OUP user menu

A respiratory-deficient mutation associated with high salt sensitivity in Kluyveromyces lactis

Paola Goffrini
DOI: http://dx.doi.org/10.1111/j.1567-1364.2006.00148.x 180-187 First published online: 1 March 2007


A salt-sensitive mutant of Kluyveromyces lactis was isolated that was unable to grow in high-salt media. This mutant was also respiratory-deficient and temperature-sensitive for growth. The mutation mapped in a single nuclear gene that is the ortholog of BCS1 of Saccharomyces cerevisiae. The BCS1 product is a mitochondrial protein required for the assembly of respiratory complex III. The bcs1 mutation of S. cerevisiae leads to a loss of respiration, but, unlike in K. lactis, it is not accompanied by salt sensitivity. All the respiratory-deficient K. lactis mutants tested were found to be salt-sensitive compared to their isogenic wild-type strains. In the presence of the respiratory inhibitor antimycin A, the wild-type strain also became salt-sensitive. By contrast, none of the S. cerevisiae respiratory-deficient mutants tested showed increased salt sensitivity. The salt sensitivity of the Klbcs1 mutant, but not its respiratory deficiency, was suppressed by the multicopy KlVMA13 gene, a homolog of the S. cerevisiae VMA13 gene encoding a subunit of the vacuolar H+-ATPase. These results suggest that cellular salt homeostasis in K. lactis is strongly dependent on mitochondrial respiratory activity, and/or that the ion homeostasis of mitochondria themselves could be a primary target of salt stress.

  • salt resistance
  • respiratory deficiency
  • Kluyveromyces lactis
  • yeast
  • BCS1
  • VMA13


Yeast species are resistant, to various degrees, to high-salt environments. In the mechanisms of this resistance, we may distinguish two processes: (1) rapid primary response of the cell to dehydration; and (2) slow adaptation to the high-salt environments.

The first process, extensively studied in Saccharomyces cerevisiae, appears to be associated with increased turnover/accumulation of glycerol (or polyalcohol) and trehalose (Hohmann, 2002), and with the exclusion from the cell of a harmful solute such as NaCl, mediated by the P-type ATPase Ena1p and the sodium–proton antiporter Nha1p (Serrano, 1997; Patterson, 1999). The second process, which involves slow adaptation to high salt, is still poorly explored. In some cases, genomic rearrangements, such as gene amplification, may be involved (Prior, 1996; Albrecht, 2000). Studies of salt-sensitive mutations have revealed a wide range of associated phenotypes. In S. cerevisiae, the expression of at least 18 genes was strongly induced by high salt (Blomberg, 1995).

In the present work, to enable the study of salt sensitivity in Kluyveromyces lactis we selected mutants with high salt sensitivity. Among them, we found a new kind of mutant that showed, in addition to its sensitivity to high concentrations of NaCl, KCl and LiCl, a complete loss of respiratory activity. The mutation mapped in the KlBCS1 gene, which codes for a mitochondrial protein involved in the assembly of respiratory complex III. A direct association of the salt sensitivity of yeast with a specific mitochondrial dysfunction has never been described previously. In the present work, the properties of the Klbcs1 mutation and the association between salt sensitivity and mitochondrial dysfunction will be described.

Materials and methods

Strains, media and growth conditions

Table 1 lists the K. lactis and S. cerevisiae strains and derived mutants used in this study. Escherichia coli DH10B was used as a cloning host and for DNA propagation.

