Plant Physiol. (1992) 99, 1695-1698 0032-0889/92/99/1695/04/$01.

00/0

Received for publication March 20, 1992 Accepted April 22, 1992

Communication

Increased Expression of a myo-lnositol Methyl Transferase in Mesembryanthemum crystallinum Is Part of a Stress Response Distinct from Crassulacean Acid Metabolism

Induction'
Daniel M. Vernon and Hans J. Bohnert* Departments of Molecular and Cellular Biology (D.M.V., H.J.B.), and Biochemistry (H.J.B.), University of Arizona, Tucson, Arizona 85721
ABSTRACT

The facultative halophyte Mesembryanthemum crystallinum responds to osmotic stress by switching from C3 photosynthesis to Crassulacean acid metabolism (CAM). This shift to CAM involves the stress-initiated up-regulation of mRNAs encoding CAM enzymes. The capability of the plants to induce a key CAM enzyme, phosphoenolpyruvate carboxylase, is influenced by plant age, and it has been suggested that adaptation to salinity in M. crystallinum may be modulated by a developmental program that controls molecular responses to stress. We have compared the effects of plant age on the expression of two salinity-induced genes: Gpdl, which encodes the photosynthesis-related enzyme glyceraldehyde 3-phosphate dehydrogenase, and ImtI, which encodes a methyl transferase involved in the biosynthesis of a putative osmoprotectant, pinitol. Imt! mRNA accumulation and the accompanying increase in pinitol in stressed Mesembryanthemum exhibit a pattern of induction distinct from that observed for CAM-related genes. We conclude that the molecular mechanisms that trigger Imt! and pinitol accumulation in response to salt stress in M. crystallinum differ in some respects from those that lead to CAM induction. There may be multiple signals or pathways that regulate inducible components of salinity tolerance in this facultative halophyte.

The common ice plant (Mesembryanthemum crystallinum) is a facultative halophyte that switches from C3 photosynthesis to CAM in response to drought or high salinity. CAM is an alternate photosynthesis pathway that serves as a water conservation mechanism (1, 9). During CAM, stomatal opening and primary carbon fixation occur at night, allowing for closure of stomata during the day and, consequently, decreased evaporative water loss. CAM induction in the ice plant has been well characterized at the biochemical level. Exposure to high salinity has been shown to trigger the accumulation of a number of mRNAs encoding CAM enzymes (5, 6, 8), resulting in increased expression and activity of these proteins (4).
1 Supported by U.S. Department of Agriculture and Arizona Agricultural Experimental Station.

One such stress-induced enzyme is PEPCase2, which catalyzes the primary carbon fixation step in CAM. Ppcl and protein expression have been used as markers to study the stress response in M. crystallinum at the molecular level. Although PEPCase expression in the ice plant is primarily environmentally regulated (11), a number of recent studies have noted that the capacity of the plants to induce PEPCase in response to stress and the rate of Ppcl induction increase with plant age (1, 3, 5, 12). Three-week-old ice plant seedlings exhibit no significant increase in PEPCase activity (5) and only a modest increase in Ppcl mRNA levels during a 5-d stress with 500 mm NaCl, whereas plants 6 or 9 weeks of age induce Ppcl to maximum levels during such a stress (3). Also, gradual increases in Ppcl transcription and PEPCase activity are observed in unstressed plants of increasing age (2, 3, 12). Such observations have led to the proposal that the stress response in M. crystallinum is modulated by a developmental program that allows maximum expression of salinity-responsive genes only after a certain plant age (approximately 5 weeks under our growth conditions) is reached. To determine whether other salt stress-responsive mRNAs exhibit expression patterns similar to that of Ppcl, we have examined the influence of plant age on the expression of two additional salinity-induced genes. One, Gpdl, encodes GAPdH (6), an enzyme that has important housekeeping functions and photosynthetic carbon reduction roles in C3 plants. GAPdH is thought to be up-regulated during salt stress in the ice plant to fulfill an increased need for the enzyme in carbon flux during the large diurnal fluctuations in organic acids and starch that are characteristic of CAM. For this reason, the increased expression of GAPdH in stressed M. crystallinum is considered 'CAM-related.' The second salinity-induced gene, Imtl, encodes a myo-inositol 0methyl transferase that catalyzes the first step in the biosynthesis of pinitol (10), a methylated cyclic sugar alcohol that is thought to serve as an intracellular osmolyte and/or os2 Abbreviations: PEPCase, phosphoenolpyruvate carboxylase; GAPdH, glyceraldehyde-3-phosphate dehydrogenase; Imtl, gene encoding myo-inositol 0-methyl transferase; Gpdl, gene encoding GAPdH; Ppcl, gene encoding PEPCase.
1695

