OUP user menu

Freeze tolerance of the yeast Torulaspora delbrueckii: cellular and biochemical basis

Cecília Alves-Araújo , Maria Judite Almeida , Maria João Sousa , Cecília Leão
DOI: http://dx.doi.org/10.1016/j.femsle.2004.09.008 7-14 First published online: 1 November 2004


The freeze stress responses to prolonged storage at −20 °C in Torulaspora delbrueckii PYCC5323 were investigated. In this yeast, no loss of cell viability was observed for at least 120 days during freezing at −20 °C, whereas a loss of 80% was observed in a commercial baker's yeast after 15 days. In the former strain, freeze resistance was dependent on an adaptation process. The primary cell target of freeze stress was the plasma membrane, preservation of its integrity being related with a lower increase of lipid peroxidation and with a higher resistance to H2O2, but not with the intracellular trehalose concentration.

  • Torulaspora delbrueckii
  • Freeze tolerance
  • Membrane integrity
  • Baker's yeast
  • Lipid peroxidation

1 Introduction

Frozen-dough technology is well established in the baking industry, making it easier for bakers to supply oven-fresh bakery products to consumers and improving labour conditions. However, storage of frozen bread dough may lead to the loss of baker's yeast cell viability as well as of its baking capacity, and consequently to economic losses. Thus, bread-making industry holds a high demand for yeast strains with improved freeze resistance [1,2]. Most research in this field has focused on strains of Saccharomyces cerevisiae, the species currently used as baker's yeast. In this species, tolerance to freezing has generally been correlated with the intracellular trehalose concentration, but no direct correlation has been found above a certain threshold value [3,4,5,6]. Prior to the frozen storage, once the yeast cells are mixed with flour the fermentation takes place and a rapid loss of stress resistance occurs [7]. This has also been associated with the degradation of intracellular trehalose [8]. However, it has been shown that retention of high trehalose levels in fermenting cells does not prevent the loss of fermentation capacity during freezing, and that other factors – not yet identified – are required for the maintenance of freeze stress resistance [9,10]. Accumulation of other solutes such as amino acids and glycerol, and expression of aquaporins were also reported to increase the freezing resistance [11,12,13,14,15]. In addition, oxidative damage has been considered to be a factor underlying freeze–thaw damage, since an oxidative burst has been predicted to occur during thawing [16]. In agreement with this, Park et al. [17] found that respiratory ability and functional mitochondria are necessary to confer full resistance to freeze–thaw stress resistance. It has also been shown that freeze tolerance is correlated with tolerance to H2O2, and free radicals were detected in S. cerevisiae after the freeze–thaw process [5,18].

The strain PYCC5323 of Torulaspora delbrueckii, isolated from traditional corn and rye bread dough from the North of Portugal, displays high freeze and osmotic tolerance besides presenting dough-raising capacity, growth rates and biomass yields similar to commercial baker's yeast [19,20,21]. Therefore, this yeast emerges as a powerful candidate for the bread-making industry, and the elucidation of such a peculiar behaviour reveals to be of great interest.

In this work, the freeze stress responses to prolonged storage at −20 °C in T. delbrueckii PYCC5323 were investigated. The results obtained were compared to the ones of a commercial baker's yeast strain of S. cerevisiae. The following cellular and biochemical parameters were analysed: cell viability, plasma membrane integrity, oxidative damages, intracellular trehalose content and trehalase activity.

2 Materials and methods

2.1 Microorganisms, growth and freezing conditions

The strains used were T. delbrueckii PYCC5323, isolated from home-made corn and rye bread dough, and S. cerevisiae PYCC5325, isolated from commercial compressed baker's yeast – both supplied by the Portuguese Yeast Culture Collection, Faculdade de Ciências e Tecnologia, Universidade Nova de Lisboa, Caparica, Portugal. Stock cultures were maintained on glucose–yeast extract–peptone–agar (2% (w/v) glucose, 0.5% (w/v) yeast extract, 1% (w/v) peptone, 2% (w/v) agar), at 4 °C. Yeast strains were grown on yeast extract–peptone–sucrose (YPS) medium containing 2% (w/v) sucrose, 4% (w/v) peptone, 2% (w/v) yeast extract, 0.2% (w/v) KH2PO4 and 0.1% (w/v) MgSO4· 7H2O, at 30 °C. Peptone (ref. 0118–17) and yeast extract (ref. 0127–17) were from Difco (Becton–Dickinson, Sparks, MD, USA) and sucrose from Merck (E. Merck, Darmstadt, Germany).