View this table:

List of yeast strains

K. lactis
PM6-7AMAT a uraA1-1 adeT-600Chen. (1992)
PM6-7A/A16MAT a uraA1-1 adeT-600 Klbcs1This study
PM6-7A/ΔKlcyc1 (KG6)MAT a uraA1-1 adeT-600, Klcyc1::URA3Chen & Clark-Walker (1993)
MW179-1DMATαuraA1-1 leu2 lac4-8 trpA1 ade1M. Wésolowski-Louvel (University of Lyon 1)
MW179-1D/ΔKlcox14MATαuraA1-1 leu2 lac4-8 trpA1 ade1 Klcox14::kanMX4Fiori. (2000)
JBD100MATαtrp1 lac4-1 ura3Heus. (1990)
JBD100/Klcox18 (M5)MATαtrp1 lac4-1 ura3 Klcox18Hikkel. (1997)
JBD100/Klcytc1 (M3)MATαtrp1 lac4-1 ura3 Klcytc1Gbelská. (1996)
WMH9802/ΔKlqcr8MATαuraA1-1 leu2 lac4-8 trpA1 ade1 Klqcr8::URA3Brons. (2001)
2360/7MATαlysAParma collection
S. cerevisiae
W303-1AMAT a SUC2 ade2 can1 his3 leu2 trp1 ura3R. Rothstein (Columbia University)
W303-1A/Δbcs1MAT a SUC2 ade2 can1 his3 leu2 trp1 ura3 bcs1::HIS3Nobrega. (1992)
BY4741MAT a ura3Δ0 his3Δ1 leu2Δ0 met15Δ0Euroscarf collection
BY4741/Δsop1 (YPR032W)MAT a ura3Δ0 his3Δ1 leu2Δ0 met15Δ0sop1:: kanMX4Euroscarf collection
MH41-7B rho+ and rho derivativesMAT a ade2 his1Institut Curie, Orsay
MH32-12D rho+ and rho0 derivativesMAT a ade2 his1Institut Curie, Orsay
IL8-8C/HF71/rho0MATαtrp1 his1, rho0Institut Curie, Orsay
IL125-10C/rho0MATαura, rho0Institut Curie, Orsay
  • * MH41-7B/HF21 rho0, MH41-7B/OI-3 rho, MH41-7B/C7 rho, MH41-7B/P1 rho, MH32-12D/rho0 (Wésolowski-Louvel & Fukuhara, 1979)

Complete medium (YP) contained 1% Bacto yeast extract (Difco) and 1% Bacto peptone (Difco). It was supplemented with a carbon source at 2% (glucose, glycerol or others as specified). Minimal medium contained 0.7% Yeast Nitrogen Base without amino acids (Difco), and 2% glucose, supplemented with appropriate auxotrophic requirements. For plate tests, these media were solidified with 2% agar. Salt resistance/sensitivity was tested on YP-glucose plates containing the indicated concentrations of salts or sugars. The culture temperature was 28°C unless specified otherwise. Antimycin A (Sigma) was used at a concentration of 5 μM throughout. Genetic procedures for mating and sporulation were done on ME plates (5% malt extract, 3% Bacto agar).

Isolation of mutants

Yeast cells were mutagenized with UV irradiation according to Wésolowski-Louvel. (1992). Cells at a density of 108 cells mL−1 were exposed to 75 J m−2 of UV radiation. Survival was 20–30%. Cells were plated for single colonies on YP-glucose, and replica-plated on NaCl-containing medium. Putative mutants (negative growth on 1.5 M NaCl) were subcloned and retested for their salt-sensitive phenotype.

Cytochrome absorption spectra

Cells, grown to early stationary phase on YP medium supplemented with 2% glucose, were harvested by centrifugation, washed twice with cold (4°C) distilled water, and suspended in a volume of cold water twice the pellet volume. Differential spectra between reduced and oxidized cells were recorded at room temperature using a Cary 219 spectrometer. The bandwidth was 1 nm and the scan speed was 0.5 nm s−1. The cell suspension was reduced by sodium dithionite.

General methods

Published procedures were used for the transformation of K. lactis (Bianchi, 1987) and E. coli (Mandel & Higa, 1970). DNA manipulation, restriction enzyme digestion, plasmid engineering and standard techniques were performed according to Sambrook & Russel (2001). Sequencing was performed using a Beckman CEQ2000 automatic sequencer. Sequence analysis was performed with the blastp program (Altshul, 1990), and sequence alignment with the clustal w program (Thompson, 1994). The GenBank accession numbers for KlBCS1 and for KlVMA13 are AJ299738 and AJ547613, respectively.

