Comparative Biochemistry and Physiology Part A 130 Ž2001. 461᎐470 Changes in major intracellular osmolytes in L-929 cells following rapid and slow application of hyperosmotic media 夽 C. Libioulle a , L. Corbesier b, R. Gilles a,U a b Laboratory of Animal Physiology, Uni¨ ersity of Liege, ` 22, quai Van Beneden, B-4020 Liege, ` Belgium Laboratory of Plant Physiology, Uni¨ ersity of Liege, ` Sart Tilman Campus (B22), B-4000 Liege, ` Belgium Received 9 March 2001; received in revised form 7 June 2001; accepted 11 June 2001 Abstract Cultured L-929 cells respond to media-made hyperosmotic Ž600 mOsmolrkg H 2 O. by addition of NaCl, sorbitol or proline by adjusting successively their intracellular level in different osmolytes: Naq, Kq, amino acids and sorbitol. In the NaCl medium, Naq and Kq are first to increase. Their concentration is then down-regulated while they are replaced by less disrupting osmolytes: amino acids and sorbitol. The amino-acid level is also adjusted with respect to the increase in sorbitol which starts only after 24 h, depending on the induction of aldose reductase. A similar evolution in the amount of these osmolytes is observed, with different time scales and amplitudes, depending on whether the osmotic shocks are applied abruptly or slowly, in a more physiological way. The interplay between the osmolytes is also different depending on their availability in the external medium. Such complex evolutions indicate that a cascade of interacting signals must be considered to account for the overall regulation process. It can hardly be fitted into a model implicating a single primary signalling event Žearly increase in ions or decrease in cell volume. as usually postulated. Also, the volume up-regulation is not significantly different in the different conditions, showing that it is not primarily dependent on the adjustment of the intracellular osmolarity which is reached immediately upon cell shrinkage and is maintained all over, independently of the availability and changes in nature of the osmolytes. 䊚 2001 Elsevier Science Inc. All rights reserved. Keywords: L-929 cells; Osmolytes; Naq; Kq; Amino acids; Sorbitol; Hyperosmotic media 1. Introduction Prokaryotic and eukaryotic cells respond to high external osmolarities by an accumulation of or- 夽 This paper was originally presented at a symposium dedicated to the memory of Marcel Florkin, held within the Belgium, July ESCPB 21st International Congress, Liege, ` 24᎐28, 2000. U Corresponding author. Tel.: q32-4366-5005; fax: q324366-5020. E-mail address: r.gilles@ulg.ac.be ŽR. Gilles.. ganic osmotic effectors such as amino acids, polyols, or oligosaccharides, thus adjusting their intracellular osmolarity to that of the external medium. As shown mostly over the last decade, increased expression of genes coding for enzymes and transporters is important in controlling the level of these compounds Žsee, for instance: GarciaPerez et al., 1989; Csonka and Hanson, 1991; Uchida et al., 1993; Yamauchi et al., 1993; Albertyn et al., 1994.. Numerous papers have been published on the intracellular signaling events involved in this process and leading to increased 1095-6433r01r$ - see front matter 䊚 2001 Elsevier Science Inc. All rights reserved. PII: S 1 0 9 5 - 6 4 3 3 Ž 0 1 . 0 0 4 1 5 - 9 462 C. Libioulle et al. r Comparati¨ e Biochemistry and Physiology Part A 130 (2001) 461᎐470 transcription Žsee, for instance: Takenaka et al., 1994; Ferraris et al., 1996; Loomis et al., 1997; Zhou and Cammarata, 1997; Aida et al., 1999; for recent reviews: Burg et al., 1997; Wood, 1999., but the question of what primarily triggers these sequences of events remains poorly documented. According to current views, the primary triggering signalŽs. could be variations in intracellular levels of inorganic ions Žor corresponding changes in ionic strength., andror changes in cell volume occurring in cells subjected to a hyperosmotic challenge. The ‘triggering-ion hypothesis’ suggests that the expression level of the enzymes and transporters involved in osmotic adjustments, and thus allowing cell volume control, is somehow determined by the levels of the major inorganic ions Žor by the ionic strength.. Epstein Ž1986. originally proposed, on the basis of his work on E. coli, that the magnitude of the cell response depends on the magnitude of the initial shock and is elicited by the major and prolonged Kq concentration changes occurring upon application of the hyperosmotic medium, once a certain threshold is reached Žsee also Uchida et al., 1989 for support to this hypothesis in a mammalian cell system.. In bacteria, where ‘osmotic’ regulation of gene transcription has been much studied, the Kq concentration usually remains higher than in control cells after acclimation to a new, hyperosmotic medium ŽEpstein and Schultz, 1965; Dinnbier et al., 1988.. This, however, is not the case in a