regulation of peroxisomal proteins and organelle proliferation by multiple carbon sources in the...

Yeast 14, 1175–1187 (1998)

Regulation of Peroxisomal Proteins and OrganelleProliferation by Multiple Carbon Sources in theMethylotrophic Yeast, Candida boidinii

YASUYOSHI SAKAI*, HIROYA YURIMOTO, HIDEAKI MATSUO AND NOBUO KATO

Division of Applied Life Sciences, Graduate School of Agriculture, Kyoto University, Kitashirakawa-Oiwake,Sakyo-ku, Kyoto 606–8502, Japan

A methylotrophic yeast, Candida boidinii, was grown on various combinations of peroxisome-inducing carbonsource(s) (PIC(s)), i.e. methanol, oleate and -alanine, and the regulation of peroxisomal proteins (both matrix andmembrane ones) and organelle proliferation were studied. This regulation was followed (1) at the protein or enzymelevel by means of the peroxisomal enzyme activity and Western analysis; (2) at the mRNA level by Northernanalysis; and (3) at the organelle level by direct observation of peroxisomes under a fluorescent microscope.Peroxisomal proliferation was followed in vivo by using a C. boidinii strain producing a green fluorescent proteinhaving peroxisomal targeting signal 1. When multiple PICs were used for cell growth, C. boidinii induced specificperoxisomal proteins characteristic of all PIC(s) present in the medium, responding to all PIC(s) simultaneously.Thus, these PICs were considered to induce peroxisomal proliferation independently and not to repress peroxisomesinduced by other PICs. Next, the sensitivity of the peroxisomal induction to glucose repression was studied. Whilethe peroxisomal induction by methanol or oleate was completely repressed by glucose, the -alanine-inducedactivities of -amino acid oxidase and catalase, Pmp47, and the organelle proliferation were not. These resultsindicate that peroxisomal proliferation in yeasts is not necessarily sensitive to glucose repression. Lastly, thisregulation was shown to occur at the mRNA level. ? 1998 John Wiley & Sons, Ltd.

— Candida boidinii; peroxisome; peroxisomal proliferation; peroxisomal membrane proteins; -aminoacid oxidase; green fluorescent protein

*Correspondence to: Yasuyoshi Sakai, Division of Applied LifeSciences, Graduate School of Agriculture, Kyoto University,Kitashirakawa-Oiwake, Sakyo-ku, Kyoto 606–8502, Japan.Tel: (+81) 75 753 6455; fax: (+81) 75 753 6385; e-mail:[email protected]/grant sponsor: Ministry of Education, Science, Sports

INTRODUCTION

Peroxisomes are single-membrane-bound organ-elles that are ubiquitously found in eukaryoticcells. In mammalian cells, peroxisomes areinvolved in various metabolic processes, such asthe â-oxidation of fatty acids, cholesterol syn-thesis, and -amino acid metabolism (van denBosch et al., 1992). Therefore, peroxisomal pro-liferation and this metabolism should be strictlycontrolled at various levels of regulation corre-sponding to cellular demand. In the last decade,proteins involved in peroxisome biogenesis have

and Culture of Japan

CCC 0749–503X/98/131175–13 $17.50? 1998 John Wiley & Sons, Ltd.

been identified in various organisms; recently theirnomenclature has been unified, and they have beendesignated as peroxines (Distel et al., 1996). Thesequence similarity of several PEX genes in yeastsand mammalian cells indicates that the basicmolecular mechanism for peroxisome biogenesishas been conserved during evolution.

In yeasts, peroxisomes generally develop inresponse to environmental stimuli. For example, amethylotrophic yeast, Candida boidinii, can grownot only on methanol but also on oleate or-alanine as a single carbon source concomitantwith peroxisomal proliferation (Goodman et al.,1990). Peroxisomes play an indispensable role inthis growth of cells on these peroxisome-inducingcarbon sources (PICs), since C. boidinii pexmutants showed impaired growth on all of thesePICs (Sakai et al., 1995a). However, the protein
composition of peroxisomes depends on the PIC in
Received 24 April 1998Accepted 24 May 1998

1176 . .

the medium, reflecting the fact that peroxisomalmetabolism differs among three PICs. So far, theprotein composition, and regulation of peroxiso-mal matrix and membrane proteins in cells grownon a single PIC have been studied in C. boidiniiATCC32195 (Goodman et al., 1990). However, inthe strain used, all the analysed peroxisomal mem-brane proteins (PMPs), Pmp20, Pmp47 andCbPex11p (Pmp30), are encoded by two differentloci (Garrard and Goodman, 1989; McCammonet al., 1990; Moreno et al., 1994).

