliseer dat uitvloei vanaf die klier ook 'n aktiewe proses is.
Keyword: Avicennia marina, adenosine triphosphatase, halophyte, ion gradients, mangrove, salt excretion, salt glands.
Introduction
Glandular salt excretion in halophytes is probably an active process (Thomson et at. 1988). Studies using various metabolic inhibitors (Ariz et at. 1955; Atkinson et at. 1967; Hill & Hill 1973; LUttge 1964, 1966; LUttge & Osmond 1970) suggest that ATP drives ion transport. However, the location of the ion pumps is not known. The large number of mitochondria in salt gland cells (Atkinson et at. 1967; LUttge 1975; Faraday & Thomson 1986) suggests that at least one active process occurs in the glands. The concentration of salt (Na+ and NaCI) in the xylem sap of the salt-excreting mangrove Avicennia marina (Forsskill) Vierh. is in the range of 25 - 200 mM (Scholander et at. 1966; Drennan & Pammenter 1982; Waisel et at. 1986) whereas that of the glandular exudate is approximately 700 mM (Scholander et at. 1962), indicating that salt excretion in A. marina is energy-requiring. Shimony et at. (1973) have suggested, on the basis of a low salt content in the adaxial glands of the mature leaf, that an ion pump is located in the region of the vascular bundles with possibly a second pump in the secretory cells transferring ions from the symplast to the exterior. Drennan et at. (1987) have predicted, on the basis of the ultrastructure of functional A. marina salt glands, that at least one pump site on the
excretory route is situated in the glands at the point of symplastic loading. The present study is an investigation into possible sites of ion pumping associated with the salt glands of A. marina and involves characterization of ion gradients using energydispersive X-ray analysis of bulk frozen material and cytochemical localization of ATPase activity. Materials and Methods Material Propagules of A vicennia marina collected from the Beachwood Nature Reserve, Durban, were germinated on vermiculite and irrigated weekly with a modified Hoagland's nutrient solution (Epstein 1972). When the developing plants had at least one pair of mature leaves, they were transferred into nutrient solutions of salinities equivalent to o and 50% seawater (Drennan & Pammenter 1982). Plants were maintained in controlled environment cabinets with a 14-h/1O-h light/dark cycle and a corresponding day/night temperature cycle of 30°C/23°C at >90% r.h.
Energy-dispersive X-ray analysis Pieces (approximately 5 mm 2 ) of fresh leaf tissue from the 50% seawater treatment were placed into grooves in large
S.AfrJ.Bot.,1992,58(6)
487
brass stubs (30 mm in diameter). The stub was then submerged in liquid nitrogen and the frozen leaf fragments fractured using a razor blade. The stub with the fractured leaf was subsequently transferred as rapidly as possible to the specimen chamber of a Jeol JSM 3S SEM with attached KEVEX' microanalyser. As the microscope used was not fitted with a cold stage, specimens were maintained in the column for approximately 20 min only, after which time they were discarded. The specimens, however, apparently remained frozen during the period of analysis. The operating conditions for analysis were as follows: emission current, 17S !-LA; Voltage, IS kV; working distance, 3S mm; take-off angle, 4So; dead time, approximately 3S%; probe raster area, 2 !-Lm2. A preset time of 30 s was used for collecting counts. The results were compared on a semi-quantitative basis by expressing the area under the Kaelemental peak of each element (elemental peak integral) as a ratio to the area under the background region 4.S - S.S keY. Harvey et al. (1981, 1986) have suggested that insufficient is known about intercellular differences in ion distribution in plant tissues to determine whether the ionic content of cells is uniform within a given population, or constant with time. Thus, although averages were used in the present study to compare ion distribution data, statistical analysis of the data is not included (Harvey et al. 1986). ATPase localization ATPase