OUP user menu

Fluorescence based assay of GAL system in yeast Saccharomyces cerevisiae

Mateja Novak Štagoj , Aleksandra Comino , Radovan Komel
DOI: http://dx.doi.org/10.1016/j.femsle.2005.01.041 105-110 First published online: 1 March 2005

Abstract

The GAL1 promoter is one of the strongest inducible promoters in the yeast Saccharomyces cerevisiae. In order to improve recombinant protein production we have developed a fluorescence based method for screening and evaluating the contribution of various gene deletions to protein expression from the GAL1 promoter. The level of protein synthesis was determined in 28 selected mutant strains simultaneously, by direct measurement of fluorescence in living cells using a microplate reader. The highest, 2.4-fold increase in GFP production was observed in a gal1 mutant strain. Deletion of GAL80 caused a 1.3-fold increase in fluorescence relative to the isogenic strain. GAL3, GAL4 and MTH1 gene deletion completely abrogated GFP synthesis. Growth of gal7, gal10 and gal3 also exhibited reduced fitness in galactose medium. Other genetic perturbations affected the GFP expression level only moderately. The fluorescence based method proved to be useful for screening genes involved in GAL1 promoter regulation and provides insight into more efficient manipulation of the GAL system.

Keywords
  • GAL system
  • Green fluorescent protein
  • Heterologous expression
  • Deletion strains
  • Saccharomyces cerevisiae

1 Introduction

Galactose utilization in the yeast Saccharomyces cerevisiae consists of a biochemical pathway that converts galactose into glucose-6-phosphate and a regulatory mechanism that controls whether the pathway is switched on or off. This process has been extensively reviewed and involves at least three types of proteins [1,2]. The transporter gene (GAL2) encodes a permease that transports galactose into the cell. GAL1, GAL7 and GAL10 genes code for three enzymes specific to galactose metabolism: galactokinase, transferase and epimerase, respectively. Regulatory genes GAL3, GAL4, GAL80 exert transcriptional control over transporter, enzymes and, to a certain extent, also over each other [3,4].

The GAL1, GAL10 and GAL7 gene promoters in the GAL gene system are among the strongest in S. cerevisiae and have been widely used for recombinant protein production. The GAL genes are tightly regulated, being repressed by glucose and induced by galactose. GAL gene expression requires the well-studied transcription activator protein Gal4p that binds to GAL gene promoters. Gal4p function is inhibited by Gal80p, which is bound to Gal4p, and by Mig1p, which represses expression of GAL1 and GAL4 in the presence of glucose [5].

Although extensively studied, many details remain unclear. Repression of the GAL system, where the Mig1 complex is believed to play an important role, is still not well understood. On the other hand, an extensive background of knowledge on galactose regulation offers the possibility of manipulating the system in a number of ways in order to improve its characteristics for protein production. In this paper, we evaluate the influence of 28 individual gene deletions on recombinant protein production using green fluorescent protein (GFP) as a well-established qualitative and quantitative reporter [68]. Based on new data sets obtained by whole genome screening [911] such as microarray data of various physiological conditions, two-hybrid screens, synthetic lethality data and availability of strains with a defined deletion of non-essential open reading frames, we built up a method for evaluating the contribution of several individual genes to protein expression from the GAL1 promoter. In addition to strains mutated in genes directly involved in galactose metabolism, we screened strains mutated in genes that are indirectly involved in the regulation of the GAL system, such as various transcription factors (MIG1, MED2, PGD1, NRG1, MIG2, CTI6) including members of the SAGA complex (NGG1, SPT3, SPT8, GCN5, ADA2) [12,13], chromatin binding (NPH6B), genes involved in carbohydrate metabolism (CSR2, SNF1, SNF4) and hexose transporter (MTH1) [http://yeastgenome.org/]. Products of UBL1 and YCL045 genes were selected as Gal80p-interacting proteins [14].

Here, we present a systemic approach to the comparative quantification of the influence of well-defined genetic background on the production level of recombinant protein. We have setup an experimental procedure for genome-scale screening to quantify the contribution of a large set of genes involved in the process of heterologous protein synthesis and processing.