In this study, we analysed the regulation ofperoxisomal proteins and organelle proliferation inC. boidinii strain S2 grown on multiple PICs tostudy the induction of peroxisomal proteins andorganelle proliferation (Sakai et al., 1995a, 1996a).The strain used was a haploid strain of C. boidiniiwith which a heterologous protein can be pro-duced (Sakai et al., 1995b, 1996b). To followperoxisomal proliferation in vivo, a C. boidiniitransformant producing a green fluorescent pro-tein (GFP) tagged with peroxisomal targetingsignal 1 (PTS1), -AKL, at the carboxyl terminal,was observed under a fluorescence microscope(Monosov et al., 1996). The present study wasconducted to determine: (1) whether there is any‘priority rule’ in the utilization of multiple PICs(i.e. do cells synthesize peroxisomal proteins char-acteristic of all PICs present in the medium orspecific PICs?); (2) whether peroxisomal prolifer-ation is sensitive to glucose repression for all PICs;and (3) whether the observed regulation occurs atthe mRNA level.

MATERIALS AND METHODS

Strains, media and cultivationC. boidinii S2 was used in all experiments (Tani

et al., 1985a). The organism was grown on thesynthetic MI medium described previously (Sakaiet al., 1991). As carbon sources, 2% glucose (w/v),1% methanol (v/v), 0·5% oleate (v/v) and 0·6%-alanine (w/v) were used. Tween 80 was added tothe oleate-medium at a concentration of 0·05%(v/v). When -alanine was used as both nitrogenand carbon sources, NH4Cl was omitted from theMI medium. The initial pH of the medium wasadjusted to 6·0. Cultivation was aerobic at 28)Cwith reciprocal shaking, and growth was followedby measuring the optical density at 610 nm.

C. boidinii strain TK62 (ura3; Sakai et al., 1991)and strain pex5Ä (ura3) were used as hosts for

? 1998 John Wiley & Sons, Ltd.

transformation. The strain pex5Ä was derivedfrom strain TK62 as PEX5 gene disruptant (Y.Sakai, unpublished results). Escherichia coli JM109(Sambrook et al., 1989) was used for plasmidpropagation.

Enzyme assaysCells were harvested by centrifugation at 500#g

for 10 min at 4)C, washed twice with ice-colddistilled water, resuspended in 0·1 -potassiumphosphate buffer, pH 7·5, and then subjected todisruption with a KUBOTA Insonator Model201M (2 MHz for 35 min). The cell debris wasremoved by centrifugation at 16,000#g for 5 minat 4)C. The resultant supernatant was immediatelysubjected to enzyme activity assays. The activitiesof catalase (CTA), alcohol oxidase (AOD),-amino acid oxidase (DAO) and acyl-CoA oxi-dase (ACO) were assayed by the methods ofBergmeyer (1955), Tani et al. (1985b), Goodmanet al. (1990) and Shimizu et al. (1979), respectively.Protein was determined by the method ofBradford (1976) with a protein assay kit (BioRadLaboratories, Hercules, CA, USA). Bovine serumalbumin was used as the standard.

SDS–polyacrylamide gel electrophoresis andWestern analysis

Standard 9% Laemmli gels (Laemmli, 1970),with separating gels of pH 9·2, were employed. Acell-free extract containing 50 ìg protein wasloaded per lane. Western analysis was performedas described by Towbin et al. (1979) using anAmersham ECL detection kit (Arlington Heights,IL, USA). The VA9 monoclonal anti-Pmp20antibody, IVA7 monoclonal anti-Pmp47 anti-body, and G358 polyclonal anti-AOD antibodywere kindly provided by Dr J. M. Goodman(University of Texas, Southwestern MedicalCenter at Dallas).