variety of animal tissues and cell types. There, levels of the major inorganic ions are quite well regulated and thus not always significantly different from control values after acclimation to the hyperosmotic environment ŽBagnasco et al., 1987; Gilles, 1987; Libioulle et al., 1996.. It would thus appear that, as animal cells acclimate, inorganic ion levels tend to return to control values while the osmotic gap left by their decrease is ‘compensated’ by increased concentrations of certain so-called ‘compensatory’ organic osmolytes. In this view, as previously mentioned ŽLibioulle et al., 1996., it remains possible that early, major changes in ion levels occurring upon immediate application of the hyperosmotic medium serve as major triggering signals perceived by the cells. To gain more insight into this problem, we have studied in L-929 cells, over the first 24 h after application of various hyperosmotic media, the changes in different parameters related to osmotic and volume control. The osmotic up-shifts were applied in two different ways: abruptly, as in all previous studies, in a single, rapid step inducing large changes in volume, ion levels, and ionic strength, or very slowly, in a more physiological manner, so that the changes could be expected to be quite minimal and to occur over a much longer time scale. We monitored in these cells, upon application of media which was made hyperosmotic by the addition of NaCl or sorbitol, the changes in the following parameters: water content, levels of the major inorganic cations ŽNaq and Kq. , levels of the principal organic osmotic effectors Žamino acids and sorbitol.. The reasons for using sorbitol in the external medium were that it increases the osmolarity without markedly affecting the INrOUT NaCl gradient and it is the major organic osmolyte found inside the cells after long-term hyperosmotic adaptation ŽLibioulle et al., 1996.. We also monitored under the same conditions induction of aldose reductase, the enzyme catalyzing the synthesis of sorbitol from glucose and abundantly expressed in L-929 cells after hyperosmotic acclimation ŽLibioulle et al., 1996.. In some experiments, the medium was made hyperosmotic with proline. We have previously shown that cells long acclimated to high-proline media essentially accumulate proline instead of sorbitol and do not display induced aldose reductase expression ŽLibioulle et al., 1996.. 2. Materials and methods 2.1. Cell cultures Mouse L-929 cells, derived from subcutaneous areolar and adipose tissue, were obtained from the American Type Culture Collection. They were cultured to confluence at 37⬚C as previously described ŽLibioulle et al., 1996. in sealed 175 cm2 Falcon flasks containing Dulbecco’s modified Eagle’s medium at pH 7.4 ŽDMEM, Flow.. The control medium was prepared from powdered DMEM containing 2.3 mM glutamine and no sodium bicarbonate ŽICN.. This medium was supplemented with 10% fetal calf serum, 20 mM C. Libioulle et al. r Comparati¨ e Biochemistry and Physiology Part A 130 (2001) 461᎐470 HEPES, 13 mM NaHCO3 , and 0.6% Žvrv. antibiotic-antimycotic solution ŽGibco-BRL. containing 10 000 Urml penicillin G, 10 mgrml streptomycin sulfate, and 25 ␮grml amphotericin B fungicide. The osmolality of this medium was 300 mOsmolrkg. Hyperosmotic media Ž600 mOsmolrkg. were made by adding to the control medium either NaCl Ž160 mM., sorbitol Ž280 mM., or proline Ž273 mM.. Proliferation rates generally dropped and many cells died during the first transfers. The cells were considered acclimated after approximately five passages, by which time they had resumed a normal growth rate. For subculture and some experimental procedures, the confluent cell monolayers were suspended by trypsinization in the adequate medium before seeding or centrifugation and further treatment. 463 Fig. 1. Changes in the water content of L-929 cells subjected abruptly Ž䢇,`. or slowly Ž',^. to an osmotic up-shift Žfrom 300 to 600 mOsmolrl. achieved by adding NaCl Ž^,`. or sorbitol Ž',䢇. to the control medium. Results are means " S.D. of four experiments. ---I---: evolution of external medium osmolarity during slow changes. 2.2. Abrupt and slow shocks Abrupt osmotic up-shifts were achieved simply by replacing the control medium with a hyperosmotic one. Slow up-shifts were achieved by adding the hyperosmotic medium at a constant flow rate of 145 ␮lrmin to Falcon flasks initially filled with 100 ml control medium, while maintaining the volume constant. The flasks were subjected to slow rotary shaking at 45 motions per min. Under these conditions, the osmolality of the medium increased slowly, reaching 565 mOsmolrkg after 24 h Žsee Fig. 1.. 