activity was localized in fully expanded leaves from both SO% seawater and salt-free treatments using a lead phosphate precipitation procedure (Hall 1971). Pieces of tissue, approximately S mm 2, were fixed in 2% glutaraldehyde in a O.OSM sodium cacodylate buffer (PH 7.2) for 1 h at 4°C. Tissue segments were then washed for 1 h in O.OSM sodium cacodylate buffer (with buffer changes every IS min) followed by two IS-min rinses in O.IM Trismaleate buffer (PH 7.2). Tissue was infiltrated at 4°C with an incubation medium consisting of 2mM ATP, 6mM MgCI2 and 3.6mM Pb(N~h in O.lM Tris-maleate buffer (PH 7.2). The tissue was then incubated at 3SoC for 30 min. Following incubation, tissue was rinsed with distilled water and post-fixed in 2% osmium tetroxide in O.OSM sodium cacodylate buffer (pH 7.2) for 2 h. Tissue was dehydrated through a graded acetone series and infiltrated and embedded in Spurr's (1969) epoxy resin. Controls consisted of leaf segmens incubated in a medium from which ATP had been omitted (Hall 1971), or pieces of leaf tissue which were boiled for 20 min prior to incubation in the ATP-
containing medium. Ultrathin sections were viewed without post-staining. Results Ion distribution gradients
The spatial relationship of the various salt gland cells is shown in Figure 1. The elemental ratios within the collecting cells were distinct from those of the adjacent spongy mesophyll cells and the stalk and excretory cens (Table 1). The chlorine ratios of the collecting cens were consistently lower than those of the spongy mesophyll and markedly lower than those of the stalk and excretory cens. They were also lower than the chlorine ratios of the xylem vessels. The potassium ratios of the collecting cells were, however, much higher than those of the spongy mesophyll cells and the stalk and excretory cells. The proportional distribution of sodium was similar to that observed for chlorine. However, much lower ratios were obtained, probably as sodium is near the lower limit of elemental detection that can be achieved using energy-dispersive X-ray analysis (postek et al. 1980). ATPase localization Staining for ATPase activity, which indicated the presence of a membrane proton pump and/or a membrane-associated soluble phosphatase (Katz et al. 1988), was marked along the plasmalemma of the collecting cells in the abaxial glands of the mature leaf (Figure 2). The cytochemical technique used, although it did not distinguish between types of phosphatase, resulted in a marked polarity in distribution of lead precipitate in the stalk cell. Heavy
Figure 1 Anatomy of the abaxial salt gland of A. marina. c, collecting cell; s, stalk cell; e, excretory cell; m, spongy mesophyll cell.
Table 1 The ratio of the elemental peak integral to the integral of the background region 4.5 - 5.5 keV for the elemental analysis of ce lls of the abaxial salt glands of leaves of A. marina grown in 50% S.W.a Cell type
n
Xylem vessel Spongy mesophyll Collecting cell Stalk/excretory cell b
4 6
3 3
Na 0.13 0.43 0.09 0.89
(0.02 (0.12 (0.07 (0.72
Cl -
0.34) 0.64) 0.12) 1.13)
1.74 4.27 0.90 7.02
(1.06 (2.56 (0.75 (5.86
• Number of cells is indicated (n), and range is given in parentheses. b Stalk cell could not always be distinguished from excretory cell.
K
-
2.85) 5.94) 0.95) 8.47)
1.17 1.51 2.12 0.78
(0.88 (0.73 (1.98 (0.50
-
1.64) 1.95) 2.32) 0.82)
488
staining occurred along the plasmalemma of the stalk cells at the stalk cell/collecting cell interface (Figure 2) and at the points of adhesion of the stalk cell plasmalemma to the cutinized lateral walls of the stalk cells (Figure 2). Weak staining was found along the plasmalemma of the excretory cells (Figure 3). The staining at this surface was much less
S.-Afr.Tydskr.Plantk., 1992, 58(6)
intense than that found at the stalk cell plasmalemma (cf. Figures 2 and 3). No staining of endomembranes in the stalk or excretory cells was apparent. The pattern and intensity of staining was similar in glands from 0 and 50% SW treatments (cf. Figures 2 and 4). Both controls showed no membrane reaction (Figure 5).