For freezing assays, cells were harvested at initial stationary phase (24 h of culture, 2.4–2.7 × 108 cells/ml for T. delbrueckii and 1.1–1.5 × 108 cells/ml for S. cerevisiae), washed twice with deionised water, and suspended in a quarter of the initial volume in sterile water to an A640 of 12–20. Aliquots (5 ml) of cells were transferred into 15-ml polycarbonate tubes, centrifuged and the pellet (400–500 mg of wet weight) suspended in 500 μl of the storage medium (LF medium) or as described in the results. LF medium is a liquid medium formulated to simulate the fermenting ability of yeast in bread dough, with the following composition: 1% (w/v) glucose, 1% (w/v) sucrose, 3% (w/v) maltose, 0.25% (w/v) (NH4)2SO4, 0.5% (w/v) urea, 1.6% (w/v) KH2PO4, 0.5% (w/v) Na2HPO4· 12H2O, 0.06% (w/v) MgSO4, 22.5 ppm nicotinic acid, 5.0 ppm pantothenic acid, 2.5 ppm thiamine, 1.25 ppm pyridoxine, 1.0 ppm riboflavin and 0.5 ppm folic acid [22]. The samples were frozen at −20 °C for different time periods (cooling rate approximately 3 °C/min) [5], and then thawed at 30 °C for 2 min. For fast freezing, 15-ml polycarbonate tubes containing the cell suspensions were directly immersed in liquid nitrogen (cooling rate approximately 200 °C/min) [5]. For pre-fermentation treatments, cells were subjected to a fermentation period before freezing as follows: the pellets in the 15-ml polycarbonate tubes were suspended in LF medium to a final OD640 nm of 0.3–0.5, and incubated for 120 min at 30 °C. After this, the suspension was centrifuged and the pellet was suspended in LF medium and frozen as described above.

2.2 Extraction and assay of trehalose

Cells were harvested by centrifugation, washed twice with cold deionised water and sampled for dry weight contents. Trehalose was extracted from cold cell pellets with 5% (w/v) trichloroacetic acid (Merck, Darmstadt, Germany) for 45 min with occasional shaking. Cells were then centrifuged at 735g, for 10 min. Extraction was repeated once more, and supernatants from the two extractions were combined and used for the determination of trehalose by high-performance liquid chromatography. The apparatus used was a Gilson chromatograph (132-RI Detector), equipped with a carbohydrate H+ column (SS-100, H+, Hypersil) which was maintained at 30 °C. A solution of H2SO4 (0.0025 M) was used as the mobile phase at a flow rate of 0.45 ml/min. The relative values (%) of the intracellular trehalose concentration after different periods of freezing were calculated by dividing the values of intracellular trehalose concentration obtained for the frozen samples by those obtained for the unfrozen samples.

2.3 Trehalase activity and protein assay

Pellets containing 75–100 mg wet weight of washed cells were suspended in 1 ml ice-cold 50 μM MES (4-morpholineethanesulfonic acid) (Boehringer–Mannheim, GmbH-Germany)/KOH buffer, pH 7, containing 50 μM CaCl2. Cells were broken by vortexing with 500 μl of glass beads (0.5 mm diameter), for four periods of 1 min, with 1-min intervals on ice between them. The crude enzyme extract was centrifuged for 3 min at 13 200g, at 4 °C. The supernatant was dialysed overnight at 4 °C against 10 mM Mes/KOH buffer, pH 7, containing 50 μM CaCl2. Trehalase was assayed as described previously [23]. The glucose liberated was determined by glucose oxidase/peroxidase method (Glucose GOD – Perid, Boehringer–Mannheim, GmbH-Germany). Protein determination was carried out according to Lowry et al. [24]. Specific activity of trehalase was expressed as units (U, nmol glucose released per min) per mg protein.

2.4 Measurement of cell viability and membrane integrity

The viability of yeast cells was determined by counting colony-forming units (CFU). For this assay, yeast cell suspensions were washed twice with deionised water and, after convenient dilution, spread on YPDA medium plates. The plates were incubated for 48 h at 30 °C before counting. The relative values (%) of viable cells after different periods of freezing were calculated by dividing the values of CFU counts obtained for the frozen samples by those obtained for the unfrozen samples.

