BIOTECHNOLOGY TECHNIQUES Volwne 7 No.5 (May 1993) pp.325-328 Received a.~ revised 9th March

ANALYSIS OF X-GAL PLATES FOR QUANTITATIVE PLASMID STABILITY DETERMINATIONS IN SACCHAROMYCES CEREVISIAE Z. Wang

and N. A. Da Silva*

Biochemical Engineering Program University of California, Irvine Irvine. CA 92717 SUMMARY X-gal plates were evaluated for the measurement of plasmid stability in yeast cells carrying the Escherichia coli 1acZ gene. The influence of the medium, cloned gene number, and cell state on the accuracy of the results was determined by comparison with traditional replica plating. With few exceptions, X-gal plates provide a good quantitative one-step assay for plasmid stability in Saccharomyces cerevisiue expressing pgalactosidase. INTROIHJCTION Plasmid stability is a primary consideration in the cultivation of recombinant microorganisms, and is often monitored during fermentations of both bacteria and yeast. The plasmid-containing fraction of the population is usually determined by replica plating from non-selective to selective medium unless a calorimetric plate assay can be used to detect a plasmid product. A common example of such an assay is the use of X-gal plates for the detection of Escherichia coli containing the 1acZ gene (Sambrook et al., 1989). X-gal (5-brom~-4-cIiloro-3-indolyl-~-D-galactopyranoside)

is a substrate of the ZacZ gene product

(p-galactosidase), and turns blue when cleaved by this enzyme. The enzyme P-galactosidase is stable and easily assayed in liquid culture by ONPG tests. For this reason, the 1acZ gene has bcen widely employed as a model heterologous gene in research with other recombinant microorganisms, including the yeast Saccharomyces cerevisiae. For S. cerevisiae, X-gal plates have also been developed to indicate lucZ expression (Rose and Botstein, 1983: Guarente, 1983); color development is sensitive to the medium used and requires buffering to pH 7. The objective of this work was to evaluate X-gal plates for the quantitative determination of plasmid stability in S.

cerevisiae

carrying a plasmid-encoded E. coli JacZ gene. When the plasmid-containing

fraction is determined, general indication of JucZ expression in the cell population is insufficient; the color change must be accurately disccmible for every colony rcgardlcss of copy number or size. To assess the accuracy of the X-gal assay, values dctcrmined from the X-gal plates were compared with those cal:culatcd from standard replica plating. Three factors were considcrcd: the medium composition of the plate, the number of 1acZ gene copies, and the state of the ccl1 sample to bc evalualcd (i.e., growth phase). Factors important during plate preparation were also detcrmincd, and a general assessment made of X-gal plates for quantitative measurements.

325

MATERIALS

AND

Yeast

and Plasmid

Strains

METHODS

Saccharomyces cerevisiaestrainYM603 (MATa ade2-101ura3-52his31~~2-801 metregl-501) hasa regl-.5OZmutationwhich inhibitscataboliterepression by glucose(Hovlandet al., 1989).StrainD603-i (Srienc et al., 1986)is the a/a homozygousdiploid of strainYM603, but with one copy of the E. coli 1acZgeneintegratedinto the chromosomes. The 1acZgeneis controlledby an S. cerevisiaeGALZOCYCl hybrid promoter.PlasmidpRY121 (West et al., 1984;Yocum et al., 1984)is a shuttlevector containingthe 2~ origin andREP3site,a URA3gene(selectable marker),andthe E. coli 1acZgenefused to the yeastGALJ promoter.The GALJO-CYCZandGALI promotersareinducedby galactose;thelatter isapproximatelythreefoldto fourfold stronger(Da Silva andBailey, 1991). Nutrient

Media

SDCmediumcontainsnitrogenbasewithout ammoacids(6.7 g/L, Difco), casaminoacids(5 g/L), Difco), glucose(20 g/L), and adenine(100mg/L). YPD mediumcontainsyeastextract (10 g/L, Difco), peptone(20 g/L, Difco), and glucose(20 g/L). Solid mediaalsocontainsBacto-agar(20 giL, Difco). Theserecipesaremodificationsof thosegiven in Methodsin YeastGenetics(Shermanet al., 1986).The liquidmediawerebufferedto pH 5.5 with 0.05M citratebuffer. Preparation