2 Materials and methods

2.1 Yeast strains, media and growth conditions

All assays were carried out with strains isogenic to the haploid reference strain BY4741 (MATa; his3Δ1; leu2Δ0; met15Δ0; ura3Δ0) in which single genes of interest had been replaced with the kanMX deletion cassette. All strains were obtained by EUROSCARF, Germany (Table 1).

View this table:
1

S. cerevisiae deletion strainsa

Strain labelGene name (ORF deleted)Function of gene product
Y03157GAL1Gakactokinase
Y02692GAL2Galactose permease, also able to transport hexoses
Y03949GAL3Protein binding, involved in galactose induction of GAL genes
Y01044GAL4Transcriptional activator, positive regulator of GAL genes
Y06545GAL5 (PGM2)Phosphoglucomutase
Y01973GAL6 (LAP3)Transcriptional regulator activity
Y03155GAL7Galactose-1-phosphate uridyl transferase
Y03156GAL10Epimerase
Y01742GAL11RNA polymerase II transcription mediator activity
Y00520GAL80Transcription co-repressor activity
Y04403MIG1Transcription factor involved in glucose repression
Y03534NGG1Transcription cofactor involved in glucose repression, member of SAGA complex
Y04228SPT3Transcription factor, member of SAGA complex
Y02666SPT8Transcription factor, probably member of SAGA complex
Y07285GCN5Histone acetyltransferase activity, member of SAGA complex
Y04282ADA2Transcription cofactor, ADA and SAGA component
Y03229NHP6BChromatin binding (regulation of transcription from Pol II and Pol III promoter)
Y13701MED2RNA polymerase II transcription mediator
Y04393PGD1RNA polymerase II transcription mediator
Y03979NRG1Transcriptional repressor, involved in regulation of glucose repression
Y04575MIG2Specific RNA polymearse II transcription factor, involved in glucose represion
Y24311SNF1SNF1A/AMP-activated protein kinase, regulation of carbohydrate metabolism
Y04482SNF4Protein kinase activator, involved in release from glucose repression
Y03636MTH1Glucose transport, negative regulator of HXT gene expression
Y04814UBR1Ubiquitin-protein ligase
Y03452YCL045cMolecular function unknown
Y02070CTI6Transcription factor, positive regulation of transcription
Y05449CSR2Potential regulator of galactose and non-fermentable carbon sources utilization

Yeasts were cultured in the standard media YPD (1% Bacto-pepton (Difco), 2% yeast extract and 2% glucose) and in induction media YPD/Gal or YPRaff/Gal with 1% glucose/2% galactose or 1% raffinose/2% galactose, respectively. The cells were cultivated in sterile 96-well microtiter plates (Porvair Sciences, Shepperton, UK) at 30 °C in a microtiter plate shaker at 400 rpm (Tehtnica, Železniki, Slovenia). For fluorescence readings, cultures were grown overnight in YPD, centrifuged and inoculated at OD600= 1 into induction media. Cultures were incubated 24 h before fluorescence readings.

2.2 Plasmid construction and transformation

All enzymes were purchased from New England Biolabs (Beverly, MA, USA). YCplac111 [15] was digested with HindIII, blunt-ended by Klenow polymerase and re-ligated to obtain YCplac111ΔHindIII. ADH terminator was amplified from pJG4-5 (Clontech Laboratories, Palo Alto, CA, USA) by PCR (PTC-100TM programmable thermal controller, MJ Research Inc., Watertown, MA, USA) adding EcoRI recognition site to the 5′-end and ClaI site to the 3′-end (oligonucleotide sequences are available upon request). The fragment was inserted into the EcoRI/NarI site of YCplac111ΔHindIII. YCpGAL1 was created by cloning the GAL1 promoter obtained by PCR from pJG4-5 using 5′-PstI and 3′-BamHI modified oligonucleotides. YCpGAL1 plasmid was then digested with BamHI, treated with Mung Bean Nuclease and subsequently digested with EcoRI. GFP gene was excised from pGFP (Clontech Laboratories) with BamHI, blunt-ended by Mung Bean Nuclease, digested with EcoRI and cloned into the vector. Escherichia coli DH5α was used for plasmid propagation. The resulting plasmids, YCpGAL1-GFP and YCpGAL1, were finally transformed into the yeast deletion strains using a large-scale transformation protocol [16].