Membrane integrity was analysed by flow cytometry using the membrane exclusion dye, propidium iodide (PI). In these assays, cells with preserved membrane integrity are not permeated by propidium iodide (PI cells), while those that lost their membrane integrity do incorporate the fluorochrome (PI+ cells) [25]. Cell suspensions (about 107 cells/ml) were incubated for 10 min in the dark with a 20 μg/ml PI solution (ratio of 2:1, respectively) and injected on a Partec PAS III flow cytometer, equipped with an argon-ion laser emitting a 15 mW beam at 488 nm. From each sample, 2 × 104 cells were analysed. Control suspensions of membrane-disrupted cells were prepared by boiling cell suspensions. The relative values (%) of PI cells after different periods of freezing were calculated by dividing the values of PI cells obtained for the frozen samples by those obtained for the unfrozen samples.

2.5 Oxidative stress evaluation

For evaluating the effect of pre-treatment with the radical scavenger N-tert-butyl-α-phenylnitrone (PBN) (Aldrich Chem. Co. Milwaukee, WI), cells were grown for 24 h to initial stationary phase in YPS medium at 30 °C. PBN was added to the culture at the final concentrations of 0.5, 5.0 and 15 mM, and cells were cultured for 30 min before being harvested. PBN was removed by washing the cells twice with deionised water and cells were frozen in LF medium as described above. In assays where a pre-fermentation was performed, PBN was added to the fermenting cell suspensions 30 min before the end of the fermentation period.

For treatment with oxidising agents, cells were grown to mid-exponential (OD640 nm 0.5–1.0) and initial stationary phase (24 h of culture), harvested and washed twice with deionised water. Subsequently, these cells were suspended in water in order to achieve a concentration of OD640 nm 0.6–0.7, and 0.1 ml of this suspension was spread on solid YPDA plates (50 mm diameter). A paper disc (6 mm diameter – BBL, 231039, Becton–Dickinson, Sparks, MD, USA) containing 10 μl of one oxidising agent, was laid on each inoculated plate. The oxidants used were: menadione at concentrations of 0.05, 0.10 and 1.0 mM; diamide at concentrations of 0.3 and 3.0 M; hydrogen peroxide at 1.1 and 11 M, all purchased from Sigma Chemical Co. (St. Louis, MO). The plates were incubated at 26 °C during 2 days, after which the diameters of the inhibition halos around the paper discs were measured. The value obtained was the average of two perpendicular diameters, excluding the disc diameter [26].

2.6 Thiobarbituric acid (TBA) reaction for lipid peroxide analysis

For quantification of TBA-reactive substances (TBARS), pellets containing 350–400 mg (wet weight) of cells were washed with ice-cold deionised water and suspended in 0.75 ml ice-cold sodium phosphate (E. Merck, Darmstadt, Germany) buffer, pH 7.2. Cells were broken by vortexing with 500 μl of glass beads (0.5 mm diameter) for six periods of 1 min, with 1 min intervals on ice between them. TBARS were then determined according to Buege and Aust [27]. The TBARS concentration was expressed as μmol malondialdehyde per mg protein.

2.7 Reproducibility of the results

All the experiments were repeated at least three times, and the data reported are mean values and SD. When statistical analyses were performed, the significance was tested by analysis of variance (Anova, Microsoft Excel 2000).

3 Results

3.1 Cell viability and membrane integrity along freeze storage

To characterise the freeze resistance of T. delbrueckii PYCC5323, we first studied cell viability, assessed as CFU, during freezing at −20 °C for up to 120 days (Fig. 1(a)). For a comparative analysis, a commercial baker's yeast, S. cerevisiae PYCC5325, was used as a reference strain (Fig. 1(b)). In T. delbrueckii, no loss of cell viability was observed for the entire storage period, whereas a loss of 80% was observed in S. cerevisiae after 15 days. In both species, loss off viability was more rapid when cells were stored in freezing medium without sugars, even when glycerol was added to keep the osmotic pressure. These results are consistent with previous reports, showing that the presence of the disaccharide trehalose in the extracellular medium has a protective effect in cell viability during freezing [28].

Figure 1

Relative values (%) of colony forming units (CFU) in stationary growth phase cells of T. delbrueckii (a) and S. cerevisiae (b) during frozen storage at −20 °C, for 120 and 30 days, respectively. Cells were grown in YPS medium during 24 h at 30 °C, harvested, washed with water and suspended in different storage media: (◻) LF medium; (●) LF medium + 0.01% cycloheximide; (○) LF medium without sugars + 0.16 M glycerol; (▲) Sugars solution (1% glucose + 1% sucrose + 3% maltose).