2.3. Intracellular water, Naq, and K q contents The same samples were used for all measurements. The cells were grown under the conditions described above and pelted by centrifugation Ž5 min at 300 = g .. They were then washed three times in HBSS containing Ca2q, Mg 2q, the amount of the appropriate solute necessary to maintain the desired osmolarity, and inulin Ž1% wrv., used as an extracellular space marker. The wet pellets were weighed, then dried at 110⬚C to constant weight in order to estimate the total amount of water in the wet pellet. Inulin and ions were then extracted for 24 h from the dry pellets in 1 ml distilled water and their amounts measured in the supernatant after centrifugation at 10 000 = g for 10 min. Naq and Kq were assayed by flame photometry. The amount of inulin was measured as in Roe et al. Ž1949.. Intracellular contents were derived from calculations taking into account the extracellular space determined from the inulin measurements. 2.4. Protein electrophoresis Total cell lysate proteins were separated by electrophoresis in a 20% polyacrylamide separating gel preceded by a 3.4% polyacrylamide stacking gel Ž140 = 180 = 1.5 mm.. Each well received 130 ␮g protein ŽLibioulle et al., 1996.. The gels were fixed and stained for 1 h in a mixture of waterracetic acidrmethanol Ž45:10:45. supplemented with 2% Coomassie Blue ŽServa.. The gels were destained by repeated soaking in a m ixture of waterracetic acidrm ethanol Ž70:10:20.. 2.5. Extraction for organic osmolytes analysis The cells were washed six times with phosphate-buffered saline in their Falcon culture flasks ŽNaCl or sucrose was added to reach 600 mOsmolrkg when necessary ., in order to clear the extracellular space of the organic solutes of the medium. After washing, the cells were scraped, lysed for 10 min in 10 ml ice-cold water, and transferred with two rinses of 5 ml ice-cold water in a Kontes glass᎐glass homogeniser. After 20 strokes, the resulting homogenates were centrifuged at 12 000 = g for 10 min at 4⬚C. A small 464 C. Libioulle et al. r Comparati¨ e Biochemistry and Physiology Part A 130 (2001) 461᎐470 aliquot of supernatant was kept for protein measurements ŽBradford, 1976. and the rest was cleared of protein by precipitation with 7% trichloracetic acid ŽTCA.. The TCA was then extracted with a tri-n octylaminerfreon solution Ž26% vrv. ŽBagnasco et al., 1987.. The aqueous upper phase was recovered, frozen at y80⬚C, and lyophilised. 2.6. Sorbitol analysis Lyophilised samples were dissolved in 400 ␮l water and filtered on a 0.2-␮m LC13PVDF Acrodisc ŽGelman.. Sorbitol was assayed in 20 ␮l aliquots by differential refractometry following high performance liquid chromatography on Aminex HPX-87C ŽBio-Rad. in a column Ž300 = 7.8 mm i.d.. kept at 80⬚C under a water flow rate of 0.8 mlrmin ŽLejeune et al., 1991.. medium induced a similar immediate major increase in Kq. The Kq concentration then decreased to near the control value within 24 h, even though the volume-restoring process was far from completed. Interestingly, the initial step of this down-regulation appeared faster in NaCl medium than in sorbitol medium. When the medium was made hyperosmotic by addition of NaCl, the Naq concentration also increased. This increase was much slower than the Kq increase, however, and continued for approximately 4 h. No significant decrease occurred over the 20 remaining h of the experiments. When the medium was made hyperosmotic with sorbitol, no significant change in the Naq concentration was observed from the time of the first measurements Ž15 min.. When the osmotic up-shifts were ap- 2.7. Amino acid analysis Lyophilised samples were dissolved in 1 ml water. A 1 ␮l aliquot of each sample was used for amino acid determination and quantification by means of a 130A PTC-amino acid analyzer equipped with a 920A Data Module ŽApplied Biosystems, Perkin Elmer.. 3. Results 3.1. Hyperosmotic shocks with NaCl and sorbitol 3.1.1. Inorganic ions and cell ¨ olume As shown in Fig. 1, the water content of the cells decreased immediately after quick, one-step application of a medium, which was made hyperosmotic by the addition of NaCl or sorbitol. The cells remained shrunken for approximately 1 h, then slowly started volume-regulating. Over the 24-h span of our experiments, the volume was not totally restored. When the osmolarity was increased gradually, the cell water content decreased slowly, reaching a minimum after approximately 8 h. The cells then initiated a slow volume-regulating process that was not completed within the 24-h span of the experiments. During these experiments, the intracellular Naq and Kq concentrations evolved as the water content changed. As shown in Fig. 2a, an abrupt one-step application of either hyperosmotic Fig. 2. Changes in intracellular Naq Ž䢇,`. and Kq Ž',^. levels in L-929 cells subjected abruptly Ža. or slowly Žb. to an osmotic up-shift Žfrom 300 to 600 mOsmolrl. achieved by adding NaCl Ž^,`. or sorbitol Ž',䢇. to the control medium. Results are means " S.D. of four experiments. C. Libioulle et al. r Comparati¨ e Biochemistry and Physiology Part A 130 (2001) 461᎐470 465 plied slowly, changes in intracellular ion concentrations, though following the same general pattern, were much less pronounced ŽFig. 2b.. Kq increased from approximately 125 to 150 mEqrl intracellular water over an 8-h period in either medium, as opposed to the immediate rise to approximately 240 mEqrl intracellular water observed when the shift was abrupt. Here again, the Naq concentration was not significantly altered when sorbitol was the osmolarity-increasing agent. A slow and slight increase was observed over the first 8 h in NaCl medium, after which the Naq level remained steady but lower than in the case of an abrupt osmolarity up-shift. 