of X-gal

plates

Four mediaof varying complexity were evaluated.Minimal platesare the standardX-gal plates describedby Roseand Botstein(1983) with minor modifications(Grant Bitter, Amgen), and contain: (NH4)2SO4(15 mM), FeS04 (2 PM), glucose(20 g/L), thiamine(0.4 mg/L), pyridoxine (0.4 mg/L), pantothenicacid (0.4 mg/L), inositol(2 mg/L), biotin (0.04 mg/L), andX-gal (40 mg/L). Adenine(100 mg/L), uracil (20 mg/L), L-histidine-HCl (20 mg/L), L-lysine-HCl (30 mg/L) and L-methionine(20 mgiL) were supplemented to complementthe auxotrophicmutationsin the host,andgalactose(20 g/L) wasaddedto induce1acZexpression.SDX platescontainnitrogenbasewithout aminoacids(6.7 g/L, Difco), glucose(20 g/L), galactose(20 g/L), X-gal (40 mg/L), andadenine,uracil, histidine,lysine and methionineat the sameconcentrations asin the minimalplates.SDCX platesaresimilarto SDX plates, but with casaminoacids(5 g/L, Difco) replacingthe histidine,lysine, andmethionine.YPDX platesare YPD platessupplemented with galactose(20 g/L) and X-gal (40 mg/L). All X-gal plateswereprepared with Bacto-agar(20 g/L, Difco). Beforeautoclaving,theglucose,galactose,salts,yeastextract,peptone,andagarin thevariousmedia weredissolvedin 0.1 M KH2PO4andthe pH adjustedto 7.0 by adding10M KOH. The vitamin stock solution(100X, filter-sterilized),nitrogenbasestocksolution(10X, filter-sterilized),andcasaminoacids, adenine,uracil,histidine,lysine,andmethioninestocksolutionswereaddedafter autoclaving.TheX-gal stock (dissolvedin N,N-dimetbylformamideat a concentrationof 20 mg/mL and storedat -2OOC)was addedimmediatelybeforepouringtheplates.All platescouldbe storedat 4OCfor 6 months. Experimental

Procedure

Seedcultures(5 mL) wereinoculatedfrom singlecoloniesandcultivatedat 30°Cfor 24 hours.Shake flaskscontaining100mL of mediumwereinoculatedfrom the seedcultures(1% v/v) andincubatedin a water bath shaker(New Brunswick,modelG76D) at 30°C and250 rpm. Optical densitiesof culture samples weremeasured at 600 nmwith a spectrophotometer (Varian,DMS 1OOS). Cell suspensions were dilutedto 600-800cells/ml, spreadon X-gal platesandYPD plates(0.2-0.25mL/plate),andincubatedat 30°C. Color developmentwasmonitoredfor the X-gal plates.Colonieson the YPD plateswerereplica plated to selectiveSDC plates.For eachmethod,a minimumof 400 colonieswere tested;thus, the maximumstatisticalerror (95% confidenceinterval) is 0.04and0.02 for plasmid-containing fractionsof 0.75 and0.95, respectively. RESULTS

AND

DISCUSSION

StrainYM603:pRY121wascultivatedin YPD andSDCliquid media,andthentransferredto thefour different X-gal plates.As expected,the cell growth rateandthe intensityof the blue color development varied with the richnessof the medium.AI1colonieshadappearedafter two dayson YPDX plates,three dayson SDCX plates,four dayson SDX plates,and5 dayson minimalplates.With the exceptionof the YPDX plates,the bluecolor developedasthecolony grew,andthe intensitywassufficientto differentiate P-galactosidase producingcells from plasmid-freecells approximatelythree days after initial colony

326

formation. The final intensity of color, however, decreased with increasing medium richness. The rich complex YPDX plate showed very poor color development and could not be used as an indicator of /3galactosidase expression. The minimal, SDX, and SDCX plates were all satisfactory for distinguishing plasmid-free and p&mid-containing

cells for this strain; a quantitative comparison of the three X-gal

plates with replica plating is given below. To test the sensitivity of the X-gal plates, a suspension of strain D603-i was spread on the minimal, SDX, and SDCX plates. D603-i contains one copy of the ZacZ gene (under the control of a hybrid GALIO-CYCl