Northern analysisTotal RNAs were extracted from cells using an

ISOGEN RNA extraction kit (Nippon Gene Co.,Ltd, Tokyo, Japan), and electrophoresed on a1·1% agarose gel made with 20 m-MOPS buffercontaining 1 m-EDTA and 2·2 -formaldehyde.Fractionated total RNA was blotted onto mem-brane filter (Gene Screen Plus; BiotechnologySystems, NEN Research Products, Boston, MA,

Yeast 14, 1175–1187 (1998)

1177

USA). Hybridization was performed under highlystringent conditions as described previously(Sakai et al., 1993). Labeling was performed by therandom primer extension method of Feinberg andVogelstein (1983). The 32P-labeled probes were a0·7-kb BglII-SalI-fragment derived from pMOX33harboring the coding region of AOD1 (Sakai andTani, 1992), a 0·4-kb StyI-HincII fragment harbor-ing C. boidinii CTA1 (catalase; Y. Sakai, unpub-lished results), a 1·9-kb PstI fragment harboring C.boidinii PMP20 (Y. Sakai, unpublished results), a1·6-kb HincII-HindIII-fragment derived frompMP471 harboring the coding region of PMP47(Sakai et al., 1996a), a 0·4-kb StyI-HindIII frag-ment of pDA7 coding for DAO (H. Yurimoto,unpublished results), a 0·7-kb fragment coding forACO (H. Yurimoto, unpublished results) and a0·9-kb ClaI-HindIII fragment harboring C. boidiniiACT1 DNA (coding for actin) (Sakai et al.,1996a).

Construction of C. boidinii GFP-AKLThe GFP gene expression cassette consisted of

the C. boidinii actin promoter, the modified GFPgene and the C. boidinii actin terminator. Themodified GFP genes were constructed through theuse of PCR. The GFP gene was tagged withGly-Gly (STOP) or Gly-Gly-Ala-Lys-Leu (AKL)at the carboxy terminus. These modified GFPgenes were ligated into pACT1 that containingpBR322, actin promoter, actin terminator andURA3. The resulting plasmids (pGFP-STOP andpGFP-AKL) were integrated into the chromo-somal DNA of C. boidinii TK62 or C. boidiniipex5Ä by the modified lithium acetate method(Sakai et al., 1993). Transformants were termedGFP-AKL/wt, GFP-STOP/wt, GFP-AKL/pex5Äand GFP-STOP/pex5Ä.

Fluorescence microscopyC. boidinii strain GFP-AKL was used to observe

peroxisomal proliferation in vivo. The cell suspen-sion was placed on a microscope slide and exam-ined using the FITC channel of a Axioplan 2fluorescence microscope (Zeiss, Oberkochen,Germany) equipped with a Plan-NEOFLUAR100#/1·30 (oil) objective and Nomarski attach-ments. Cells were photographed on Fuji NEO-PAN SS 135 black-and-white film. The number ofperoxisomes per cell was determined for 40 to 80cells in a randomly selected field.


RESULTS

Regulation of peroxisomal proteins by PIC(s)

Regulation by a single PIC In C. boidinii strainS2, three PMPs (Pmp47, CbPex11p and Pmp20)and at least seven other proteins are encoded by asingle locus (Sakai et al., 1995c, 1996a). At first, westudied the regulation of these PMPs and severalperoxisomal enzymes by a single PIC in C. boidiniiS2.

C. boidinii S2 was grown on glucose or aperoxisome-inducing carbon source (PIC), i.e.methanol, -alanine, or oleate, and then peroxiso-mal enzyme activities were determined (Figure1a–d, columns G, M, D and O). The regulation ofPMPs was also followed by Western analysis(Figure 2). CTA activity and Pmp47 were signifi-cantly induced by all PICs when compared tothose in glucose-grown cells. (The amount ofPmp47 in oleate- and methanol-grown cells washigher than that in -alanine-grown cells.) In con-trast to CTA activity and Pmp47, other proteinswere induced by a specific PIC: the activities ofAOD and Pmp20 were induced by methanol, theactivity of ACO was induced by oleate, and theactivity of DAO was induced by -alanine. When-alanine was used as the single carbon and nitro-gen source (Figure 1c, D ("NH4Cl)), DAO ac-tivity was about twice as high as when it was usedas the single carbon source (Figure 1c, D).

