Plasmid 69 (2013) 114–117

Contents lists available at SciVerse ScienceDirect

Plasmid

journal homepage: www.elsevier .com/ locate /yplas

Short Communication

Construction of integrative plasmids suitable for genetic modificationof industrial strains of Saccharomyces cerevisiae

Fernanda Cristina Bezerra Leite a, Rute Salgues Gueiros dos Anjos a,Anna Carla Moreira Basilio a, Guilherme Felipe Carvalho Leal a,Diogo Ardaillon Simões a,c, Marcos A. de Morais Jr. a,b,⇑a Interdepartmental Metabolic Engineering Research Group, Federal University of Pernambuco, Recife, PE, Brazilb Department of Genetics, Federal University of Pernambuco, Recife, PE, Brazilc Departament of Biochemistry, Federal University of Pernambuco, Recife, PE, Brazil

a r t i c l e i n f o

Article history:Received 26 April 2010Accepted 24 September 2012Available online 4 October 2012Communicated by Dhruba K. Chattoraj

Keywords:Ethanol fermentationIndustrial strainIntegrative cassetteLactose assimilationYeast transformation

0147-619X/$ - see front matter � 2012 Elsevier Inchttp://dx.doi.org/10.1016/j.plasmid.2012.09.004

⇑ Corresponding author address: DepartmentUniversity of Pernambuco, Av., Moraes Rego, 123550.670-901 Recife, PE, Brazil. Fax: +55 81 21268522

E-mail address: marcos.morais@pesquisador.cnpURL: http://www.ufpe.br/nem (M.A. de Morais).

a b s t r a c t

The development of efficient tools for genetic modification of industrial yeast strains is oneof the challenges that face the use of recombinant cells in industrial processes. In thisstudy, we examine how the construction of two complementary integrative vectors canfulfill the major requirements of industrial recombinant yeast strains: the use of lactoseassimilation genes as a food-grade yeast selection marker, and a system of integration thatdoes not leave hazardous genes in the host genome and involves minimal interference inthe yeast physiology. The pFB plasmid set was constructed to co-integrate both LAC4-basedand LAC12-based cassettes into the ribosomal DNA (rDNA) locus to allow yeast cells to beselected in lactose medium. This phenotype can also be used to trace the recombinant cellsin the environment by simply being plated on X-gal medium. The excisable trait of theLAC12 marker allows the introduction of many different heterologous genes, and makesit possible to introduce a complete heterologous metabolic pathway. The cloned heterolo-gous genes can be highly expressed under the strong and constitutive TPI1 gene promoter,which can be exchanged for easy digestion of enzymes if necessary. This platform wasintroduced into Saccharomyces cerevisiae JP1 industrial strain where a recombinant withhigh stability of markers was produced without any change in the yeast physiology. Thus,it proved to be an efficient tool for the genetic modification of industrial strains.

� 2012 Elsevier Inc. All rights reserved.

The yeast Saccharomyces cerevisiae is one of the mostimportant industrial microorganisms and well-known forits activity in the production of fermented beverages,bread, bioethanol and food ingredients (Ostergaard et al.,2000). Great advances have been made in the academicworld regarding the biology of this organism in the lastfew decades, including its complete genome sequencingin 1996 (Goffeau et al., 1996) and the setting up of well-

. All rights reserved.

of Genetics, FederalCidade Universitária,.

q.br (M.A. de Morais).

established protocols and methods for the genetic modifi-cation of laboratory strains (Ausubel et al., 1989; Romanoset al., 1992). However, the construction and use of geneti-cally modified (GM) derivatives of industrial strains is stillin its early stages (Pronk, 2002; Schuller and Casal, 2005).

This study describes the construction of integrative vec-tors for heterologous gene expression in industrial strainsof S. cerevisiae for the use of GM strains in fermentationprocesses. The main characteristics of these vectors includethe following: (a) the possibility of high-copy number inte-gration in the target genome and high genomic stability ofthe recombinant, (b) the use of an environmentally-friendly selection marker and the possibility of markerrescue for multiple integrations, (c) the traceability of the

F.C.B. Leite et al. / Plasmid 69 (2013) 114–117 115

recombinant strains regarding intellectual propertymatters, (d) the minimization of metabolic burden causedby the integrative cassette and (e) the high-level constitu-tive expression of heterologous genes.