The amplification of the mutated Klbcs1 allele was obtained by PCR with PM6-7A/A16 genomic DNA as a template and the primer pair KlBCS1F, 5′-AATCCGAGGCCTCGATTTCC-3′, and KlBCS1R, 5′-GGATGGACAACGAACGATAT-3′.


Isolation and phenotypic characterization of mutants of K. lactis with high salt sensitivity

The strain PM6-7A was UV-mutagenized, and about 25 000 cells were plated on YP-glucose plates. The colonies were replica-plated onto YP-glucose containing 1.5 M NaCl. Many colonies showed slow, leaky growth. Only three colonies (A8, A10 and A16) were clearly incapable of growing on the high-salt plates, as shown in Fig. 1. They were submitted to further tests of osmosensitivity on high-salt and high-sugar media. The results are shown in Table 2.


Salt sensitivity of Kluyveromyces lactis mutants. Strain PM6-7A and the three salt-sensitive mutants derived from it were streaked on YP-glucose plates with (b) or without (a) 1.5 M NaCl, and allowed to grow at 28°C for 2 days.

View this table:

Salt sensitivity of three mutants

StrainNaCl 1.5 MKCl 1.5 MLiCl 0.3 MGlucose 2 MSorbitol 2 MSucrose 1 M
  • The three mutants, A8, A10 and A16, are subclones of PM6-7A obtained by UV mutagenesis. All the tests were performed on YP-glucose plates supplemented with high-osmolarity solutes as indicated. The signs + and indicate the occurrence and absence of growth, respectively, as recorded after 3 days of incubation.

All the mutants were sensitive to 1.5 M NaCl, but not to high concentrations of sugars. They could be distinguished by their different sensitivities to 1.5 M KCl and to 300 mM LiCl. In particular, a mutant, called A16, was sensitive to all three salts, and moreover its growth was sensitive to high temperature (36°C) on YP-glucose. This mutant was also unable to grow on nonfermentable substrates (glycerol, ethanol and lactate). It was indeed respiratory-deficient, displaying an 80-fold decrease in oxygen consumption rate compared to the wild-type parental strain (data not shown). The cytochrome absorption spectra (Fig. 2) indicated that cytochrome b and cytochrome a+a3 were reduced by 25% and 65%, respectively. Our study focused on the particular pleiotropic mutant A16.


Cytochrome spectra of Kluyveromyces lactis PM6-7A and its respiratory-deficient mutant A16. The absorbance peaks at 550, 560 and 602 nm correspond to cytochrome c, cytochrome b and cytochromes a+a3, respectively.

The A16 mutant appears to have a single gene mutation, as suggested by two observations: (1) all spontaneous back mutations restored a complete wild-type phenotype, and (2) transformation of the mutant with a single gene (KlBCS1), as described below, fully complemented all the deficient phenotypes of A16. Although genetic crosses with wild-type laboratory strains gave diploids severely impaired in sporulation, we were able to obtain a diploid able to sporulate by crossing the mutant with the 2360/7 strain. Tetrad analysis demonstrated that the pleiotropic phenotype was due to a single nuclear mutation, as we obtained a 2 : 2 Mendelian segregation.