Next, Northern analysis was performed to con-firm that the observed regulation occurred at themRNA level. As shown in Figure 3, mRNAs ofCTA1 and PMP47 were detected in cells grown onall PICs. In contrast, mRNAs of AOD1 andPMP20 were observed only in methanol-growncells and mRNA of ACO was detected only inoleate-grown cells (Figure 5b, lanes G, M, D andO). These mRNAs were not detected in glucose-grown cells. All hybridizing bands were detected atthe expected sizes (data not shown), and ACT1mRNA was almost constant. These band intensi-ties precisely reflected the regulatory profile esti-mated from the enzyme activities and the results ofWestern analysis.

From the results obtained, a haploid strain of C.boidinii, strain S2, was shown to have a similarprofile of peroxisomal protein regulation by asingle PIC to that of C. boidinii ATCC 32195(Goodman et al., 1990; Moreno et al., 1994), i.e.peroxisomal proteins could be classified into (1)metabolism-specific proteins (AOD and Pmp20

Yeast 14, 1175–1187 (1998)

1178 . .

for methanol, ACO for oleate, and DAO for-alanine), and (2) metabolism non-specificproteins (CTA and Pmp47).

Figure 1. Peroxisomal enzyme activities of C. boidinii S2 grown on a single PIC or variouscombinations of PICs. Cells were grown to the early-log phase on the indicated carbon source(s), andthen enzyme activities (a, catalase (CTA); b, alcohol oxidase (AOD); c, -amino acid oxidase (DAO);d, acyl-CoA oxidase (ACO)) were measured as described in Materials and Methods. The enzymeactivities are expressed as units/mg protein in the crude cell-free extracts. The results from threeindependent experiments are given. The carbon sources were: G, glucose; M, methanol; D, -alanine;O, oleate.

Regulation by multiple PICs To determinewhether or not there is a priority rule among PICsfor peroxisomal metabolism, cells were grown onvarious combinations of PICs, i.e. methanol (M),-alanine (D) and oleate (O). Enzyme activities(Figure 1) and induced PMPs (Figure 2b) wereanalysed.

The activities of a peroxisomal marker enzyme,CTA, and Pmp47, were induced when cells weregrown on any combination of PICs (Figure 1a forCTA; Figure 2b for Pmp47). Methanol-specificproteins, AOD and Pmp20, were induced whenmethanol was present in the medium (Figure 1b;Figure 2b), and the induction of these methanol-


specific proteins was not affected by the presenceof other PICs, i.e. oleate and/or -alanine. In thecase of M+O and M+D+O (Figure 2b), the AODactivities seemed to be reduced by the presence ofoleate. However, in this case, the peroxisomemetabolism was diverse, and consequently, thespecific activities of each enzyme decreased. Infact, compared with the repressed level by glucoseas described below, these activities were still at theinduced level.

Similarly, the activities of oleate-inducibleenzymes, ACO was induced when oleate waspresent in the medium (Figure 1d), and the induc-tion was not repressed in the presence of methanoland/or -alanine. DAO activity was induced when-alanine was present in the medium (Figure 1c),and the induction was not repressed by the coex-istence of methanol and/or oleate. When -alanine

Yeast 14, 1175–1187 (1998)

1179

was used as both the carbon and nitrogen source(Figure 1c, denoted as ("NH4Cl)), DAO activitywas higher than when -alanine was used as thecarbon source and NH4Cl was used as the nitrogensource.

Thus, the three PICs induced peroxisomal pro-teins independently, and there was no priority ruleamong the PICs, i.e. C. boidinii seems to utilizeany combination of PICs simultaneously whenmultiple PICs are present in the medium.