3.1.2. Aldose reductase expression As previously shown ŽLibioulle et al., 1996., acclimation to media, which is made hyperosmotic with NaCl or sorbitol results in induction of aldose reductase ŽAR. enzyme catalyzing the synthesis of sorbitol from glucose. Sorbitol is in fact the major osmolyte accumulated in L-929 cells long acclimated to such media. On SDSpolyacrylamide gels, AR can be simply visualized as a band at 40 kDa. Since it has been shown previously ŽLibioulle et al., 1996. that cells showing no AR band on gels have no detectable AR enzyme activity, we have simply monitored AR induction over the first 24 h of exposure to the hyperosmotic conditions by its appearance on the electrophoretic gels. Fig. 3a shows the appearance of the AR-containing protein band 10᎐15 h after an abrupt shift to high-NaCl medium. The enzyme also appeared within 24 h after an abrupt shift to high-sorbitol medium Žnot shown.. This confirms results obtained previously on long-acclimated cells ŽLibioulle et al., 1996.. As shown in Fig. 3b, AR expression was similar for both hyperosmotic media over a 24-h period when the osmolarity was increased slowly. 3.1.3. Sorbitol and amino acid concentrations As mentioned above, AR did not appear on the gels until approximately 15 h post-shift. By this time the Kq concentration had already dropped considerably and the cell water content was increasing. This could mean that molecules other than sorbitol play a part in the intracellular osmolarity increase that must take place to account for concomitant volume up-regulation and Kqlevel reduction. Ion and water content changes Fig. 3. Electrophoretic profiles of L-929 cells subjected to different conditions. Ža. Evolution after abrupt application of high-NaCl medium. For control and long-acclimated cells refer to part b. Žb. Effect, over 24 h, of high-NaCl or highsorbitol medium applied either slowly Žslow. or abruptly Žfast.. For NaCl fast, refer to part a Ž20 h.. AR: aldose reductase. being greatest when the hyperosmotic shocks were applied rapidly, we concentrated on this situation, measuring the evolution of the levels of sorbitol and free amino acids. As shown in Fig. 4, sorbitol remained undetectable during the first 24 h in high-NaCl medium. It did not increase until later, slowly reaching, more than 48 h post-shift, the concentration found in acclimated cells. The sorbitol concentration increased far more rapidly in cells 466 C. Libioulle et al. r Comparati¨ e Biochemistry and Physiology Part A 130 (2001) 461᎐470 Fig. 4. Changes in intracellular sorbitol Ž䢇,`. and amino acid Ž',^. levels in L-929 cells subjected abruptly to an osmotic up-shift Žfrom 300 to 600 mOsmolrl. achieved by adding NaCl Ž^,`. or sorbitol Ž',䢇. to the control medium. Results are means " S.D. of four experiments. exposed to the high-sorbitol medium, reaching within 4 h the level observed in cells fully acclimated to NaCl medium. Interestingly, AR was induced in high-sorbitol medium exactly as in the high-NaCl medium Žsee point 2 above.. In high-NaCl medium, free amino-acid levels increased considerably over the first 24 h ŽFig. 4., then dropped as the sorbitol level began to rise. During a sorbitol shock, the increase in amino acids was lesser, reaching a much lower maximum level after 4 h. By this time the intracellular sorbitol concentration was already quite high ŽFig. 4.. 3.1.4. Complementary experiments with proline The above results point to an interaction between sorbitol and amino acids in the response of L-929 cells to hyperosmotic media. We therefore studied the behavior of these cells in a medium made hyperosmotic by addition of proline, an amino acid that they can tolerate at high concentration and which readily enters the intracellular fluid. Previous experiments have shown that L-929 cells can be acclimated for long periods Žyears. to this hyperosmotic proline medium ŽLibioulle et al., 1996.. Fig. 5 summarizes the changes in cell water content and ion concentrations upon abrupt application of high-proline medium. As observed with the high-NaCl and high-sorbitol media, a large immediate decrease in water content was recorded, concomitant with an increase in Kq. As with the