2.3 Protein extraction

Protein extracts were made from cells resuspended at a final density of 2 × 108 cells ml−1 in TE buffer (10 mM Tris–HCl pH 8.0, 10 mM EDTA) with acid-washed glass beads (0.45 mm diameter; Sigma, St. Louis, MO, USA). Protein concentration was determined by a protein assay reagent (Bio-Rad, Hercules, CA, USA) with bovine serum albumin as standard.

2.4 GFP fluorescence determination

A standard dilution series (0.80–5.00 μg ml−1) of GFP provided by Clontech was made in TE buffer as well as in untransformed BY4741 protein extract containing 0.50 μg μl−1 total proteins.

For the studies of protein production in viable cells, cultures were centrifuged and cell pellets re-suspended in 100 μl TE buffer in order to remove fluorescence background of the culture medium. Wells containing TE buffer only were used to detect fluorescence of the medium. The control wells contained YCpGAL1 transformed yeast strains. GFP florescence intensity was measured using a fluorescence spectrophotometer (LS55 microplate reader; Perkin–Elmer, Wellesley, MA, USA) at an excitation wavelength of 395 nm and an emission wavelength of 509 nm (2.5- and 5-nm slit widths). A 5-s reading was set for each micro-well. Standards and samples were measured in triplicate and averaged. All samples had a reading within the range of the standard curve.

2.5 Cell number determination

The cell number of yeast cultures was determined by microtiter plate assay using Calcofluor White M2R [17]. Cells were stained with Calcofluor White M2R after GFP fluorescence reading to prevent fluorescence interference. Fluorescence analysis was performed at an emission wavelength of 460 nm, with excitation of 360 nm.

To determine specific productivity, GFP emission fluorescence readings were normalized to the unit of OD600.

3 Results and discussion

3.1 GFP amount determination

Standard curves of purified GFP were established in either TE buffer or in a lysate of untransformed yeast cells to estimate possible quenching effect (Fig. 1). The GFP has an absorbance/excitation peak at 395 nm with a minor peak at 475 nm and has a peak emission at 509 nm, with a shoulder at 540 nm. The plot of the emission fluorescence against the amount of GFP in a well (Fig. 1) was linear over the range of 80–500 ng GFP with a high degree of confidence. Quantitation of GFP fluorescence from a calibration curve of GFP in protein extracts of untransformed cells determined by linear regression could also be applied, but with a slightly lower degree of confidence (R2= 0.979).

1

Standard curve using purified GFP in TE buffer (∘); and purified GFP mixed with protein extract of untransformed BY4741 (△). Fluorescence is given in arbitrary units (AU).

The basal level of fluorescence should be expected to affect the standard curve by increasing the fluorescence uniformly to each standard. This would result in a parallel shift of the standard curve measured in the presence of yeast protein extract above that measured in TE. We repeatedly observed, however, a change in nearly parallel shifts below the standard curve. As shown in Fig. 1, the slope of the curve of GFP standard dilutions in the protein extract is a little lower than that of the standard dilution in TE buffer. This indicates a quenching of fluorescence in the presence of yeast extract and results in lower fluorescence readings than expected from GFP concentration. The quenching effect may be due to the yeast substances that absorb UV-light thereby making it unavailable for GFP chromophore.

The fluorescence signal of extracted proteins from cells expressing GFP was measured and compared to the fluorescence intensity of viable cells. Equivalent results were obtained from both measurements (data not shown).

3.2 Level of protein production in various yeast mutant strains

The availability of complete genomic sequence and technologies that allow comprehensive analysis of global regulatory networks has greatly expanded our ability to monitor the internal state of cells. Here, we describe fluorescence based screening of various S. cerevisiae deletion strains (Table 1) and its use as a tool to study and compare the contribution of several individual gene products to heterologous protein production. We used the existing biochemical model of regulation in the galactose utilization pathway [1,2] to predict the effects of genetic perturbations on the expression level of GFP from the GAL1 promoter. Many of these predictions would have been obvious from previous experiments, however, the majority of them were less evident and still never quantified.

Guided by the current model wild-type (wt) and nine genetically altered yeast strains were examined, each one with a complete deletion of one of the nine GAL genes; transporter (GAL2), enzymatic (GAL1, GAL7, GAL10, GAL5) or regulatory (GAL4, GAL3, GAL80 or GAL6). We compared and quantified the effects of these GAL mutations on protein expression using established fluorescence-based method in different environmental conditions (Fig. 2). Measuring kinetics of GFP, we observed that the maximal level of protein accumulation was achieved in stationary phase, after 24 h of cultivation. Since similar results were obtained in non-repressing media containing raffinose (instead of glucose) under the same experimental conditions, we assumed that in stationary-phase cultures, glucose (repressor) was exhausted from the media.