The present cloning system is not employed as a form ofresistance to antibiotics or toxic compounds but is based onmetabolic selection by lactose assimilation instead. As S.cerevisiae is not able to assimilate this disaccharide, recom-binant cells harboring LAC genes can be easily selected inlactose-containing medium. The biotechnological applica-tions of LAC genes from Kluyveromyces have been fullyanalyzed by Rubio-Teixeira (2006). However, despite thelarge number of lactose-assimilating S. cerevisiae recombi-nants, the main application focused on the cultivation ofwhey permeate for biomass production of baker’s yeast orfor use in ethanol production. Heterologous LAC4 has beenlargely used as a reporter gene for transcription regulationstudies of endogenous yeast genes (Sassi et al., 2009). Inaddition, the components of the LAC regulon have beenused for protein–protein interaction studies by means ofthe two-hybrid approach (examined by Williamson andSutcliffe, 2010). Interestingly, the idea of using one or bothLAC genes as selective markers has hardly been discussed inthe literature (Romanos et al., 1992; Pronk, 2002; Rubio-Teixeira, 2006). For example, the expression of bacteriallacO operator was used to select Pichia pastoris recombi-nants (Rao et al., 2010), and while this work was being con-ducted, an integron-based gene cassette was reported thatcontained the lacZY operon for the selection of Pseudomonasstutzeri in lactose medium (Gestal et al., 2011).

As currently conceived, genome integration is targetedat the non-transcribed site 2 (NTS2) of the rDNA locus, thatoffers as many as 300 copies per haploid genome. This al-lows either multiple integrations of the same heterologousgene or a single integration of many different heterologousgenes (Lopes et al., 1991, 1996). Another significant featureis that only one recombinant is needed after the transfor-mation, an important factor when it is taken into accountthat most industrial strains are averse to genetic transfor-mation (Silva-Filho et al., 2005). This selection of the re-combinant can be achieved by plating transformed cellsonto lactose medium. At the same time, these recombinantyeasts will have to face very complex and unstable indus-trial environments, which involve a variation in their phys-ico-chemical and microbiological parameters (Silva-Filhoet al., 2005). Thus, any genetic modification should onlyinterfere to a minimal degree with the fitness and physio-logical traits of the cells (Pronk, 2002).

The set of pFB plasmids was constructed in the follow-ing stages: (1) the construction of pBASE with minimumbacterial plasmid sequences for the plasmid manipulationin Escherichia coli; (2) cloning of rDNA sequences togetherwith yeast TPI1 gene promoter amplified directly from theindustrial S. cerevisiae JP1 strain genome; (3) cloning fromGRAS Kluyveromyces lactis yeast the LAC4 gene (in the caseof pFB-LAC4) that encodes b-galactosidase or the LAC12gene (in the case of pFB-LAC12) that encodes lactose per-mease; (4) adaptation of cloning sites. The pFB plasmidset is shown in Fig. 1 (see Supplementary Material for a de-tailed description of the plasmid construction). The result-ing integrative vectors can be digested with SmaI to

produce the 6,751-bp (pFB-LAC4) and 5,623-bp (pFB-LAC12) integrative fragments that can then recombineand be integrated into the yeast rDNA locus, while theremaining 2,765-bp bacterial ampr/Ori fragments will belost. After being integrated in the yeast genome, the plas-mid set produces recombinant cells without any bacterialfragments or antibiotic-resistance genes. The heterologousgene, which in theoretical terms is referred to here as a Hetgene, is cloned into pFB-LAC12 so that it can be expressedin the industrial cells. After the SmaI restriction, the yeastcells will be transformed to produce GM cells that are[LAC4+LAC12+Het1+] genotype and lactose-assimilatingLac+ phenotype. Since the LAC genes were obtained fromthe GRAS yeast Kluyveromyces marxianus and no bacterialsequence is expected to be present in the final arrange-ment of the integrative cassette, there should be nobio-safety problems to impede the release of these recom-binant cells for industrial purposes. Moreover, it has beenshown that horizontal transference of heterologous genesbetween the recombinant yeast and the bacterial cells inthe environment is a rare phenomenon in nature (Adamet al., 1999). In addition, the LAC12-containing cassettewas constructed so that it could be flanked by loxP sites.This site is the target for bacterial Cre recombinase andtwo in tandem sequences can recombine to pop-out theinternal sequence, leaving a small 30-bp sequence as arecord of that process (Gueldener et al., 2002). The left loxPsite has no bio-safety consequence and in addition can beused as a signature of the recombinant cells. A newheterologous Het2 gene can be cloned into pFB-LAC12and the integration cassette can be used to transform[LAC4+ lac12� Het1 Lac�] cells to restore the lactoseassimilation phenotype [LAC4+ LAC12+ Het1 Het2 Lac+]. Thisprocedure can be repeated several times for multipleheterologous gene integration.