Sensitivity to glucose In yeasts, methanol- andoleate-inducible peroxisomal enzymes and peroxi-somes have been reported to be sensitive to glucoserepression. Furthermore, glucose triggers anautophagic process to degrade methanol- andoleate-inducible peroxisomes (Chiang et al., 1996;Tuttle and Dunn, 1995; Tuttle et al., 1993).

Figure 2. Western blots of crude extracts of cells of C. boidinii S2grown on (a) a single PIC or (b) various combinations of PICs. Cellswere grown on G, glucose; M, methanol; D, -alanine; or O, oleate,or on combinations of them. Each lane was loaded with 50 ìgprotein.


First, the effect of glucose on the formation ofperoxisomal proteins was studied by growing cellson glucose-containing medium in combinationwith various PICs (Figure 4 a–b). When glucosewas present in the medium, the induction of AOD-and ACO-activities was repressed in all casestested (data not shown). On the contrary, DAOactivity was highly induced in all media containing-alanine, i.e. G+D, G+M+D, G+D+O andG+M+D+O (Figure 4b). And CTA activities in-alanine-containing media (G+D, G+M+D,G+D+O, or G+M+D+O) was higher than thatin media without -alanine (G+M,, G+O,or G+M+O) (Figure 4a). Thus, methanol-and oleate-induced peroxisomal enzymes wererepressed by glucose, but the induction of DAO-and CTA-activities was not repressed in the pres-ence of glucose. When glucose was present

Yeast 14, 1175–1187 (1998)

1180 . .

Figure 3. Northern analysis of peroxisomal matrix and mem-brane proteins in C. boidinii S2. Total RNAs (20 ìg) extractedfrom cells grown on M, methanol; O, oleate; D, -alanine; orG, glucose, as a single carbon source were loaded on each lane,and then probed with the indicated DNA fragment as describedin Materials and Methods. The C. boidinii ACT1 DNA codingfor actin was used as a control for constitutive expression.

together with various PICs in the medium, theAOD and Pmp20 proteins were not detected onWestern analysis (Figure 4c). However, in spite ofthe presence of glucose, Pmp47 could be detectedwhen -alanine was present as a PIC (Figure 4c,lanes G+D and G+M+D). Thus, although AODand Pmp20, which were massively induced bymethanol, were completely repressed by glucose,-alanine-induced Pmp47 was not.

To confirm that the DAO1 gene (coding forDAO) is insensitive to glucose repression at themRNA level, cells were grown on G, M, D, G+M,G+D or M+D as carbon source(s), and Northernblot analysis was performed using DAO1- orAOD1-DNA as probe. As shown in Figure 5a, theDAO1 mRNA was detected not only in D andM+D but also in G+D. In contrast, the AOD1mRNA was detected in M and M+D, but not inG+M. Similarly, the ACO1 (coding for ACO)mRNAs were detected in O, M+O and D+O


(Figure 5b). These results suggested that while theexpression ofAOD1 and ACO1 are repressed byglucose, that of the DAO1 is not, and that theirexpression is regulated at the mRNA level.

Peroxisomal proliferation by C. boidinii strainGFP-AKL

The fact that the induction of three peroxisomalproteins (DAO, CTA and Pmp47) by -alaninewas not repressed by glucose strongly suggestedthat the proliferation of -alanine-induced peroxi-somes is also insensitive to glucose repression.To confirm this, we introduced a GFP-AKL-expression plasmid into C. boidinii to visualizeperoxisomes and to follow peroxisomal prolifer-ation in vivo (strain GFP-AKL/wt). GFP-AKL isGFP tagged with -Ala-Lys-Leu (-AKL) at theC-terminal. It was constitutively expressed in C.boidinii S2 under the C. boidinii ACT1 (actin)promoter. The C-terminal three amino acid resi-dues, -AKL, comprise a typical motif of PTS1(Gould et al., 1990; Marshall et al., 1996), and issufficient for peroxisomal transport in C. boidinii.This GFP analysis enabled us to identify peroxi-somes even when they were low in number, as in-alanine- or glucose-grown cells.