Cloning and characterization of the KlBCS1 gene

The A16 mutant was transformed with a K. lactis genomic library constructed on a centromere-based vector KCp491 carrying a URA3 marker (Prior, 1993). The Ura+ transformants were tested for their salt sensitivity. Among them, two clones were capable of growing on 1.5 M NaCl. The Ura+ phenotype and the salt resistance cosegregated in the course of spontaneous loss of the marker. The two transformants also recovered temperature resistance and respiratory competence. Both carried an identical plasmid (named pOSME) that contained a DNA insert of 6.0 kb. Fragments of this DNA were subcloned, and a segment spanning a 2.7-kb EcoRI fragment (carried by a plasmid named pOSME/E27) was found to be responsible for the transformed phenotype. The insert contained a putative single ORF of 1.35 kb. The predicted product of this DNA was a 450-amino-acid protein that showed 69% identity with the Bcs1p of S. cerevisiae. This is a nuclear-encoded mitochondrial protein of the AAA family (ATPase associated with diverse cellular activities) that in S. cerevisiae controls the assembly of the cytochrome bc1 complex and leads to a total absence of complex III activity. It has been proposed that Bcs1p acts as an ATP-dependent chaperone maintaining the precomplex in a competent state for the subsequent assembly of Rieske FeS and Qcr10p proteins (Nobrega, 1992; Cruciat, 1999).

The complex III deficiency in the bcs1 mutant also affects the amount of complex IV assembly and activity. In the cytochrome absorption spectra of the Klbcs1 mutant (Fig. 2), the effect on cytochromes a+a3 of complex IV was more evident; the effect on cytochrome b was less pronounced but distinguishable from that of the wild type. With regard to cytochrome c1, the negative effect of the mutation is evident from the reduced shoulder on the right side of the peak at 550 nm. Taken together, these results indicated that there was the same pleiotropic effect on cytochrome assembly in the K. lactis bcs1 mutant as was observed in the S. cerevisiae mutant.

As expected, the structural organization of the cloned K. lactis gene was similar to that of BCS1, showing the presence of two supposed ATP-binding motifs and a mitochondria-targeting signal (Fig. 3).


Structure of KlBcs1p as deduced from the DNA sequence. The amino acid sequence of the KlBCS1 product was deduced from the nucleotide sequence. Comparison with the Saccharomyces cerevisiae ortholog (Nobrega, 1992) clearly identified characteristic sequence motifs as indicated. The lower half of the figure shows the DNA sequence-deduced structure of KlBcs1p in the A16 mutant.

The mutant allele of the A16 strain was also sequenced by means of triplicate PCR amplification of the mutant DNA. The mutation corresponded to the introduction of a stop codon in the middle of the gene, resulting in a large deletion of the C-terminal half of the protein, with a loss of the putative ATP-binding motifs (Fig. 3).

Functional complementation of S. cerevisiae bcs1 mutation by the cloned K. lactis gene

Given the structural similarity of the cloned gene to BCS1, we carried out a functional complementation experiment. The cloned K. lactis DNA on the KCp491 vector that can also replicate in S. cerevisiae was transformed into the S. cerevisiaeΔbcs1 mutant. The transformants fully recovered the ability to grow on nonfermentable carbon sources (Table 3). The K. lactis gene was therefore named KlBCS1, and the mutant allele Klbcs1. Having observed a functional homology between KlBCS1 and BCS1, we wanted to know whether in S. cerevisiae the BCS1 gene is involved in salt resistance. In contrast to what was observed in K. lactis, the S. cerevisiaeΔbcs1 mutant did not show increased sensitivity to 1 M NaCl compared to its isogenic wild type (1 M was the highest concentration tolerated by the wild-type S. cerevisiae strains used here, as compared to 1.5 M for K. lactis strains) (Table 3). Moreover, the mutant was also not temperature-sensitive for growth. Thus the phenotypes of the bcs1 mutation clearly differed between the two species.