Figure 2 Portion of an abaxial gland from a mature leaf of a 50% SW plant. Dense staining for ATPase activity (-+) is associated with the collecting cell (c) plasmalemma and the stalk cell (s) plasmalemma at the collecting cell/stalk cell interface and at the plasmalemma! lateral wall adhesion (pa). Figure 3 ATPase activity (-+) localized at the surface of the excretory cells (e) of a gland from a 50% SW treatment. Figure 4 Portion of an abaxial gland from a mature leaf of a 0% SW plant. The staining pattern for ATPase activity (-+) is similar to that found in 50% SW material. c, collecting cell; s, stalk cell; e, excretory cell. Figure 5 Control tissue (spongy mesophyll) in which the ATPase has been deactivated by boiling, showing an absence of staining (cf. Figure 4). Similar results were obtained in controls when substrate was omitted.
S.AfrJ.Bot., 1992,58(6)
Discussion Salt excreted via the glands is probably derived from both the apoplast and the symplast. The low chlorine and sodium ratios of the collecting cells relative to the xylem vessels and surrounding spongy mesophyll cells suggest that the collecting cells are at the endpoint of a symplastic and possibly an apoplastic downhill ion gradient towards the glands. Shimony et al. (1973) suggested that there was a downhill gradient from the vascular bundles to the excretory surface (as opposed to the collecting cells). These authors made the assumption that the adaxial glands on the mature leaf excrete salt. However, as the leaf matures, the glands on the adaxial surface undergo autolysis (Drennan & Berjak 1982). The high ratio for potassium in the collecting cell in the absence of high concentrations of sodium and chlorine, suggests that the accumulation of potassium maintains a low osmotic potential in the collecting cells. The collecting cells probably provide the immediate source of water for movement through the gland cells. The reservoir cell of the digestive glands of Pinguicula ionantha which would be the collecting cell equivalent, is suggested to provide the immediate source of water of the excreted fluid in that species (Heslop-Harrison & Heslop-Harrison 1980). As the collecting cells in A. marina and the equivalents in many species are distinguishable from the adjacent spongy mesophyll cells only by their spatial relationship to the gland and a lack of well-developed chloroplasts (Thomson 1975), their possible role in the excretory mechanism has generally been overlooked. However, the current study suggests that the collecting cell may provide a crucial link between mesophyll salt regulation and salt export. In only one other species, Pinguicula ionantha, has a definite role been ascribed to the collecting cell equivalent (reservoir cell) which is suggested to act as a site of preferential chlorine accumulation, the chlorine subsequently being pumped through the gland on stimulation by prey capture (Heslop-Harrison & Heslop-Harrison 1980). The implied high chlorine content of the reservoir cell of P. ionantha by contrast to the measured low chlorine ratio of the A. marina collecting cell, may be a reflection of inducible ion pumps in the former as opposed to continuous-functioning ion pumps in the latter. Similarity in staining patterns for ATPase activity in non-excreting and excreting A. marina suggests that pumps in A. marina are constitutive. The mechanisms by which the characteristic ion ratios of the collecting cells are regulated are, however, not known. It would be of particular interest to ascertain whether the elemental ratios of the collecting cell vary with changes in salt loading of the surrounding symplast and apoplast, or whether they provide a 'set point' in the excretory system. The maintenance of a constant level of cr and Na+ in the collecting cell with all incoming Na+ and cr over and above that amount moving into the stalk cell, could explain how an increase in salt loading of the leaf results in an increased excretion rate (Drennan & Pammenter 1982). Furthermore, a constant K+ level in the collecting cell might result in a more concentrated saline solution being excreted under conditions of water stress. The possible significance of the ionic ratios of the collecting cell with respect to the salt and water