We constructed four GFP-expressing C. boidiniistrains (GFP-AKL/wt, GFP-STOP/wt, GFP-AKL/pex5Ä and GFP-STOP/pex5Ä). These fourstrains induced on methanol-containing mediumwere observed under the fluorescent microscope.GFP-AKL/wt cells contained intrinsic green fluor-escent punctate structures (Figure 6a). In contrast,cells of GFP-AKL/pex5Ä, GFP-STOP/wt, andGFP-STOP/pex5Ä did not have any punctatestructures, but green fluorescence was diffused inthe whole cytosol (Figure 6 b–d). The punctatestructures we observed in GFP-AKL/wt cells wereidentified as peroxisomes (see Discussion).

When cells were grown on glucose as a singlecarbon source, the cells contained one to two(1·42&0·0019 (S. D.)) very small peroxisomes(Figure 7a). On the other hand, cells grown onPICs contained peroxisomes corresponding toeach carbon source; methanol-grown cells hadthree to six large peroxisomes in a cluster (Figure6a), oleate-grown cells had 8 to 12 (10·3&0·85)small peroxisomes (Figure 7d), and -alanine-grown cells had two to three (2·28&0·17) smallperoxisomes (Figure 7b). When -alanine was usedas both the carbon and nitrogen source, cells hadtwo to five (3·17&0·13) small peroxisomes (Figure7c).

Yeast 14, 1175–1187 (1998)

1181

Figure 4. (a,b) Enzyme activities and (c) Western analysis of C. boidinii S2 grown on variouscombinations of PICs and glucose. Cells were grown to the early-log phase on the indicated carbonsources, and then enzyme activities were measured as described in Materials and Methods. (a) Catalase;(b) -amino acid oxidase. The enzyme activities are expressed as units/mg protein in the crude cell-freeextracts. The results from three independent experiments are given. The carbon sources were: G,glucose; M, methanol; D, -alanine; O, oleate. (c) Each lane was loaded with 50 ìg protein. Westernanalysis was performed as described in Materials and Methods.

When cells were grown on G+M or G+O, thecells contained one or two very small peroxisomes(1·41&0·0082, 1·43&0·058, respectively) (Figure8a,d), which is similar to the number observed inglucose-grown cells (Figure 7a). But cells grown onG+D contained two to three small peroxisomes(2·03&0·042) (Figure 8b), which is similar to thenumber in -alanine-grown cells (2·28&0·17)(Figure 7b). In addition, cells grown on G+D("NH4Cl) contained two to five small peroxi-somes (2·27&0·12) (Figure 8c), which is similar tothe number in -alanine ("NH4Cl) (2·28&0·17)(Figure 7c). Thus, glucose did not inhibit theproliferation of -alanine-induced peroxisomes.


When cells were grown on M+O, the cells con-tained five to ten peroxisomes (8·2&0·30) (Figure8e). Although these cells contained methanol-inducible peroxisomal proteins, there was noperoxisomal cluster typical of methanol-growncells, and the number of peroxisomes was moresimilar to that in oleate-grown cells (10·3&0·85)(Figure 7d).

DISCUSSION

In this study, we examined the regulatoryprofile of peroxisomal proteins and peroxisomal

Yeast 14, 1175–1187 (1998)

1182 . .

Figure 5. Induction of DAO1 (-amino acid oxidase) was insensitive to glucoserepression in C. boidinii S2. Total RNAs (20 ìg) were extracted from cells grown onthe indicated carbon sources, loaded on each lane, and then probed with AOD1-,DAO1-, CTA1- (a) or ACO1- (b) DNA. G, glucose; M, methanol; D, -alanine. TheC. boidinii ACT1 DNA coding for actin was used as a control for constitutiveexpression.

proliferation in C. boidinii strain S2 when cellswere grown on various combinations of metaboli-cally distinguishable PICs (methanol, oleate and-alanine), and glucose, a potent repressor ofperoxisomal proliferation.