View this table:

Respiratory deficiency and salt sensitivity: comparison between Kluyveromyces lactis and Saccharomyces cerevisiae

K. lactis
PM6-7A+antimycin A
MW179-1D+antimycin A
JBD100+antimycin A
S. cerevisiae
MH41-7B, rho+++
MH41-7B/HF21, rho0+
MH41-7B/OI-3, rho+
MH41-7B/C7, rho+
MH41-7B/P1, rho+
MH32-12D, rho+++
IL8-8C/HF71, rho0+
W303-1A+antimycin A+
BY4741+antimycin A+
  • Kluyveromyces lactis and Saccharomyces cerevisiae strains were streaked on YP plates containing glycerol (test for respiratory competence) or glucose and NaCl (test for salt sensitivity), as indicated.+[KlBCS1] indicates the presence of the monocopy plasmid carrying the KlBCS1 gene. Note that for S. cerevisiae, the salt sensitivity test was performed on 1 M NaCl, the maximal salt concentration tolerated by the wild-type strains, as compared to 1.5 M for K. lactis. The K. lactis wild-type strains used here cease to grow on 1.7 M NaCl.

Does the salt-sensitive phenotype always accompany the respiratory deficiency in K. lactis?

In order to know whether the salt sensitivity was due to the Klbcs1 mutation per se or to the respiratory deficiency resulting from the mutation, we examined the salt sensitivity of available K. lactis respiratory-deficient mutants. Several respiratory-deficient mutants of S. cerevisiae were also included for comparison. The results obtained (Table 3) indicated that all the respiratory-deficient mutants of K. lactis were sensitive to 1.5 M NaCl (Fig. 4). These were mutants of cytochrome c, cytochrome oxidase subunits and complex III subunits, respectively. One exception was a cytochrome c1 mutant that grew on the high-salt medium (see Discussion). The strong correlation between respiratory deficiency and salt sensitivity was further supported by the observation that the wild-type K. lactis strains became salt-sensitive in the presence of the respiratory inhibitor antimycin A (Table 3). By contrast, none of the respiratory-deficient mutants of S. cerevisiae, including bcs1, were salt-sensitive compared to isogenic wild-type strains. As a negative control, we used the sop1 mutant of S. cerevisiae, a well-known salt-sensitive strain (Larsson, 1998).


Salt-sensitive phenotype of Kluyveromyces lactis respiratory-deficient mutants. The salt sensitivity of various respiratory mutants of K. lactis was tested against either 1.5 M NaCl or 1.5 M KCl included on YP-glucose plates. The respiratory deficiency is shown by their absence of growth on YP-glycerol plates.

A multicopy suppressor of Klbcs1

The role of KlBcs1p in salt resistance is not obvious. In order to find possible linked elements, we looked for a multicopy suppressor of Klbcs1 mutation. The Klbcs1 mutant was transformed with a K. lactis genomic library carried by the multicopy K. lactis/S. cerevisiae shuttle vector pSK1 (Wésolowski-Louvel, 1988). Among the 6000 Ura+ transformants, one single clone recovered the ability to grow on 1.5 M NaCl (in YP-glucose). However, this clone remained respiratory-deficient and temperature-sensitive for growth. Therefore, the suppressor appeared to be extragenic. The suppressed clone contained a plasmid with a DNA insert of about 6 kb. The predicted product of the gene found in this segment has an identity of 43% with the protein encoded by the VMA13 gene of S. cerevisiae. The Vma13p of S. cerevisiae has been known to form part of the vacuolar H+-ATPase complex (V-ATPase), an ATP-dependent proton pump that acidifies the vacuolar compartment. Vma13p is thought to be an activator or a stabilizer of the multimeric V-ATPase complex (Anraku, 1992). The functional equivalence of KlVMA13 and VMA13 was confirmed by complementation of the vma13 mutation by KlVMA13 (recovery from Ca2+ sensitivity of the mutant; data not shown).

A contribution of V-ATPase to the mechanism of salt tolerance in yeast has been reported (Hamilton, 2002). ATP hydrolysis is coupled with active proton transport inside the vacuole, thus generating a chemical gradient that drives transport of ions such as Na+ and Ca2+, which is mediated by the Na+/H+ antiporter, NHX1 (Nass & Rao, 1998; Hirata, 2002).