To elucidate whether the loss of membrane integrity along freezing was directly conditioning the cell viability, membrane damage was monitored by flow cytometry using PI cell staining. The decrease of cells with preserved plasma membrane (PI cells) was much less pronounced in T. delbrueckii (Fig. 2(a)) than in S. cerevisiae. As shown in Fig. 2(b), there was a direct correlation between the percentage of PI cells and the percentage of CFU counts for both T. delbruekii and S. cerevisiae yeast strains. This was evident either in cells subjected to a short fermentation period (120 min) or in cells not subjected to fermentation before freezing. The results show that, independent of the physiological state of the cells, membrane integrity is directly conditioning the cell viability, expressed as CFU counts, and therefore the plasma membrane seems to be one of the first freezing targets.

Figure 2

Relative values (%) of negative propidium iodide cells (PI cells) of T. delbrueckii (close symbols) and S. cerevisiae (open symbols) during storage at −20 °C (a) and correlation between relative values (%) of PI cells and colony forming units (CFU) (b). Cells were grown in YPS medium during 24 h at 30 °C (initial stationary growth phase), harvested, washed with water, suspended in LF medium and frozen directly (▲, △) or after a pre-fermentation period of 120 min (●, ○), as described in Section 2. In (a), for some results error bars are within the data point labels.

3.2 Cell oxidative stress responses during the freeze–thaw process

The results described above indicate that T. delbrueckii displays a higher freeze resistance when compared with S. cerevisiae, which is mainly due to plasma membrane integrity. To examine whether this capacity to preserve plasma membrane integrity was correlated with oxidative stress resistance, we directly assessed the oxidative damage in the membranes of both T. delbrueckii and S. cerevisiae during frozen storage by measuring TBARS. The TBARS test quantifies lipid peroxides in the TBA derivatised form. As shown in Table 1 for both yeast strains, frozen cells presented higher amounts of products of lipid peroxidation than control cells (cells before freezing). However, the percentage increase in the TBARS levels was significantly enhanced (P < 0.05) much earlier in S. cerevisiae (approximately 61% for the fifth day), which was associated to its rapid decrease in cell membrane integrity. As shown in the previous section, the presence of glycerol did not protect cells of T. delbrueckii during freezing. Therefore, we tested T. delbrueckii under these conditions to evaluate the increase in the percentage of TBARS. The results showed that the values of percentage increase of TBARS in T. delbrueckii frozen in a medium with glycerol, for 60 and 84 days, were similar to those observed for S. cerevisiae (respectively 64% and 83%). Moreover, samples of T. delbrueckii frozen under conditions where no loss of cell viability was observed, presented a much lower percentage increase in TBARS production (Fig. 1, Table 1).

View this table:
Table 1

TBA-reactive substances (TBARS) of yeast extracts expressed as μmol malondialdehyde (MDA) per mg proteina

Freeze timeWithout freezing5 days30 days60 days84 days
Freeze mediaLFLFSugars sol.bLF + Glyc.cLFSugars sol.LF + Glyc.LFSugars sol.LF + Glyc.
T. delbrueckii PYCC532325.5 ± 1.230.1 ± 4.329.7 ± 1.525.7 ± 0.732.7 ± 0.733.3 ± 4.633.2 ± 0.441.9 ± 0.536.4 ± 0.832.3 ± 1.646.6 ± 0.9
S. cerevisiae PYCC532519.3 ± 3.531.1 ± 1.6d
  • a Cells were grown on YPS, during 24 h, at 30 °C and prepared as described in Section 2.

  • b Sugars solution (1% glucose + 1% sucrose + 3% maltose).

  • c LF medium without sugars + 0.16 M glycerol.

  • d Not determined since cell viability was already very low (see Fig. 1). Values are means from three independent experiments ± SD.