Waterham et al. (1992) reported that C. boidiniiperoxisomes could have two metabolicallydifferent enzymes, i.e. AOD and ACO, in onecompartment, when cells are grown in oleate-methanol-limited continuous cultures or when

Yeast 14, 1175–1187 (1998)

Fi a, GFP-AKL/wt; b, GFP-STOP/wt; c, GFP-AKL/pex5Ä;d, fluorescent microscope. Pictures were obtained usingN

1183

?1998

JohnW

iley&

Sons,L

td.Y

east14,

1175–1187(1998)

gure 6. Observation of peroxisome proliferation in vivo by use of GFP. C. boidinii strains (GFP-STOP/pex5Ä) were placed on methanol-containing medium and observed under a

omarski (upper) and fluorescence optics (lower).

F ii strain GFP-AKL was grown on various carbon sources(a uorescent microscopic observation. Pictures were obtainedu

1184.

.

?1998

JohnW

iley&

Sons,L

td.Y

east14,

1175–1187(1998)

igure 7. Peroxisomes of cells grown on various carbon sources labeled with GFP. C. boidin, glucose; b, -alanine; c, -alanine ("NH4Cl); d, oleate). Cell suspensions were subjected to fl

sing Nomarski (upper) and fluorescence optics (lower).

Figure 8. Peroxisomes of cells grown on various combinations of carbon sources labeled with GFP. C. boidiniistrain GFP-AKL was grown on various carbon sources (a, glucose+methanol; b, glucose+-alanine; c, glucose+-alanine ("NH4Cl); d, glucose+oleate; e, methanol+oleate). Cell suspensions were subjected to fluorescentmicroscopic observation. Pictures were obtained using Nomarski (upper) and fluorescence optics (lower).

1186 . .

methanol is added at the stationary phase ofoleate-grown cells. However, under their condi-tions, the concentration of oleate in the mediumwas supposedly insufficient for any repressionof methanol-inducible proteins, and it was stillunclear as to the presence of a priority rule amongPICs. In this study, cells were collected at theearly-log phase to ensure that all carbon sourceswere still present in the medium. Our results showthat there is no priority rule among PICs as to theinduction of peroxisomes, and that C. boidinii isable to utilize multiple PICs simultaneously.

While peroxisomal induction by methanol oroleate was completely repressed by the coexistenceof glucose, that by -alanine was not. This wasproved in terms of (1) induction of Pmp47, DAO-and CTA-activities, (2) DAO1 expression, and (3)organelle proliferation. We followed peroxisomalproliferation using a C. boidinii strain producingGFP-AKL. The punctate structures we observedwere identified as peroxisomes from the followingobservations: (1) wild-type GFP (without PTS1)did not show a punctate structure but rather acytosolic diffusion pattern; (2) GFP-AKL in the C.boidinii pex5 mutant (defective in the PTS1 recep-tor) gave a cytosolic diffusion pattern; (3) GFP-punctate structures proliferated when cells wereshifted from glucose to PICs; (4) their morphologyand number on PICs were consistent with previousobservations on electron microscopy; and (5)oleate-induced cells of the C. boidinii pex11 mutantcontained large peroxisomes (Sakai et al., 1995c;Sakai et al., manuscript in preparation). Therefore,peroxisomal proliferation followed by GFP-AKLwas not an artifact but was indeed a reflection ofperoxisomal proliferation in C. boidinii cells. Sinceperoxisomal assembly and proliferation requiremany peroxine gene products, and matrix andmembrane proteins, their regulation needs to beproperly coordinated. So far, peroxisomal prolifer-ation in yeast has been believed to be sensitive toglucose repression. However, this study showedthat all peroxisomal proliferation may not neces-sarily be sensitive to glucose repression, at leastin the case of -alanine-induced peroxisomes inC. boidinii.

It is also suggested herein that the synthesis ofperoxisomal matrix enzymes and PMPs is mainlyregulated at the mRNA level by PICs and glucose.We have cloned these peroxisomal genes, and theirregulation is being studied at the transcriptionallevel by placing the reporter genes under theirpromoters. Through such studies we hope (1) to


clarify the unique regulatory profile of -alanine-induced peroxisomes; (2) to determine the strengthof each promoter; and (3) to answer the questionof whether or not the precedence of PMP synthesisto matrix enzymes is indeed controlled at thetranscriptional level (Sakai et al., 1996a; Veenhuisand Goodman, 1990).