promoter) stably integrated into the chromosomes (Srienc er al., 1986). On all

three plates, 100% of the colonies showed sufficient color change to be definitively scored as lucZcontaining. Therefore, the three X-gal plates are sensitive enough to detect cells harboring only one copy of a GAL regulated 1acZ gene. The accuracyof plasmidstability determinations obtainedwith the minimal,SDX, andSDCX plates was assessed by comparisonwith standardreplica plating. YM603:pRYl21 was cultivated in nonselective YPD medium for 50 hours of batch culture; the plasmid-containingfraction was then determined.The fraction of the populationretainingthe plasmidwasmeasured at 0.73 with the minimal plates,0.74 with the SDX plates,and 0.75 with the SDCX plates.Thesevaluesareconsistentwith the valueof 0.74 obtainedby replicaplatingfrom YPD platesto selectiveSDCplates. To further test the X-gal plates,plasmidstabilitieswere determinedduring the courseof a batch fermentation.Strain YM603:pRYl21 wascultivatedin SDCmedium.In this selectivemedium,plasmidfree cellswill still arise(e.g.,by segregation error),but will be unableto sustaintheir growth. Therefore, the health/viability of the plasmid-freecellstransferredto the X-gal plateswill be lower thanif the cells hadbeencultivated in a non-selectivemedium(asabove).To determinewhetherthe composition(i.e., richness)of the X-gal plate influencesthe ability of thesecells to grow, and thereforethe plasmid stability measurement, cell suspensions were transferredto the threeX-gal platesandYPD plates.The coloniesfrom the YPD plateswerethenreplicaplatedlo SDCplates.Plasmidstabilitiesweredetermined for samplestaken at the time of inoculation (0 hour), at the end of exponentialphase(17 hours),in stationaryphase(37 hours), and at the end of batch culture (60 hours).The resultsare shownin the following table. TableI. Plasmidstability for strain YM603:pRY 121during batch culture in selectiveSDC medium. Comparison of measurements via X-gal platesandtraditionalreplica-plating.

Sample1 (0 hour)

Fractionof Plasmid-Containing Cells Sample2 Sample3 (17 hours) (37 hours)

Sample4 (60 hours)

Minimal X-gal Plate

0.95

0.93

0.94

0.94

SDX Plate

0.95

0.94

0.94

0.94

SDCX Plate

0.94

0.95

0.95

0.9s

ReplicaPlating

0.95

0.94

0.96

0.94

327

The data in Table I indicate that all three X-gal plates provide accurate measurements of plasmid stability (within statistical error) regardless of the growth phase of the cells. In addition to saving time, another advantage of X-gal plates is that more cells cau be transferred from liquid culture (150-200 per plate). Overlapping colonies can be clearly differentiated, while those colonies cannot be easily transferred to selective plates via traditional replica plating. While all thee plates were adequate for the strain tested in this work, a few notes on plate preparation should be added. All plates must be buffered to pH 7.0 for sufficient color development. To ensure precise measuremeuts, it is also important that good viability be attained on the poorer plates (minimal aud SDX). The success of the minimal plates was very sensitive to the method of plate preparation. For example, strain YM603:pRY121 grew very slowly (or not at all) on the minimal plate when BiTek agar (Difco) was used. Furthermore, the cells grew poorly on the minimal and SDX plates when the glucose aud galactose stock solutions were addedafter autoclaving even when Bacto-agar (Diico) was utilized. CONCLUSIONS Plasmid stabilities in cultures of S. cerevisiae carrying GAL-regulated E.coli lucZ genes were accurately determined by a one-step method using various yeast media supplemented with X-gal and buffered to pH 7. The results from minimal, SDX, aud SDCX plates were consistent with those from staudard replica plating. Rich YPDX plates, however, showed weak color development and could not be used. When the minimal or SDX plates are utilized, cell viability should be tested in advance; this is especially important for the miuimal plates. Although the X-gal plate assay is more expensive than conventional replica plating, it is an alternate, time-saving, aud reliable method for determining plasmid stabilities in yeast expressing p-galactosidase. REFERENCES Da Silva, N.A., and Bailey, J.E. (1991). Biotechnol. Bioeng. 37,318-324. Guareute, L. (1983). Yeast Promoters aud ZucZ Fusions Designed to Study Expression of Cloned Genes in Yeast. In: Methods in Enzymology, R. Wu, L. Grossman, and K. Moldave, eds. vol. 101. pp. 181-191, New York: Academic Press. Hovlaud, P., Flick, J., Johnston, M., and Sclafani, R.A. (1989). Gene. 83, 57-64. Rose, M., aud Botsteiu, D. (1983). Construction and Use of Gene Fusions to lucZ (B-Galactosidase) that are Expressed in Yeast. In: Methods in Enzymology, R. Wu, L. Grossman, and K. Moldave, eds. vol. 101. pp. 167-180, New York: Academic Press. Sambrook J., Fritsch, E.F., and Mauiatis, T. (1989). Molecular Cloning. 2nd Ed. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. Sherman, F., Fink, G.R., and Hicks, J. (1986). Methods in Yeast Genetics, Cold Spring Harbor, New York: Cold Spriug Harbor Laboratory Press. Srienc, F., Campbell, J.L., and Bailey, J.E. (1986). Cytometry. 7, 132-141. West Jr., R.W., Yocum, R.R., and Ptashne, M. (1984). Mol. Cell. Biol. 4, 2467-2478. Yocnm, R.R., Hauley, S., West Jr., R.W., aud Ptashne, M. (1984). Mol. Cell. Biol. 4, 19851998.

328