How do cells cope with drying or exposure to excessive salt concentrations? In plants, adaptation to desiccation and adaptation to salt stress can involve the same mechanism in which increased transcription of betaine aldehyde dehydrogenase leads to synthesis of the compatible organic osmolyte, glycine betaine (Kotchoni & Bartels, 2003). In this issue of The Journal of Physiology, Huang & Tunnacliffe (2004) have now examined the possibility that a common mechanism exists in mammalian cells by comparing their responses to water loss when this is caused by hypertonicity versus desiccation. Little is known about the adaptive response to desiccation in mammalian cells. The present studies not only provide lessons about the signalling systems involved, but also raise the possibility that mammalian cells could be dried in a viable form, which could have significant applications in both medicine and basic research.
In mammalian renal cells, responses to hypertonicity induced by high concentrations of NaCl or non-ionic impermeant solutes such as raffinose involve adaptive accumulation of several compatible organic osmolytes, including sorbitol, betaine and inositol (Burg et al. 1997). Accumulation of these solutes occurs as a result of increased transcription of aldose reductase (AR, which catalyses the conversion of glucose to sorbitol), the betaine–γ-aminobutyric acid transporter (BGT1) and the sodium–myo-inositol co-transporter (SMIT), respectively, all of which are up-regulated by a transcription factor called TonEBP/OREBP (for tonicity-responsive enhancer/osmotic response element-binding protein; reviewed by Irarrazabal et al. 2004).
It is not entirely clear what initiates hypertonicity-induced activation of TonEBP/OREBP. Hypertonicity shrinks cells and increases intracellular ionic strength, and previous studies have attempted to distinguish the relative importance of these two stimuli. Huang & Tunnacliffe start with the assumption that desiccation also should have both effects, and therefore should lead to adaptive accumulation of compatible osmolytes. Interestingly, they find that hypertonicity (caused by high concentrations of sodium salts or mannitol, but not KCl or sorbitol) increases AR, BGT1 and SMIT mRNAs in 293T cells whilst desiccation does not. The negative result during desiccation renews the question of whether hypertonicity activates the adaptive response by cell shrinkage or increased intracellular ionic strength.
Previous attempts to resolve this question did not produce a straightforward answer. Thus, in response to high NaCl or raffinose, with or without ouabain, AR activity correlates with cell potassium concentration and even more strongly with the sum of cell sodium plus potassium concentration or ionic strength, but not with cell sodium concentration or water content alone. The conclusion was that intracellular ionic strength, rather than cell volume, is the stimulus (Uchida et al. 1989). Later experiments, in which mRNAs for TonEBP/OREBP target genes were measured as an end point, revealed a correlation with ionic strength but also showed that simply raising intracellular ionic strength with ouabain or high KCl is not sufficient to elevate the mRNAs (Neuhofer et al. 2002). The authors suggested that cell swelling might have occurred under those conditions, which could potentially counteract the effect of increased ionic strength. Finally, a sufficiently slow rise in osmolality generally results in isovolumetric regulation in which intracellular ionic strength increases without cell shrinkage. Under those conditions AR and BGT1 mRNAs increase greatly, emphasizing the importance of intracellular ionic strength (Cai et al. 2004).
Based on their results, Huang & Tunnacliffe conclude that the stimulus for osmotic regulation of these genes must involve factors other than intracellular ionic strength and cell shrinkage. Why else would the effects of hypertonicity and desiccation differ so profoundly? Other factors are also needed to explain why hypertonicity induced by sodium salts has effects that are different from those produced by elevation of KCl or sorbitol. The authors suggest that sodium concentration, per se, may be most important. However, this is unlikely to be a general phenomenon because the non-ionic, organic solute raffinose is as effective as NaCl in up-regulating these genes in other cell types (Burg et al. 1997). We agree that the osmotic regulation of these genes must involve factors other than intracellular ionic strength and cell shrinkage, but disagree that sodium is key. What might those ‘other factors’ be?
We suggest that signals arising from other effects of hypertonicity affect the activity of TonEBP/OREBP, adding to or negating the influence of ionic strength and cell volume. Two candidates come from recent observations that hypertonicity causes DNA damage and increases reactive oxygen species (ROS), both in cell culture and in vivo (reviewed in Zhang et al. 2004). Multiple agents, including high NaCl, high urea, UV and ionizing radiation, generate DNA damage and activate ATM, a DNA damage response protein. Among these agents, only high NaCl activates TonEBP/OREBP. The increased activity of ATM is necessary for full osmotic activation of TonEBP/OREBP (Irarrazabal et al. 2004). ROS, which are increased by high NaCl, contribute to the regulation of transcription factors other than TonEBP/OREBP, and we suggest that in the context of hypertonicity, ROS may contribute to the osmotic regulation of TonEBP/OREBP. Along the same lines, p38 mitogen-activated protein kinase (MAPK) is known to mediate many stress signalling pathways, including activation of TonEBP/OREBP by hypertonicity. Huang & Tunnacliffe now find that desiccation also activates p38, but it does not activate TonEBP/OREBP. Evidently, activation of TonEBP/OREBP by p38 requires additional accompanying signals that desiccation does not provide. Thus, hypertonicity has multiple effects and triggers multiple signalling pathways. Those signals need to act in concert rather than individually to cause full activation of TonEBP/OREBP.
By examining the previously little explored area of signals arising from desiccation in mammalian cells and comparing the effects to those of hypertonicity, Huang & Tunnacliffe have opened an exciting new area of research. Their studies shed new light on the under-studied problem of desiccation in mammalian cells, and also the extensively studied (but still incompletely understood) area of signals arising from hypertonicity.