high-sorbitol medium, no significant change in Naq was observed. As in the other media tested, slow volume regulation was observed, concomitantly with a rapid decrease in Kq and a major increase in proline, which approached equilibrium with the external medium within 24 h. Sorbitol remained undetectable in these experiments and AR was never expressed, either transiently during the first 24 h of shock Žpresent work: not shown. or in cells acclimated for long periods Žmonths: Libioulle et al., 1996.. AR expression was observed, however, when cells long acclimated to high-proline medium were subjected for 24 h to an abrupt INrOUT Naq gradient increase. The gradient increase was produced in two different ways: either iso-osmotically Ž600 mOsmol., by adding NaCl for 200 mOsmol and decreasing the amount of proline by 200 mOsmol, or by up-shifting the osmolarity to 800 mOsmol, by simply adding sufficient NaCl to the high-proline medium ŽFig. 6.. Under iso-osmotic conditions, only a slight, if significant, increase in intracellular Naq was recorded over 24 h Žfrom 44.4" 18.7 to 52.7" 6.6 mmolrkg cell water.. No change in the Kq level or cell water content was recorded Žnot shown.. 4. Discussion The above results demonstrate the complexity of the response of L-929 cells to a shift to hyperosmotic conditions. They also indicate close interactions between different mechanisms for con- Fig. 5. Changes in cell water content Ž䉫. and in intracellular levels of Naq Ž䢇., Kq Ž'., and proline ŽI. in L-929 cells subjected abruptly to an osmotic up-shift Žfrom 300 to 600 mOsmolrl. achieved by adding proline to the control medium. Results are means " S.D. of four experiments. C. Libioulle et al. r Comparati¨ e Biochemistry and Physiology Part A 130 (2001) 461᎐470 Fig. 6. Effect on AR induction, over 24 h, of an INrOUT Naq gradient increase applied abruptly to high-prolineacclimated cells. AR: aldose reductase. NaCl-H: hyperosmotic Žaddition of NaCl to the high-proline medium.. Na-I: isoosmotic Žpartial replacement of proline with NaCl.. trolling the levels of the major effectors of osmotic and volume regulation. To illustrate this, let us first consider immediate, one-step transfer of the cells to high-NaCl medium. Here, clearly, increases in the levels of both sorbitol and amino acids allow up-regulation of volume and downregulation of ion concentrations. Furthermore, since sorbitol increases significantly only after 24 h, it appears that an early increase in amino acids is responsible for the early steps of these regulatory processes. Once the sorbitol level starts to rise, amino acid levels are down-regulated. Interestingly in this case, the Naq and Kq levels evolve quite differently from the time of our earliest measurements Ž15 min post-shift. onward. By 15 min Kq has risen markedly. Only part of this increase can be ascribed to the osmotic efflux of water and to cell shrinkage. The Kq level then drops rapidly, as the Naq level increases. This phase is followed by a further, slower decrease in Kq, resulting in a level close to the control value 24 h post-transfer. Over this period, the Naq level does not change significantly. The Naq concentration must, however, be down-regulated later on since, as we have previously shown, it resumes control values in cells acclimated to this hyperosmotic medium ŽLibioulle et al., 1996.. The early decrease in Kq 467 might be related to the increase in Naq. It could involve, for instance, adjustments in the activity of the NaqrKqATPases. This, however, might not be the case, since rapid down-regulation of Kq is also observed in high-sorbitol and high-proline media where the Naq level does not change significantly. In these media we note an increase in organic osmotic effectors Žeither sorbitol or amino acids. that is faster than their increase in highNaCl medium. It is also worth noting that the early decrease in Kq is faster in high-NaCl or high-proline medium than in high-sorbitol medium. This could be indicative of two different processes, one depending on an increased INrOUT Naq gradient, the other somehow related to the availability of an organic osmotic effector Žproline: fast; sorbitol: slower.. These results clearly show interactions between at least four systems: one involved in rapid down-regulation of the Kq level, one in slower down-regulation of the Kq and Naq levels, one in the up-, then down-regulation of amino acid levels, and one in up-regulation of the sorbitol level. Our results further show that these interactions differ according to the availability of osmotic effectors. When sorbitol is present externally for instance, it increases in the intracellular medium much faster than in cells having to cope with the high NaCl medium; amino acid levels also rise more slowly on high-sorbitol medium. Apparently, the cells do not take advantage of the greater availability of sorbitol