2

Level of GFP in GAL mutated strains cultured 24 h in induction media. Yeast were cultivated in YPD, inoculated at 1.0 OD600 in YPGlu/Gal (2% galactose, 1% glucose) and cultured for 24 h. Experiments were performed in triplicate. Fluorescence is given in arbitrary units (AU).

Gal4p activates genes necessary for galactose metabolism and is among the best-characterized transcriptional activators. By the established method we confirmed that perturbation of the GAL4 or GAL3 genes totally abolished protein synthesis from the GAL1 promoter.

We repeatedly observed the highest level of protein production from the strain with a deletion of the GAL1 gene encoding galactokinase. GFP production was 2.4-fold higher in this gal1 mutant strain. We verified previously reported data that galactose is a gratuitous inducer of the GAL promoter in a gal1 mutant strain [1821]. Moreover, we also proved that in the gal1 mutant strain, protein production was significantly higher not only in comparison to the wild-type, but also in comparison to the other 27 deletion strains.

The level of protein was also higher in gal80 and slightly elevated in gal6 mutant strains. It is known that Gal6p produces another regulatory factor thought to repress the GAL genes in a manner similar to Gal80p [22]. By fluorescence measurement we confirmed that in a gal80 mutant, lacking the regulatory protein Gal80, GAL genes do not require galactose for the induction; they are constitutively expressed, during 24 h growth on glucose (Fig. 2) or raffinose (data not shown).

By the established fluorescence assay we compared the effects of an additional 20 gene deletions on GFP expression from the GAL1 promoter. The mutant strains were selected using data accessible via the Saccharomyces genome database (http://www.yeastgenome.org/) and references therein (Table 1). Data are presented in Fig. 3.

3

Correlation between optical density (OD600) and fluorescence of different mutant strains expressing GFP. Yeast were cultivated in YPD, inoculated at 1.0 OD600 in YPRaff/Gal (2% galactose) and cultured for 24 h. Fluorescence and cell number determinations were performed as described in Section 2.

The majority of screened genes did not appear to be required for heterologous protein expression and a high proportion of gene deletions seemed to have no detectable effects on the biochemical processes that result in alteration of growth rate on galactose. However, growth of some deletion strains exhibited reduced fitness in galactose medium. Consequently, in those strains the heterologous protein production was diminished. Deletion of the gene encoding either uridyl transferase (Gal7p) or UDP-galactose epimerase (Gal10p), two enzymes needed to convert galactose 1-phospate into glucose 1-phosphate, resulted in decreased and altered growth on galactose medium. This was probably due to the accumulation of the potentially harmful metabolite galactose 1-phosphate [23]. The product of the MTH1 gene is a negative regulator of some HXTs gene expression [9] and deletion of that gene surprisingly caused significant repression of GFP production. Nevertheless, alteration of regulation of the genetic system may have unexpected consequences, since regulatory proteins may have functions beyond current knowledge.

We demonstrated that fluorescence measurements in a microtiter plate could be used as a fast, simple and reliable method for high-throughput quantitation of GFP. We performed a fluorescence based approach allowing simultaneous detection and quantification of the corresponding cellular response in relation to several genetic perturbations. By the established method, we were able to screen and emphasize the effect of 28 gene deletions on protein synthesis from the GAL1 promoter. The method combines the advantages of a fast, simple and accurate technique that is of low cost and is reliable for whole genome screening. Accuracy, reproducibility, sensitivity and speed make the assay well suited for large-scale as well as automated fluorescence-based applications.

Acknowledgements

The authors thank Jelka Lenarčič for excellent technical assistance. This work was supported by the Ministry of Education, Science and Sports of Slovenia, Grant No. 3311-01-831711 to M.N.Š.

Footnotes

  • 1 It was early in January 15, 2005 when Dr. Aleksandra Comino tragically passed away. This article is dedicated to her memory, for her remarkable scientific contribution and for being a wonderful colleague and friend.

References

  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].
View Abstract