489
budget of the leaf underscores the need for further research in this area. The marked ultrastructural differences between the collecting and stalk cells of A. marina as well as the cutinized lateral thickenings of the stalk cell, suggest that apoplastic salt must move into the gland symplasm at least before reaching the collecting/stalk cell interface (Drennan et al. 1987). These authors suggested that loading of the stalk cell symplasm involves transport of salt ions across both the stalk cell plasmalemma and the tubular endoplasmic reticulum (tER) membranes in series. The high chlorine ratios of the stalk/excretory cells, as well as the large amount of ATPase activity localized at the stalk cell plasmalemma at the stalk cell/collecting cell interface, suggest that symplastic loading is active and possibly represents concentration against an electrochemical gradient. This would enable the formation of a downhill electrochemical gradient between the point of loading at the stalk cell and the plasmalemma of the excretory cells, as well as a gradient between the collecting cells and mesophyll cells and apoplast (xylem). The localization of some ATPase activity at the excretory cell surface suggests that the unloading of the gland is also an active process. Unloading does not necessarily occur against an electrochemical gradient. However, it is possible that energy is required for the eccrine transfer of ions across the membranes (tER and plasmalemma) at the gland surface. Faraday and Thomson (1986) have suggested that the high mitochondrial volume and elaborated surface area of the excretory cells of the salt glands of Limonium perezzi are indicative of the fact that ion efflux rather than influx is the active step in the excretory process. Similar suggestions have been made, on an ultrastructural basis, by several other researchers (Hill & Hill 1973, 1976; Shimony et al. 1973; LUuge 1975). However, trans-tissue microelectrode insertion through the salt glands of the mangrove Aegicerus corniculatum (Bostrom & Field 1973; Billard & Field 1974) have shown that a major change in potential occurs between the mesophyll and the basal gland cell (stalk cell equivalent), which led these authors to conclude that the inner basal cell membrane is of importance in the electrophysiology of the excretory process. The present study strongly suggests that although efflux from the salt glands of A . marina may be an active process, influx of ions into the stalk cell symplasm is a primary site of ion pumping. Acknowledgements We gratefully acknowledge the financial assistance of the Foundation for Research Development, Pretoria, and the technical assistance of P. Evers (EDX) of the University of Natal, Durban, Electron Microscope Unit, and J. Lawton (A TPase localization) of the University of Durban-Westville Electron Microscope Unit. C. Snyman is thanked for the Afrikaans translation of the abstract. References ARIZ, W.H., CAMPHUIS, I.J., HEIKENS, H. & VAN TOOREN, AJ . 1955. The secretion of the salt glands of Limonium lalifolium Ktze. Acla Bal. Neerl. 4: 322 - 338 .