to regulate their volume faster. They rather decrease the activity of the pathways involved in increasing amino acid levels, so that no significant change in the speed of volume readjustment is observed. The rapid increase in osmolyte levels Žsorbitol q amino acids. would first allow a rapid decrease in Kq. Similarly, when proline is present in the external medium, volume up-regulation is not significantly faster despite the rapid increase in intracellular proline. In this situation, sorbitol synthesis is totally depressed. The cells essentially use proline as organic osmotic effector and its rapid increase in concentration first compensates for the decrease in Kq. Clearly then, the volume evolution is not dependent on the adjustment of the intracellular osmolarity which is reached almost immediately by cell shrinkage and is then maintained all over, independently of the availability and changes in nature of the osmolytes. 468 C. Libioulle et al. r Comparati¨ e Biochemistry and Physiology Part A 130 (2001) 461᎐470 As mentioned in the introduction, transient major changes in intracellular cation levels andror cell volume have been viewed as major primary signals triggering cell volume up-regulation. The complex interplay between different major osmotic effectors in L-929 cells can hardly fit such a simple model. Clearly, a cascade of signals must be considered to account for such interactions. What these signals are and in which processes, exactly, they intervene, remains unclear. In our abrupt osmotic up-shift experiments, the earliest osmoregulatory event recorded was a fast down-regulation of the Kq concentration. This was followed by slow up-regulation of cell volume. These first steps reflect the increasing levels of other osmotic effectors: Naq and amino acids in high-NaCl medium, amino acids or sorbitol in high-proline or high-sorbitol medium. In the case of adaptation to high NaCl, the increase in inorganic ions has been classically associated with increased amino acid levels. A variety of enzymes of amino-acid metabolism are indeed affected by an increase in NaCl or KCl. An increase in the INrOUT Naq gradient also increases amino acid uptake from the external medium. These changes may account for early up-regulation of the amino acid pool Žsee, for instance, Gilles and Delpire, 1997 for a review of the problem.. This idea is in agreement with the fact that amino acid levels do not rise markedly in high-sorbitol medium, since in this medium there is no change in the INrOUT Naq gradient and an early but transient rise in Kq is the only marked change in intracellular ion levels. On the other hand, the availability of sorbitol early in the process may account for down-regulation of the Kq level, just as this down-regulation may be linked, in high-NaCl medium, to an increase in Naq and amino acids. In high-NaCl medium, sorbitol intervenes much later. Its concentration increase does not become significant until 48 h post-shift and is concomitant with a decrease in amino acid levels. The increase in sorbitol might thus be a signal to initiate the decrease in amino acids. To our knowledge, however, nothing is currently known as to a possible repressive effect of sorbitol on amino acid synthesis or transport in the framework of osmotic regulation. Nor is there any information regarding primary signalŽs. triggering down-regulation of the Kq level. As we have seen, this down-regulation is initiated in L-929 cells concomitantly with an increase in either Naq, amino acids, or sorbitol. It also starts prior to any significant volume regulation, which suggests that volume regulation plays no part. It is unclear how sorbitol and amino acids might affect, directly or indirectly, the activity of the transport pathways involved in the decrease in Kq. Interestingly sorbitol, although it appears to depress the increase in amino acids, does not depress its own synthesis. Aldose reductase, the enzyme catalyzing its synthesis from glucose, is expressed in L-929 cells in both high-NaCl and high-sorbitol medium. Some early signal must be involved, however, in regulating AR expression and sorbitol accumulation, since the enzyme is not induced and sorbitol does not accumulate in a high-proline medium. The signal does not seem to be the large increase in inorganic ions or the concomitant rapid decrease in volume occurring early upon immediate, one-step transfer to hyperosmotic medium. These changes are similar, both in magnitude and timing, whether AR is induced and sorbitol accumulates or not. Furthermore, AR is induced when sorbitol or NaCl is added slowly to the medium, even though ion-level and volume changes are slower and less marked than after an abrupt shift, even to high-proline medium in which AR is not induced. A possibility is that in high-proline medium, the major early increase in amino acids might prevent AR induction. This