The stability of the integrations depends on the size ofthe integration cassette, and cannot be larger than therepetitive unit of the rDNA locus (9.1 kb) itself. Bigger frag-ments will make the integration cassete unstable (Lopeset al., 1996). In the case of pFB-LAC12 (5.6 kb), it shouldbe possible to clone heterologous genes as large as 4 kbin order to maintain the length of the final construct belowthat limit. This is much larger than most of the bacterialand yeast genes and enough for the majority of plant andanimal open reading frames. Moreover, heterogeneity inthe rDNA repeats by introduction of genome alterations(insertions, deletions or point mutations) tends to behomogeneized by gene conversion events (Ganley andKobayashi, 2011). Thus, higher stability of the geneticmodification is expected if more than one copy of theexpression cassette is integrated (Lopes et al., 1991). S.cerevisiae JP1 strain was isolated from the industrial pro-cess for ethanol production (Silva-Filho et al., 2005). Thisis diploid and homothallic yeast that can be successfullytransformed with episomal plasmids (Reis et al., 2012).JP1 cells were transformed by means of the LiCl2 method(Sambrook et al., 1989) with digested pFB-LAC4 and pFB-LAC12 and the recombinant cells were selected in platescontaining YPLac (10 g L�1 yeast extract, 20 g L�1 peptone,20 g L�1 lactose, 20 g L�1 agar) medium. The transforma-tion efficiency was around 5 transformants (lg of DNA)�1,

Fig. 1. Physical map of pFB plasmid set (pFB-LAC4 and pFB-LAC12) designed for this study. The green boxes represent the integration sites at rDNA locusafter the digestion of the plasmids with SmaI. TPS-MCS sequence represents the promoter region of the TPS1 gene followed by the multiple-cloning sites forheterologous gene cloning. The main restriction sites and their respective positions are shown.

116 F.C.B. Leite et al. / Plasmid 69 (2013) 114–117

similar to what was previously reported for industrialstrains (Silva-Filho et al., 2005). To conduct the analysisof recombinant stability, JP1Lac+ cells were cultivated inYPD (10 g L�1 yeast extract, 20 g L�1 peptone, 20 g L�1 glu-cose) medium in agitated shake-flask batches for 24 h at33 �C for two weeks. For each batch, fresh medium wasinoculated with the preceding culture to initial concentra-tion of A660 = 0.2 and allowed to grow for another 24 h. Ini-tial and final cell concentration of each batch were used tocalculate the number of cell generations. In addition, thecells were collected after 24 h, washed in sterile water, di-luted, spread onto plates containing YPGal (10 g L�1 yeastextract, 20 g L�1 peptone, 20 g L�1 galactose, 20 g L�1 agar)medium supplemented with X-gal at 40 lg/mL (Sambrooket al., 1989) and incubated at 30 �C for two days. The per-centage of reverting cells, in which the integrated LACgenes were pop-out from the yeast genome via loxP sites,was calculated by the number of white colonies over thetotal of yeast colonies and the stability of recombinantcells was estimated as the percentage of blue colonies overthe cell generations. The results showed that only 1.3% ofthe recombinant cells lost their b-galactosidase activity,which made them turn white in X-gal plates, after 100 gen-erations of successive cell cultivation (see SupplementaryMaterial). Spontaneous intrachromosomal recombinationof loxP may occur after cultivation in YPD medium, andin the present case we observed a recombination rate ofc.a. 10�8 per cell generation. However, a faster removal ofthe LAC12 cassette could be achieved by transformationwith pSH65 episomal plasmid that carries the bleomicinresistance gene bler (Gueldener et al., 2002). The trans-formed cells are screened for bleomicin-resistance lac-tose-negative phenotype [LAC4+ lac12� Het1 Lac� bler]and then cultivated in non-selective medium for plasmidloss. Recombinant cells were tested for cell growth rateby cultivation in synthetic complete medium (1.7 g L�1

YNB, 5 g L�1 ammonium sulfate, 20 g L�1 glucose) over-night at 33 �C with constant agitation (200 rpm). Pre-growing cells were used to inoculate the same mediumto A660 = 0.2 initial concentration. Samples of the cultiva-tion were withdrawn every hour for absorbance measure-ment and the slopes of semi-Log growth curves were used

to calculate the specific growth rate (l). The resultsshowed that JP1 and its recombinant JP1Lac+ cells had spe-cific growth rates of 0.38 and 0.42 h�1 in YPD medium,respectively, indicating that the integration of both frag-ments at the rDNA locus did not affect cell growth. More-over, when recombinant JP1Lac+ cells were spread ontoYPD + X-gal plates, all the colonies had a white color. Thisshowed glucose repression of LAC genes which may reducethe possibility of metabolic burden being imposed by thecombination of marker and heterologous gene expressionson the host cells (Romanos et al., 1992) during the cultiva-tion and fermentation in glucose-rich medium such as su-gar cane juice and molasses (see Supplementary Material).

