The basal Murashige and Skoog (MS) medium was used for multiplication and selection of lines tolerant to PEG [11] . Cell lines were incrementally adapted to osmotic stress by using polyethylene glycol (up to 20% PEG) following Shah et al. (2012) [12] .

The media was sterilized by autoclaving (at pressure 15 p.s.i and temperature 121˚C) for 20 minutes. Other equipments (scalpel, forceps, Pasteur pipettes, spatulas and glassware) were wrapped in aluminum foil and sterilized by autoclaving.

The cell line from the rice was adapted to 20% PEG by a multistep adaptation procedure following Shah et al. (2012) [13] . Cell lines were subjected to an incremental increase in PEG concentration. Initially calli was inoculated in suspension medium containing 5% PEG. After 2 cultures (10 days each), flasks with maximum growth were sub-cultured on medium containing 10% PEG. In similar way, stable lines were sub-cultured on medium with higher stress until a stable line was obtained on medium containing 20% PEG.

For heat and cold stress, the cell lines were subjected to 36˚C and 18˚C, respectively for two weeks.

Proline was extracted by the method of Singh et al. (1973) [14] and determined by the method of Bates et al. (1973) [15] .

Total cellular fatty acid content from rice cell lines was extracted by following the method of Bligh and Dyer (1959) [16] . The fatty acid analysis was performed using gas chromatography-mass spectrometry (GC-MS).

The fatty acid methyl esters (FAMEs) were determined by gas chromatography on a GCMS-QP2010 plus, equipped with an electron ionization (EI) detector. Chromatography was performed by using a TRB-FFAP (free fatty acid phase) column (30 m × 0.32 m × 0.25 µM diameter). One microliter of the sample was injected with a 1:50 split ratio. The temperature of injection was 240˚C, the ion source was kept at 250˚C and the interface was kept at 240˚C. Oven temperature program was on hold for 1 min at 50˚C, followed by a 15˚C∙min⁻¹ rise to 150˚C, 5 min at 175˚C, and a final 5 min at 220˚C prior to the next injection. Helium was used as the carrier gas. Compounds were identified by comparison of retention time with external standard and further confirmed by similarity search using mass spectral libraries. NIST (National Institute of Standards and Technology) database was used as external standard.

The growth of calli was measured following Shah et al. (1990) [17] . A mesh scoop was used to add a similar amount of calli to flasks containing 80 ml suspension culture medium. The average amount of calli in one scoop was measured and the initial fresh weight was obtained. The cultures were incubated in dark. After 2 weeks the media was discarded and the total calli was weighted to obtain final fresh weight.

Our results show that control cell lines showed a slow growth rate at lower and higher temperatures tested, in contrast, adapted cell lines maintained the same growth rate at the tested temperatures compared to permissible temperature (28˚C) (Figure 1). This indicates that cell line adapted to osmotic stress show the phenomenon of co-adaptation to heat and cold stress.

An increase in the level of proline was observed in PEG adapted line on optimal temperature and the levels were maintained during temperature stress conditions. However, the proline levels in un-adapted line seem to be unaffected with low and high temperature. This indicates that a higher accumulation of proline in adapted lines is a response to osmotic stress rather than temperature stress. Higher proline levels seem to be associated with higher growth rates of adapted cell lines whereas low proline associated with poor growth rates of un-adapted cell lines (Figure 2).

These results are in agreement with the previous results of Shah et al. (2012); Wahid and Close (2007) [12] [18] ; who reported that accumulation of proline in the cell, may buffer cellular redox potential under cold, heat and other environmental stresses.

A total of fifteen fatty acids were detected in adapted and un-adapted cell lines which were classified as saturated and unsaturated fatty acids as shown in the Table S1.

Among these 10 were saturated fatty acids, all these were not detected in both lines. The common saturated fatty acids among both cell lines were Myristic acid (C14:0), Palmitic acid (C16:0), Stearic acid (C18:0) while the common unsaturated fatty acid found were Oleic acid (C18:1c) and Linoleic acid (C18:2c).

Palmitic acid was the most abundant fatty acid detected in all cell lines, followed by Oleic acid, linoleic acid, stearic acid and myristic (Table 1).

In our experiments, the un-adapted line showed higher level of saturated fatty acid as compared to the adapted cell line under optimal conditions. Under cold stress a decrease was observed in both lines while under heat stress the percent level was higher in adapted cell lines than the un-adapted cell line (Figure 3).

Furthermore, our data reveals that the level of fatty acid unsaturation increased at low temperature while at high temperature the level decreased as shown in Figure 4.

It is reported that plasma membrane having a higher proportion of unsaturated

fatty acids is usually tolerant to low temperature stress while a higher proportion of saturated fatty acids renders tolerance to higher temperature stress [9] [19] . Furthermore, the degree of lipid unsaturation is related to cold tolerance due to its effect on PM stability [20] .

Plant basic physiological response to drought stress overlaps temperature, as well as salt stress; that ultimately leads to dehydration in growing plants [5] . Thermal adaptation is related to biochemical and physiological responses of the plant which are mainly caused by changes in lipidic fluidity of membranes [21] . Plants respond to different abiotic stresses (drought, heat and salinity) by the changes in the degree of un-saturation of fatty acids [22] .

It is interesting to note that PEG adapted cell lines have shown differential response to low and high temperature. Under low temperature it accumulated a high concentration of unsaturated fatty acids while under high temperature it had a high percentage of saturated fatty acids as compared to the un-adapted cell line. Similar results have been reported by [23] and [24] that increase in temperature increase the level of saturated fatty acids in membrane lipids and enhance the heat stability of cell membrane. While with decrease in temperature the level of saturated fatty acid decreased. This suggests that the cold temperature activates fatty acid desaturation and the saturated fatty acids were converted to unsaturated fatty acids to stabilize PM under cold environment [25] . An increase in total saturated and unsaturated fatty acids under high and low temperatures respectively, have obviously affected the fluidity of the plasma membrane [26] .