seems unlikely, however, since AR is induced in cells acclimated to high proline, when NaCl is abruptly added either iso-osmotically, in replacement of part of the proline, or in the same amount, as an additional osmolyte causing an osmotic up-shift. In both situations levels of amino acids, particularly proline, are high from the start. Another argument against the idea of an adjustment of AR expression to the intracellular level of inorganic ions is of the course the fact that AR remains high in cells long acclimated to high-NaCl medium, where both cell volume and intracellular ion concentrations have been regulated ŽLibioulle et al., 1996.. Maintenance of high AR expression after regulation of ion levels has also been shown in GBR-PAP cells ŽBagnasco et al., 1987.. Osmolarity per se is another clear candidate as a primary triggering signal for AR induction and maintenance of high expression. In PAP cells, Uchida et al. Ž1989. demonstrated a direct relation between the level of AR and the osmolarity C. Libioulle et al. r Comparati¨ e Biochemistry and Physiology Part A 130 (2001) 461᎐470 of the external medium after a week of acclimation, i.e. after regulation of ion levels. This seems unlikely in L-929 cells: the enzyme is not induced in high-proline medium despite the high osmolarity of this medium, and despite the apparent absence of repression of AR synthesis by proline or other amino acids. Changes in the external level of NaCl are another possibility. A direct effect of NaCl seems unlikely, however, since AR is also induced on high-sorbitol medium. Ion effects could also be indirect, leading for instance to changes in membrane potential. In this framework, the immediate hyperpolarizing effects of one-step application of the hyperosmotic medium should be approximately the same for all three media used, the volume decrease and concomitant rise in intracellular Kq being in the same range. In preliminary long-term experiments, however, when acclimated cells were given time to regulate both their volume and ion concentrations, L-929 cells acclimated to high NaCl or high sorbitol Žwhere AR remains expressed. appeared to be less polarized than cells acclimated to high-proline or control medium Žwhere AR is not expressed.. The recorded membrane potential values are: control cells: y16.3" 7.4 mV Ž n s 18.; high-proline cells: y17.7" 9.6 mV Ž n s 12.; high-NaCl cells: y5.4" 1.9 mV Ž n s 21.; high-sorbitol cells: y9.5" 5.1 mV Ž n s 22.. This indicates that changes in membrane potential could be important signals involved in maintaining high AR expression in acclimated cells, and possibly involved in other events associated with volume up-regulation. These signals could be induced by changes in the external NaCl level, in the activity of transport pathways, andror in membrane permeability. In relation to this, recent experiments show that activation of the E. coli proline transporter ProP, reconstituted in proteoliposomes, depends on both a membrane potential and a hyperosmotic up-shift ŽRacher et al., 1999.. ProP is a major transporter involved in proline accumulation in E. coli during hyperosmotic acclimation. To our knowledge, this is the first experimental evidence that a protein can act in both osmosensing and osmoregulation once a membrane potential and an osmotic shift have been established. The idea that the membrane potential might act in L-929 cells as a ‘permissive’ signal necessary for induction of osmoregulatory processes by events such as those mentioned 469 above must of course be confirmed by further experiments. In summary, it is clear that a cascade of different signals eliciting different specific responses must account for the complexity and flexibility of the adaptive processes involved in coping with a hyperosmotic shock. Our results suggest the possibility that a change in membrane potential, combined with events related to the osmolarity of the medium and to the nature of the osmolytes present, might affect the macromolecules involved in osmoregulation and thereby trigger osmoregulatory mechanisms. We cannot exclude, however, that an early change in Kq Žor ionic strength. might also serve as an ‘initiating’ or ‘permissive’ signal. If so, our results on AR induction following a gradual osmolarity up-shift suggest that this signal would become effective at quite a low threshold value that could be reached very slowly, over several hours. Our results anyway exclude the possibility that the regulatory mechanisms involved adjust simply and directly to the magnitude of the change in Kq Žor ionic strength.. They are indicative of a far more complex mechanism implicating a cascade of several, interacting, signals. Acknowledgements This work has been aided by University and FNRS grants to R.G., L.C. was responsible for the sorbitol analysis. Amino acids analysis have been run by the laboratory of