1 696

VERNON AND BOHNERT

Plant Physiol. Vol. 99, 1992

moprotectant during osmotic adjustment in salt-stressed M. crystallinum (7). In this study, we show that the expression of Imtl, as well as the accumulation of pinitol, exhibit a pattern of induction that differs from the CAM-related increase in expression of the Gpdl gene. Although the behavior of Gpdl during development and stress reflects the trend previously observed for Ppcl, Imtl and pinitol levels are readily induced by salt stress in M. crystallinum regardless of plant age. Furthermore, there is no obvious gradual accumulation of the Imtl transcript in unstressed plants during development. We conclude that in regulation as well as function, this mRNA is part of a stress response distinct from the induction of CAM in the ice plant.

MATERIALS AND METHODS

Plant Materials
Plants were grown in 1-L pots in soil (5) and either left unstressed (controls) or irrigated daily for 5 d with 500 mm NaCl in the nutrient solution. Stress treatments were started 3, 6, or 8.5 (8 weeks plus 4 d) weeks after seeding. Leaf and axillary tissue of stressed plants and controls was harvested on the fifth d of stress (102 h), ground by mortar and pestle in liquid N2, and stored at -700C. Material from three to five plants was combined for each sample. To avoid possible variation of mRNA levels due to diurnal regulation, all harvests were carried out at the same time of day, 4 h before the end of the light period.
RNA Extraction and Analysis

we examined the expression of two salt stress-induced transcripts, Gpdl and Imtl, in 3-, 6-, and 8.5-week old M. crystallinum (Fig. 1). Plants of all three ages responded to a 5-d stress with 500 mm NaCl by significantly increasing the expression of Imtl to approximately equal levels. This result contrasted with what was observed for Gpdl. Gpdl was upregulated to a greater extent in older plants in response to stress. Poststress Gpdl transcript levels were approximately 3 times higher in stressed 6- and 8.5-week-old M. crystallinum than in stressed 3-week-old plants. An additional difference between Gpdl and Imtl expression was apparent in unstressed controls, where there was a gradual accumulation of Gpdl mRNA with age. Gpdl levels were at least twofold higher in 8.5-week-old unstressed plants than in 3-week-old unstressed plants. No reproducible significant increase was observed for Imtl. The pattern of expression described here for Gpdl resembles that described previously for Ppcl (3). The expression before and after salt stress of both CAM-related transcripts seems to be influenced similarly by plant age. Imtl exhibited a different pattern of accumulation; it was dramatically induced to maximum levels in young and old plants alike without increasing significantly with age in unstressed plants. The different effects of salt stress and development on the expression of Imtl and the CAM-related mRNAs Gpdl and Ppcl are

Imtl

Gpd 1
4

MM

Total RNA was extracted from leaf/axillary tissue as previously described (5), and dilution series were slot blotted and probed according to Vernon et al. (11). Blots were probed with random-primed full-length Imtl cDNA (10) or a 650base pair partial Gpdl cDNA (McUA1) (6).

3 wk

-mm0

Carbohydrate Extraction and Analysis Carbohydrates were extracted by grinding 50 mg of leaf/ axillary tissue in 1.0 mL of methanol/chloroform/water (12:5:3) in a glass homogenizer. After grinding, 1 mL of water was added and phases were separated by centrifugation. The aqueous phase was desalted on a column of AG50WX4 (BioRad) and Amberlite IRA-68 in the OH- form (Sigma). Samples were dried, dissolved in deionized water, and filtered through a nylon Acrodisc 13 (Gelman Sciences, Ann Arbor, MI). Equal amounts of extracts were resolved by HPLC on a 300 X 7.8 mm HPX87C calcium-form ligand exchange column (BioRad) as described (10). Three replicate runs of each extract were carried out. Pinitol levels were quantified by comparing HPLC peak areas obtained with a Spectrophysics SP4290 integrator (Spectrophysics Analytical, San Jose, CA) to peak areas of pinitol standards of known
concentration.