The heterologous target genes must be cloned under theregulation of the constitutive promoter of the gene TPI1,which is one of the strongest promoters in S. cerevisiae gen-ome (Romanos et al., 1992). In the present work we usedthe alaD gene (1.2 kb) that encodes an alanine dehydroge-nase from Bacillus subtilis. The gene was amplified frompPK243 plasmid, which was kindly provided by Prof. JackPronk, TU Delft, The Netherlands, and cloned into pFB-LAC12. The GM cells [LAC4+ LAC12+ alaD Lac+] were culti-vated in synthetic complete medium (as described above)to late log phase in triplicate. Cell-free extracts were pre-pared for enzyme activity determination (see Supplemen-tary Material for detailed procedure). The results showedthat the GM cells produced alanine dehydrogenase at 6.1(±0.83) U mg�1, which proved the strength of the constitu-tive TPI1 gene promoter. The specific cell growth rateswere calculated as 0.44 and 0.45 h�1 for JP1 strain and itsalaD+ recombinant, respectively. Thus, the heterologousgene expression did not affect the cell metabolism. Lopeset al. (1996) affirmed that the expression of heterologousgenes within rDNA repeat does not affect mitotic stabilityof the recombinant either. Another important characteris-tic of this system is that the design of the cloning sitesmakes it easy to replace the promoter with double diges-tion of pFB-LAC12 with PmeI and any other enzyme thattargets one of the restriction sites at the MCS. This processallows further modifications to be made and enables theuser to regulate the expression of the target gene. Forexample, the use of the glucose-repressible ADH2 gene

F.C.B. Leite et al. / Plasmid 69 (2013) 114–117 117

promoter should offer a good platform for controlled het-erologous gene expression (Cunha et al., 2006). Thus, dur-ing the fermentation of the glucose-containing substratessuch as sugarcane, corn, wheat, etc., both the heterologousgene and the marker genes will be switched-off. Upon glu-cose exhaustion, at the end of fermentative phase, the cellscan be further used for heterologous protein production.The advantages are clear: there is no cost for biomass pro-duction since it is derived from a previous industrial pro-cess, no problem with recombinant instability duringfermentation phase, and a huge supply of biomass for het-erologous protein production of the industrial plant.

In addition to acting as a selective marker, when inte-grated with the LAC4 gene, the yeast genome may also beused to identify and quantify recombinant cells in theyeast population. Two experiments were conducted forthis purpose. First of all, JP1 cells were mixed with JP1Lac+

cells in different proportions and plated onto YPGal + X-galplates to 107 CFU per plate. After 24 h of incubation, onesingle recombinant dark-blue colony could be observedin a layer of white colonies. Secondly, cells of 38 yeast spe-cies identified from the ethanol fermentation process(Basílio et al., 2008) were plated onto YPGal + X-gal plates.Only cells of Dekkera bruxellensis produced pale-blue colo-nies after 48 h of incubation, which contrasted with the24 h growth of dark-blue colonies of JP1Lac+ and of K.marxianus CBS 6556 used as a control strain (see Supple-mentary Material). As a result, Lac+ is a fairly distinguish-able phenotype in complex populations and the presenceof recombinant cells in the yeast population can be tracedand quantified among the huge diversity that are found inthe fermentation process (Basílio et al., 2008).

Thus, the platform which has been outlined here fulfillsall the requirements for stable transformation and easyrecovery of recombinant cells derived from industrial S.cerevisiae strains. Moreover, it can meet all the bio-safetyrequirements needed to release them in low-contentionindustrial environments such as those for ethanol-fuelplants for fermentation.

Acknowledgments

The authors would like to express their thanks to thefollowing Brazilian agencies CNPq, CAPES, BNB/FUNDECIand FACEPE for their financial support and scholarships.

Appendix A. Supplementary data

Supplementary data associated with this article can befound, in the online version, at http://dx.doi.org/10.1016/j.plasmid.2012.09.004.

References

Adam, A.C., González-Blasco, G., Rubio-Texeira, M., Polaina, J., 1999.Transformation of E. coli with DNA from Saccharomyces cerevisiae celllysates. Appl. Environ. Microbiol. 65, 5303–5306.

Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith,J.A., Struhl, K., 1989. Current Protocols in Molecular Biology, vols. 1–2.John Wiley & Sons, Inc., New York, USA.