Dr J. Van Beeumen ŽEiwitbiochemie-Gent Universiteit.. References Aida, K., Tawata, M., Ikegishi, Y., Onaya, T., 1999. Induction of rat aldose reductase gene transcription is mediated through the cis-element, osmotic response element ŽORE.: increased synthesis andror activation by phosphorylation of ORE-binding is a key step. Endocrinology 140 Ž2., 609᎐617. Albertyn, J., Hohmann, S., Thevelein, J.M., Prior, B.A., 1994. GPD1, which encodes glycerol-3-phosphate dehydrogenase, is essential for growth under osmotic stress in Saccharomyces cere¨ isiae, and its expression is regulated by the high-osmolarity glycerol response pathway. Mol. Cell. Biol. 14 Ž6., 4135᎐4144. Bagnasco, S.M., Uchida, S., Balaban, R.S., Kador, P.F., Burg, M.B., 1987. Induction of aldose reductase and 470 C. Libioulle et al. r Comparati¨ e Biochemistry and Physiology Part A 130 (2001) 461᎐470 sorbitol in inner medullary cells by elevated extracellular NaCl. Proc. Natl. Acad. Sci. USA 84, 1718᎐1720. Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of ␮g quantities of protein utilizing the principle of dye-binding. Anal. Biochem. 72, 248᎐254. Burg, M.B., Kwon, E.D., Kultz, ¨ D., 1997. Regulation of gene expression by hypertonicity. Ann. Rev. Physiol. 59, 437᎐455. Csonka, L.N., Hanson, A.D., 1991. Prokaryotic osmoregulation: genetics and physiology. Annu. Rev. Microbiol. 45, 569᎐606. Dinnbier, U., Limpinsel, E., Schmid, R., Bakker, E.P., 1988. Transient accumulation of potassium glutamate and its replacement by trehalose during adaptation of growing cells of Escherichia coli K-12 to elevated sodium chloride concentrations. Arch. Microbiol. 150, 348᎐357. Epstein, W., 1986. Osmoregulation by potassium transport in Escherichia coli. FEMS Microbiol. Rev. 39, 73᎐78. Epstein, W., Schultz, S.G., 1965. Cation transport in Escherichia coli: V. Regulation of cation content. J. Gen. Physiol. 49, 221᎐234. Ferraris, J.D., Williams, C.K., Jung, K.Y., Bedford, J.J., Burg, M.B., Garcia-Perez, A., 1996. ORE, a eukaryotic minimal essential osmotic response element. The aldose reductase gene in hyperosmotic stress. J. Biol. Chem. 271 Ž31., 18318᎐18321. Garcia-Perez, A., Martin, B., Murphy, H.R. et al., 1989. Molecular cloning of cDNA coding for kidney aldose reductase. J. Biol.Chem. 264, 16815᎐16821. Gilles, R., 1987. Volume control and adaptation to changes in ion concentrations in cells of terrestrial and aquatic species: clues to cell survival in anisosmotic media. In: Dejours, P., Bolis, L., Taylor, C.R., Weibel, E.R. ŽEds.., Comparative Physiology: Life in Water and on Land. Springer Verlag, Heidelberg, pp. 485᎐502. Gilles, R., Delpire, E., 1997. Variations in salinity, osmolarity and water availability. In: Dantzler, W.H ŽEd.., Handbook of Physiology, section 13: Comparative Physiology, 2. Oxford University Press, New York, Oxford, pp. 1523᎐1586. Lejeune, P., Bernier, G., Kinet, J.M., 1991. Sucrose levels in leaf escudate as a function of floral induction in the long day plant Sinapsis alba. Plant Physiol. Biochem. 29, 153᎐157. Libioulle, C., Llabres, G., Gilles, R., 1996. Protein patterns, osmolytes and aldose reductase of l-929 cells exposed to hyperosmotic media. J. Cell Physiol. 168, 147᎐154. Loomis, W.F., Shaulsky, G., Wang, N., 1997. Histidine kinases in signal transduction pathways of eukaryotes. J. Cell Sci. 110, 1141᎐1145. Racher, K.I., Voegele, R.T., Marshall, E.V. et al., 1999. Purification and reconstitution of an osmosensor: transporter ProP of Escherichia coli senses and responds to osmotic shifts. Biochemistry 38, 1676᎐1684. Roe, J.H., Epstein, J.H., Goldstein, N.P., 1949. A photometric method for the determination of inulin in plasma and urine. J. Biol. Chem. 178, 838᎐845. Takenaka, M., Preston, A.S., Kwon, H.M., Handler, J.S., 1994. The tonicity-sensitive element that mediates increased transcription of the betaine transporter gene in response to hypertonic stress. J. Biol. Chem. 269 Ž47., 29379᎐29381. Uchida, S, GarciaPerez, A, Murphy, A.H., Burg, M.B., 1989. Signal for induction of aldose reductase in renal medullary cells by high external NaCl. Am. J. Physiol. 256, C614᎐C620. Uchida, S., Yamauchi, A., Preston, A.S., Kwon, H.M., Handler, J.S., 1993. Medium tonicity regulates expression of the Naq and Cly-dependent betaıne ¨ transporter in Madin-Darby Canine Kidney cells by increasing transcription of the transporter gene. J. Clin. Invest. 91, 1604᎐1607. Wood, J.M., 1999. Osmosensing by bacteria: signals and membrane-based sensors. Microbiol. Mol. Biol. Rev. 63, 230᎐262. Yamauchi, A., Uchida, S., Preston, A.S., Kwon, H.M., Handler, J.S., 1993. Hypertonicity stimulates transcription of gene for Naq-myo-inositol cotransporter in MDCK cells. Am. J. Physiol. 264, F20᎐F23. Zhou, C., Cammarata, P.R., 1997. Cloning the bovine Naqrmyo-inositol cotransporter gene and characterization of an osmotic responsive promoter. Exp. Eye Res. 65 Ž3., 349᎐363.