490 ATKINSON, M.R., FINDLAY, c.P., HOPE, A.B., PITMAN, M.G., SADDLER, H.D.W. & WEST, K.R. 1967. Salt regulation in the mangroves Rhizophora mucronata Lam. and Aegialitis annulata R. Br. Aust. 1. Bioi. Sci. 20: 589 - 599. BILLARD, B. & FIELD, C.D. 1974. Electrical properties of the salt gland of Aegiceras. Planta 115: 285 - 296. BOSTROM, T.E. & FIELD, C.D. 1973. Electrical potentials in the salt gland of Aegiceras. In: Ion Transport in Plants, ed. W.P. Anderson, pp. 385 - 392. Academic Press, London. DRENNAN, P.M. & BERJAK, P. 1982. Degeneration of the salt glands accompanying foliar maturation in Avicennia marina (Forssk!l.l) Vierh. New Phytol. 90: 165 - 176. DRENNAN, P.M., BERJAK, P., LAWTON, J.R. & PAMMENTER, N.W. 1987. Ultrastructure of the salt glands of the mangrove, Avicennia marina (Forssk.) Vierh., as indicated by the use of selective membrane staining. Planta 172: 176 - 183. DRENNAN, P.M. & PAMMENTER, N.W. 1982. Physiology of salt excretion in the mangrove Avicennia marina (Forssk.) Vierh. New Phytol. 91: 597 - 606. EPSTEIN, E.L. 1972. Mineral Nutrition in Plants: Principles and Perspectives. Wiley, New York. FARADA Y, C.D. & THOMSON, W.W. 1986. Morphometric analysis of Limonium salt glands in relation to ion flux. 1. Exp. Bot. 37: 471 - 481. HALL, J.L. 1971. Cytochemical localization of ATPase activity in plant root cells. 1. Microsc. 93: 219 - 225. HARVEY, D.M.R., HALL, J.L., FLOWERS, T.J. & KENT, B. 1981. Quantitative ion localization within Suaeda maritima leaf mesophyll cells. Planta 151: 555 - 560. HARVEY, D.M.R., STELZER, R., BRANDTNER, R. & KRAMER, D. 1986. Effects of salinity on ultrastructure and ion distributions in roots of Plantago coronopus. Physiol. Pl. 66: 328 - 338. HESLOP-HARRISON, Y. & HESLOP-HARRISON, J. 1980. Chloride ion movement and enzyme secretion from the digestive glands of Pinguicula. Ann. Bot. 45: 729 - 731. HILL, A.E. & HILL, B.S. 1973. The electrogenic chloride pump of the Limonium salt gland. 1. Membr. Bioi. 12: 129 - 144. KATZ, D.B., SUSSMAN, M.R., MIERZWA, R.J. & EVERT, R.F. 1988. Cytochemical localization of ATPase activity in oat
S.-Afr.Tydskr.Plantk., 1992,58(6) roots localizes a plasma membrane-associated soluble phosphatase, not the proton pump. Pl. Physiol. 86: 841 - 847. LUTTGE, U. 1964. Untersuchungen zur Physiologie der Carnivoren-Driisen. Ber. dt. bot. Ges. 63 (Supp!.): 181 - 187. LUTTGE, U. 1966. Untersuchungen zur Physiologie der Carnivoren-Driisen. V. Mikroautoradiographische Untersuchung der Chloridsekretion durch das Driisengewebe von Nepenthes. Planta 68: 269 - 285. LUTTGE, U. 1975. Salt Glands. In: Ion Transport in Plant Cells and Tissues, eds. D.A. Baker & J.L. Hall, pp. 335 - 336. North Holland Publishing Company, Amsterdam. LUTTGE, U. & OSMOND, C.B. 1970. Ion absorption in Atriplex leaf tissue. III. Site of metabolic control of light-dependent chloride secretion to epidermal bladders. Aust. 1. bioi. Sci. 23: 17 - 25. POSTEK, T.P., HOWARD, K.S., JOHNSON, A.H. & McMICHAEL, K.L. 1980. Scanning Electron Microscopy. A Student's Handbook. Ladd Research Industries, Inc. SCHOLANDER, P.F., BRADSTREET, E.D., HAMMEL, H.T. & HEMMINGSON, E.A. 1966. Sap concentrations in halophytes and some other plants . Pl. Physiol. 41: 529 - 532. SCHOLANDER, P.F., HAMMEL, H.T., HEMMINGSON, E. & GAREY, W. 1962. Salt balance in mangroves. Pl. Physiol. 3: 722 - 729. SHIMONY, c., FAHN, A. & RHEIN HOLD, L. 1973. Ultrastructure and ion gradients in the salt glands of A vicennia marina (Forssk.) Vierh. New Phytol. 72: 27 - 36. SPURR, A.R. 1969. A low viscosity epoxy resin embedding medium for electron microscopy. 1. Ultrastruet. Res. 26: 31 - 43. THOMSON, W.W. 1975. The structure and function of salt . glands. In: Plants in Saline Environments, Ecological Studies 15, eds . A. Poljakoff-Mayber & J. Gale, pp. 118 - 146. Springer Verlag, Berlin, Heidelberg. THOMSON, W.W., FARADAY, C.D. & OROS, J.w. 1988. Salt glands. In: Solute Transport in Plant Cells and Tissues, eds. D.A. Baker & J.L. Hall, ch. 13. Longman Scientific and Technical, Essex. WAISEL, Y., ESHEL, A. & AGAMI, M. 1986. Salt balance of the leaves of the mangrove Avicennia marina. Physiol. Pl. 67: 67 - 72.