The role of oxidative stress during freezing was examined by using the oxygen radical scavenger PBN. In S. cerevisiae PYCC5325, pre-incubation of cells with PBN (1 or 5 mM) resulted in higher levels of membrane integrity preservation for the first days of frozen storage, but no protective effect by PBN was observed in cells frozen after a period of pre-fermentation (results not shown). This result might reflect the different physiological status of the cells and it is conceivable that oxidative stress is being carried over by other factors [17]. Next, the oxidative stress responses of both strains to hydrogen peroxide (H2O2, toxicity mainly due to hydroxyl radicals), menadione (a superoxide-generating agent) and diamide (thiol-oxidizing drug) were evaluated. The results obtained in a disc diffusion assay are shown in Fig. 3. T. delbrueckii was more sensitive to menadione (P < 0.001) and diamide (P < 0.05), while more tolerant to H2O2 (P < 0.01) when compared with S. cerevisiae. Therefore, like it was previously reported for S. cerevisiae, in T. delbrueckii freeze resistance is also correlated with H2O2 resistance [5].

Figure 3

Effect of oxidising agents on the radial growth of T. delbrueckii (black bars) and S. cerevisiae (open bars). Cells were grown in YPS medium at 30 °C, to mid-exponential growth phase (b) or to initial stationary growth phase (a), and 0.1 ml of a suspension (OD640 nm 0.6–0.7) of these cells was spread on YPDA plates. A paper disc, 6 mm diameter, containing 10 μl of one oxidising agent was laid on each inoculated plate. The plates were incubated during 2 days, after which the diameters of the inhibition halos were measured. The values shown are those obtained for the highest concentration of oxidant tested.

3.3 Intracellular trehalose content during the storage period

The results described above pointed to the possibility that oxidative stress was not the only condition influencing the cell's freeze resistance, mainly in cells frozen after a short period of pre-fermentation. Intracellular trehalose accumulation has been described to protect cells from oxygen radicals and also to be involved in the stabilisation of the plasma membrane structure during freezing [29,30,31]. Hence, it was also investigated whether trehalose was involved in the higher capacity of T. delbrueckii to maintain plasma membrane integrity compared to S. cerevisiae. Immediately before freezing, the values of the intracellular trehalose content were high and similar in both species: 109.34 ± 8.72 and 112.79 ± 20.97 mg/g dry wt for S. cerevisiae and T. delbrueckii, respectively. During the entire freezing storage period, these values were kept high and constant in T. delbrueckii but decreased quickly in S. cerevisiae. For both species, the relative values of the intracellular trehalose content during freezing closely followed the percentage of PI cells (Fig. 4). In S. cerevisiae, the decrease in the intracellular trehalose content was accompanied by an increase of the extracellular trehalose content and of the medium OD260 nm (used as a measure of leakage of cell contents) (Fig. 4(b)). At the end of the assay, the amount of trehalose found in the media was about the same as the one lost from the cells, indicating that the total trehalose amount (intracellular + extracellular) remained unchanged. Thus, cell leakage appears to be responsible for the decrease in trehalose content of the cells, which is consistent with the observed loss of plasma membrane integrity. In addition, a decrease in the intracellular trehalose content was observed in cells that were subjected to a short fermentation period before freezing (cells more sensitive to freeze stress), which was similar for both strains (from 109.34 ± 8.72 to 9.79 ± 0.93 mg/g dry wt for S. cerevisiae and from 112.79 ± 20.97 to 17.16 ± 6.24 mg/g dry wt for T. delbrueckii, during a 120 min fermentation period).

Figure 4

Cell adaptation during the freeze period: cells of T. delbrueckii (a) and S. cerevisiae (b) were grown in YPS medium during 24 h at 30 °C, harvested, washed with water, suspended in LF media and frozen at −20 °C. During freezing samples were taken and assessed for: (○) relative values (%) of negative propidium iodide cells (PI cells); (▲) relative values (%) of intracellular trehalose concentration (Tin); (◻) extracellular trehalose concentration (Tout, g/l); (●) medium optical density at 260 nm, used as a measure of leakage of cell contents (OD260 nm). The relative values (%) were estimated as described in Section 2.

These results seem to indicate that a similar activation pattern of trehalase(s) by glucose was present in both yeast strains. To address this point, trehalase activities were monitored in cell free extracts prepared from cells in the absence of glucose (stationary phase cells) as well as after a glucose pulse. A similar behaviour was presented by both strains, with a two to threefold increase in activity almost immediately after glucose addition (from 15.0 ± 0.4 to 32.1 ± 1.3 mU/mg protein for S. cerevisiae and from 11.1 ± 0.2 to 31.0 ± 0.7 mU/mg protein forT. delbrueckii, before and after glucose addition, respectively). The results are in agreement with the mobilisation of trehalose observed during the pre-fermentation period. Moreover, they are consistent with the conclusion that the higher freeze resistance displayed by T. delbrueckii can not be attributed to higher intracellular trehalose contents.