Basílio, A.C.M., Araújo, P.R.L., Morais, J.O.F., Silva-Filho, E.A., de Morais Jr,M.A., Simões, D.A., 2008. Detection and identification of wild yeastcontaminants of the industrial fuel ethanol fermentation process.Curr. Microbiol. 56, 322–326.

Cunha, A.F., Missawa, S.K., Gomes, L.H., Reis, S.F., Pereira, G.A., 2006.Control by sugar of Saccharomyces cerevisiae flocculation for industrialethanol production. FEMS Yeast Res. 6, 280–287.

Ganley, A.R., Kobayashi, T., 2011. Monitoring the rate and dynamics ofconcerted evolution in the ribosomal DNA repeats of Saccharomycescerevisiae using experimental evolution. Mol. Biol. Evol. 28, 2883–2891.

Gestal, A.M., Liew, E.F., Coleman, N.V., 2011. Natural transformation withsynthetic gene cassettes: new tools for integron research andbiotechnology. Microbiology 157, 3349–3360.

Goffeau, A., Barrell, B.G., Bussey, H., Davis, R.W., Dujon, B., et al., 1996. Lifewith 6000 genes. Science 274, 563–567.

Gueldener, U., Heinisch, J., Koehler, G.J., Voss, D., Hegemann, J.H., 2002. Asecond set of loxP marker cassettes for Cre-mediated multiple geneknockouts in budding yeast. Nucleic Acids Res. 30, e23.

Lopes, T.S., Hakkaart, G.J., Koerts, B.L., Raué, H.A., Planta, R.J., 1991.Mechanism of high-copy-number integration of pMIRY-type vectorsinto the ribosomal DNA of Saccharomyces cerevisiae. Gene 105, 83–90.

Lopes, T.S., de Wijs, I.J., Steenhauer, S.I., Verbakel, J., Planta, R.J., 1996.Factors affecting the mitotic stability of high-copy-numberintegration into the ribosomal DNA of Saccharomyces cerevisiae.Yeast 12, 467–477.

Ostergaard, S., Olsson, L., Nielsen, J., 2000. Metabolic engineering ofSaccharomyces cerevisiae. Microbiol. Mol. Biol. Rev. 64, 34–50.

Pronk, J.T., 2002. Auxotrophic yeast strains in fundamental and appliedresearch. Appl. Environ. Microbiol. 68, 2095–2100.

Rao, B., Zhong, X., Wang, Y., Wu, Q., Jiang, Z., Ma, L., 2010. Efficient vectorsfor expression cloning of large numbers of PCR fragments in P.pastoris. Yeast 27, 285–292.

Reis, V.C., Nicola, A.M., de Souza Oliveira Neto, O., Batista, V.D., de Moraes,L.M., Torres, F.A., 2012. Genetic characterization and construction ofan auxotrophic strain of Saccharomyces cerevisiae JP1, a Brazilianindustrial yeast strain for bioethanol production. J. Ind. Microbiol.Biotechnol. 39, 1673–1683.

Romanos, M.A., Scorer, C.A., Clare, J.J., 1992. Foreign gene expression inyeast: a review. Yeast 8, 423–488.

Rubio-Teixeira, M., 2006. Endless versatility in the biotechnologicalapplications of Kluyveromyces LAC genes. Biotechnol. Adv. 24, 212–225.

Sambrook, J., Fristsch, E.F., Maniats, T., 1989. Molecular Cloning aLaboratory Manual, second ed. Cold Spring Harbor LaboratoryPress, Cold Spring Harbor, New York, USA.

Sassi, H.E., Bastajian, N., Kainth, P., Andrews, B.J., 2009. Reporter-basedsynthetic genetic array analysis: a functional genomics approach forinvestigating the cell cycle in Saccharomyces cerevisiae. Methods Mol.Biol. 548, 55–73.

Silva-Filho, E.A., Melo, H.F., Antunes, D.F., Dos Santos, S.K.B., Resende,M.A., Simões, D.A., de Morais Jr, M.A., 2005. Isolation by genetics andphysiological characteristics of a fuel-ethanol fermentativeSaccharomyces cerevisiae strain with potential for geneticmanipulation. J. Ind. Microbiol. Biotechnol. 32, 481–486.

Schuller, D., Casal, M., 2005. The use of genetically modifiedSaccharomyces cerevisiae strains in the wine industry. Appl.Microbiol. Biotechnol. 68, 292–304.

Williamson, M.P., Sutcliffe, M.J., 2010. Protein–protein interactions.Biochem. Soc. Trans. 38, 875–878.