am_

6 wk

mm

8.5 wk
-

_ -

RESULTS AND DISCUSSION Because it has been proposed that the stress response in M. crystallinum is controlled by a developmental program,

Figure 1. Levels of Imtl and Gpdl mRNA in stressed and unstressed M. crystallinum of different ages. Plants were grown in soil and either left unstressed (-) or stressed with 500 mm NaCI in the nutrient solution (+). Plant ages at the beginning of stress treatments are shown. All plants were harvested at the end of the 5-d stress period. tRNA was isolated from leaf and axillary shoot material pooled from three plants for each sample. Twofold serial dilutions of RNA (5-0.625 Mg) were hybridized to 32P-labeled Gpdl or ImtI cDNA probes. Autoradiograms of blots are shown.

DISTINCT STRESS RESPONSES IN M. CRYSTALLINUM

A

1697

Ppc I
Max

Gpdl
-Max-

Imta

Table I. Pinitol Levels in Stressed and Unstressed M. crystallinum of

Various Ages

CL
3 6
3

L3
6 85

Plant Age (weeks)

8 5 3 6 Plant Age (weeks)

Figure 2. The effects of plant age and salt stress on the relative expression of various salinity-responsive mRNAs. The behavior of the photosynthesis-related mRNAs Ppcl and Gpdl (A) with respect to development and response to stress is compared to that of lmtl, a non-CAM gene (B). Transcript levels (relative to maximum expression for each) from unstressed plants are E. Levels after a 5-d stress with 500 mm NaCI are Ages at start of stress are noted. Ppcl expression information is from Cushman et al. (3).
U.

Carbohydrates were extracted from leaf and axillary tissue and resolved by HPLC as described in "Materials and Methods." Material from three to five plants was pooled for each sample. Values shown are the mean of three HPLC runs of each sample. SD were calculated from variation between HPLC runs. Pinitol Content Plant Age at Start of Stress Stressed Unstressed umol/g fresh wt weeks 3.0 0.2 NDa 3 2.40.1 6 0.70 2.80.1 8.5 0.60.1 a Not detected; detection limit: -25 pmol/injection.

summarized in graphic form in Figure 2. The relatively high level of Gpdl in unstressed controls is not unexpected (6) since GAPdH carries out glycolytic housekeeping functions and photosynthetic carbon reduction roles in unstressed tissue. The protein encoded by Imtl catalyzes the initial step in the stress-induced biosynthesis of the methylated cyclic sugar alcohol pinitol (10). To verify that the increased expression of Imtl mRNA results in the expected physiological response (the appearance of pinitol) in young as well as old stressed plants, carbohydrates were extracted from leaf/axillary tissue of plants of designated ages and subjected to HPLC analysis. Values for pinitol detected in plant extracts are shown in Table I. Pinitol did indeed increase as much in 3-week-old M. crystallinum as it did in older plants, reflecting the induction of the Imtl transcript seen in Figure 1. Some pinitol was present in older control plants. Low levels of pinitol have been observed previously in unstressed leaf tissue from mature plants (7). This accumulation is likely due to the fact that pinitol, being relatively metabolically inert, has a slow turnover time; the pinitol produced in small amounts by unstressed plants, therefore, may accumulate over time (7, 10, E.J. DeRocher, P. Adams, D.M. Vernon, unpublished observations). In any case, it is clear from both the RNA data (Fig. 1) and the large increase of pinitol in stressed, young seedlings that this aspect of the M. crystallinum stress response is not dependent on plant age as is the expression of CAM-related genes.
It is not known what causes the observed influence of age CAM induction and the gradual increase of CAM gene expression seen in unstressed plants. Cushman et al. (3) suggest that the response to salt stress in M. crystallinum is dependent on a 'developmental program.' Others hypothesize that the ice plant's stress response is triggered by decreases in turgor. They attribute differences in PEPCase expression between young and old plants to the cumulative effects of diurnal changes in water status that become more dramatic as plants age (12). Our data are not easily accommodated by either model. The putative developmental program that modulates the CAM response at the molecular
on