3.4 Cell adaptation during the freeze period

We evaluated the ability of T. delbrueckii and S. cerevisiae to adapt to freezing by inhibiting protein synthesis with cycloheximide. The results obtained when cells were frozen at −20 °C showed that for T. delbrueckii, in contrast to S. cerevisiae, the presence of cycloheximide in the freezing medium increased the loss of cell viability throughout frozen storage (Fig. 1). To assess if cycloheximide could have a toxic effect in T. delbrueckii cells, being responsible for the observed decrease in cell viability, a cell suspension of this yeast was incubated in LF medium with cycloheximide, and the number of CFU counts was determined without previous freezing the suspension. No differences were found between the number of CFU counts estimated for this cell suspension and for the control without cycloheximide. These results indicated that at the concentration tested, cycloheximide was not having a toxic effect. In addition, when cells of T. delbrueckii were frozen at a much faster rate, in liquid nitrogen (−196 °C), and subsequently stored at −80 °C, loss of cell viability rapidly increased and the CFU counts decreased to approximately 50% after one day of frozen storage. Together, these results suggest that cell viability of T. delbrueckii is dependent on de novo protein synthesis, and therefore that yeast cells can adapt in a slow freezing process, most probably during the initial cooling period. In accordance with these findings, the capacity of T. delbrueckii cells to adapt might be essential for the yeast's high freeze tolerance phenotype.

4 Discussion

Our findings clearly indicate that the strain T. delbrueckii PYCC5323 exhibits a high freeze tolerance when compared to the baker's yeast S. cerevisiae, which reinforces previous reports claiming its useful exploitation in baking industry. Evidence was presented that the primary cell target of freeze stress is the plasma membrane and that the capacity to preserve membrane integrity displayed by T. delbrueckii PYCC5323 is correlated with a higher resistance to lipid oxidative damage. Hydroxyl radicals appear to be the agents responsible for cell membrane damage in freeze stress in this yeast. This is the opposite of a previous report on S. cerevisiae, where superoxide radicals were considered to be the agents responsible for cell damage under these stress conditions [18].To ascertain the key role of these radicals during freezing we are developing a catalase null mutant in T. delbrueckii PYCC5323. Loss of cell viability seems to be correlated with the percentage increase in the TBARS levels and not with their absolute values, suggesting that a higher content in unsaturated fatty acids could allow the cell to cope with higher absolute levels of lipid oxidation without compromising membrane integrity. This hypothesis would also agree with previous work, showing that freeze-tolerant yeast strains have larger amounts of unsaturated fatty acids when compared to freeze-sensitive strains [32]. Previous studies have pointed out a lack of capacity in S. cerevisiae to adapt to cold stress [17], although more recently, evidence for cold-induced expression changes associated with improved cryoresistance has also been provided [33,34]. According to our results, in T. delbrueckii, contrasting with S. cerevisiae, the surviving capacity (evaluated by cell viability) is dependent on de novo protein synthesis. An adaptation process during slow freezing appears to be determinant for the yeast's high freeze tolerance phenotype. In the light of these observations, the yeast's response to freeze stress seems to be strongly dependent on the yeast strain, culture and freeze conditions. However, further studies will be necessary to clarify the molecular basis of cell adaptation to cold/freeze. In the case of the strain T. delbrueckii PYCC5323, the observed behaviour is particularly relevant in view to its utilisation for frozen dough production, and implies that the dough should be frozen at a slow, rather than at a fast rate.

In addition, and noteworthy from a methodological point of view, the results regarding the correlation observed along freezing between the loss of membrane integrity and the cell proliferative capacity, validate the application of flow cytometry and the use of the fluorochrome PI as a measure of viability of cells subjected to freeze stress. Therefore, and contrary to other stress conditions [25], the assessment of PI cells by flow cytometry, as a method to determine cell viability either in T. delbrueckii or S. cerevisiae, can replace the more laborious and time consuming determination of CFU counts.


We are grateful to Prof. Victor Costa for laboratory support and for critical reading of the manuscript. Cecília Alves-Araújo was supported by a PhD grant (PRAXIS XXI/BD/21543/99) from Fundação para a Ciência e a Tecnologia, Portugal.


  • Editor: B.A. Prior


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
View Abstract