If the observed suppression resulted from a major role played by V-ATPase in the ionic homeostasis of mitochondria, we might expect that the salt-sensitive phenotype of other K. lactis respiratory-deficient mutants might also be complemented by the multicopy KlVMA13 gene. However, this was not the case. The suppressor effect of KlVMA13 was specific for the Klbcs1 mutation, suggesting that if the V-ATPase contributes to the ionic balance of the mitochondria, this might occur through a mechanism involving Bcs1p.


Relationship between respiratory deficiency and salt resistance in K. lactis

Salt resistance involves many genes. Studies on several yeast species, in particular S. cerevisiae, have led to the identification of different types of mutation that display different phenotypes. A striking finding was the identification of BCS1 as a genetic determinant involved in salt resistance in K. lactis. Because it is difficult to imagine a specific direct involvement of Bcs1p in salt resistance, we hypothesized that the salt sensitivity of the Klbcs1 mutant may not be a specific phenotype of this particular mutation, but rather a general consequence of the respiratory deficiency. The observation that wild-type K. lactis strains became salt-sensitive when respiration was specifically blocked by antimycin A was in favor of this interpretation. When we investigated the osmotic response of other respiratory-deficient mutants of K. lactis, all of them, as expected, were salt-sensitive, except for a cytochrome c1 mutant that maintained the resistant phenotype of the parental strain. However, this mutant retained 20% of the respiratory capacity of its wild-type parent (Gbelská, 1996). Thus, this mutant may not be considered to be strictly respiration-negative. For this reason, it should not be considered an exception to the relationship between respiratory deficiency and salt sensitivity. Therefore, the results with all the respiratory-deficient mutants, as well as those obtained with antimycin A, indicate that the lack of respiratory activity was correlated with the salt sensitivity.

Difference between K. lactis and S. cerevisiae

Unlike in K. lactis, in S. cerevisiae the bcs1 mutant as well as all other respiratory-deficient mutants retain high salt resistance, which is similar to the level observed in the wild-type strains. Therefore, the two yeast species clearly differ by the presence/absence of a link between respiratory deficiency and salt sensitivity. A possible reason could be that the laboratory strains of S. cerevisiae originate mainly from fermentation media, and hence these strains show a preference for a fermentative life rather than a respiratory mode of growth. We would expect such yeasts to possess stress-resisting mechanisms that do not require respiratory metabolism. Conversely, K. lactis strains have a strong respiratory activity, and normally this microorganism has a respiratory mode of life. The reducing potential generated by its strong glucose 6-phosphate shunt has to be recycled by active respiratory activity. Therefore, this species may have developed a stress response mechanism that is more tightly associated with mitochondrial functions, in comparison with S. cerevisiae. A respiratory deficiency or a mitochondrial mutation may then result in increased sensitivity to certain stresses. Such an interpretation, perhaps oversimplified, can be experimentally tested by the use of other yeast species showing a high dependence of growth on respiratory activity. In those species, salt resistance may also be linked to active mitochondrial functions, and in this regard, the responses of S. cerevisiae to high salt may reflect an exceptional physiology of this species.

Finally, the specific suppression of multicopy KlVMA13 in the Klbcs1 mutation has not yet been explained. Moreover, a relationship of V-ATPase with the mitochondrial system has been reported by Ohya. (1991), who observed that some of the mutations of the V-ATPase complex were accompanied by a respiratory-deficient phenotype in S. cerevisiae. Recently, links between iron and copper metabolism and mitochondrial and vacuolar function have also been found. In particular, a role for VMA13 in metal trafficking has been demonstrated (van Bakel, 2005).


I would like to thank Hiroshi Fukuhara for encouragement and helpful discussions, and Marco Ventura for critical reading of the manuscript. I thank Drs Julius Šubík, Alexander Tzagoloff, Xin-Jie Chen, Janike Brons, Claudio Falcone and Micheline Wésolowski-Louvel for providing mutant strains. I thank Tiziana Monduzzi for skillful technical assistance. This work was supported by Fondazione Adriano Buzzati-Traverso.


  • Editor: Monique Bolotin-Fukuhara


View Abstract