level does not suppress the induction of Imtl or the accumulation of pinitol in young, stressed M. crystallinum, nor does it significantly affect the levels of Imtl mRNA in unstressed plants. Likewise, the inducibility of Imtl in older plants and the background levels of Imtl in plants of increasing age do not appear to be augmented by diurnal transient reductions in water status. The regulation of the CAM response (as gauged by Ppcl and Gpdl mRNA levels), therefore, differs in some respects from the regulation of other aspects of stress adaptation (such as osmoregulation and/or osmoprotection, as gauged by Imtl mRNA expression). It appears as though there is not a single signal or pathway that coordinately regulates molecular responses to osmotic stress in the ice plant. Rather, the response to the environment in M. crystallinum is likely to be complex, and the regulation of stressresponsive genes may involve several signals or the interplay of multiple control mechanisms.
ACKNOWLEDGMENTS
We thank Dr. John C. Thomas and Dr. John C. Cushman for helpful discussions and critical reading of the manuscript. We are very grateful to Pat Adams for carrying out HPLC analysis. LITERATURE CITED
1. Bohnert HJ, Ostrem JA, Cushman JC, De Rocher EJ, Michalowski CB, Rickers J, Meyer G, Vernon DM, Krueger M, Vasquez-Moreno L, Velton J, Hoefner R, Schmitt JM (1988) Mesembryanthemum crystallinum: a higher plant model for the study of environmentally induced changes in gene expression. Plant Mol Biol Reporter 6: 10-28 2. Chu C, Dai Z, Ku MSB, Edwards GE (1990) Induction of crassulacean acid metabolism in the facultative halophyte Mesembryanthemum crystallinum by absiscic acid. Plant Physiol 93: 1253-1260 3. Cushman JC, Michalowski C, Bohnert HJ (1990) Developmental control of crassulacean acid metabolism inducibility by salt stress in the common ice plant. Plant Physiol 94: 1137-1142 4. Holtum JAM, Winter K (1982) Activity of enzymes of carbon metabolism during the induction of crassulacean acid metabolism in Mesembryanthemum crystallinum L. Planta 155: 8-16 5. Ostrem JA, Olsen SW, Schmitt JM, Bohnert HJ (1987) Salt stress increases the level of translatable mRNA for phosphoen-

1698

VERNON AND BOHNERT

Plant Physiol. Vol. 99, 1992

olpyruvate carboxylase in Mesembryanthemum crystallinum. Plant Physiol 84: 1270-1275 6. Ostrem JA, Vernon DM, Bohnert HJ (1990) Increased expression of a gene coding for NAD-glyceraldehyde-3-phosphate dehydrogenase during the transition from C3 photosynthesis to crassulacean acid metabolism in Mesembryanthemum crystallinum. J Biol Chem 256: 3497-3502 7. Paul MJ, Cockburn W (1989) Pinitol, a compatible solute in Mesembryanthemum crystallinum L?. J Exp Bot 40: 1093-1098 8. Schmitt JM, Michalowski CM, Bohnert HJ (1988) Gene expression during CAM induction under salt stress in mesembryanthemum: cDNA library and increased levels of mRNA for phosphoenol pyruvate carboxylase and pyruvate orthophosphate dikinase. Photosynth Res 17: 159-171

9. Ting IP (1985) Crassulacean acid metabolism. Annu Rev Plant Physiol 36: 595-622 10. Vernon DM, Bohnert HJ (1992) A novel methyl transferase induced by osmotic stress in the facultative halophyte Mesembryanthemum crystallinum. EMBO J. 11: 2077-2085 11. Vernon DM, Ostrem JA, Schmitt JM, Bohnert HJ (1988) PEPCase transcript levels in Mesembryanthemum crystallinum decline rapidly upon relief from salt stress. Plant Physiol 86: 1002-1004 12. Winter K, Gademann R (1991) Daily changes in CO2 and water vapor exchange, chlorophyll fluorescence, and leaf water relations in the halophyte Mesembryanthemum crystallinum during the induction of crassulacean acid metabolism in response to high salinity. Plant Physiol 95: 768-776