ACKNOWLEDGEMENTS

We acknowledge Prof. J. M. Goodman, Universityof Texas, Southwestern Medical Center at Dallas,for providing antibodies and other reagents. Thiswork was partly supported by a research grantfrom the Ministry of Education, Science, Sportsand Culture of Japan, to Y.S.

REFERENCES

Bergmeyer, H. U. (1955). Zur Messung von Katalase-Aktivitaten. Biochem. Z. 327, 255–258.

Bradford, M. M. (1976). A rapid and sensitive methodfor the quantification of microgram quantities ofprotein utilizing the principle of protein-dye binding.Anal. Biochem. 72, 248–254.

Chiang, H. L., Schekman, R. and Hamamoto, S. (1996).Selective uptake of cytosolic, peroxisomal, andplasma membrane proteins into the yeast lysosome fordegradation. J. Biol. Chem. 271, 9934–9941.

Distel, B., Erdmann, R., Gould, S. J., et al. (1996). Aunified nomenclature for peroxisome biogenesisfactors. J. Cell Biol. 135, 1–3.

Feinberg, A. P. and Vogelstein, B. (1983). A techniquefor radiolabeling DNA restriction endonuclease frag-ments to high specific activity. Anal. Biochem. 132,6–13.

Garrard, L. J. and Goodman, J. M. (1989). Two genesencode the major membrane-associated protein ofmethanol-induced peroxisomes from Candida boidinii.J. Biol. Chem. 264, 13929–13937.

Goodman, J. M., Trapp, S. B., Hwang, H. andVeenhuis, M. (1990). Peroxisomes induced in Candidaboidinii by methanol, oleic acid and -alanine vary inmetabolic function but share common integralmembrane proteins. J. Cell Sci. 97, 193–204.

Gould, S. J., Keller, G. A., Schneider, M., et al. (1990).Peroxisomal protein import is conserved betweenyeast, plants, insects and mammals. EMBO J. 9,85–90.

Laemmli, U. K. (1970). Cleavage of structural proteinsduring the assembly of the head of bacteriophage T4.Nature (Lond.) 227, 680–685.

Marshall, P. A., Dyer, J. M., Quick, M. E. andGoodman, J. M. (1996). Redox-sensitive homodimeri-zation of Pex11p: a proposed mechanism to regulate
peroxisomal division. J. Cell. Biol. 135, 123–137.
Yeast 14, 1175–1187 (1998)

1187

McCammon, M. T., Dowds, C. A., Orth, K., Moomaw,C. R., Slaughter, C. A. and Goodman, J. M. (1990).Sorting of peroxisomal membrane protein PMP47from Candida boidinii into peroxisomal membranes ofSaccharomyces cerevisiae. J. Biol. Chem. 265, 20098–20105.

Monosov, E., Wenzel, T. J., Luers, G. H., Heyman,J. A. and Subramani, S. (1996). Labeling of peroxi-somes with green fluorescent protein in living P.pastoris cells. J. Histochem. Cytochem. 44, 581–589.

Moreno, M., Lark, R., Campbell, K. L. and Goodman,J. M. (1994). The peroxisomal membrane proteins ofCandida boidinii: Gene isolation and expression. Yeast10, 1447–1457.

Sakai, Y., Saiganji, A., Yurimoto, H., Takabe, K., Saiki,H. and Kato, N. (1996a). The absence of Pmp47, aputative yeast peroxisomal transporter, causes adefect in transport and folding of a specific matrixenzyme. J. Cell. Biol. 134, 37–51.

Sakai, Y., Akiyama, M., Kondoh, H., Shibano, Y. andKato, N. (1996b). High-level secretion of fungal glu-coamylase using the Candida boidinii gene expressionsystem. Biochim. Biophys. Acta 1308, 81–87.

Sakai, Y., Matsuo, H., He, K.-Z., et al. (1995a). Isola-tion and characterization of mutants of the methylo-trophic yeast, Candida boidinii S2 that are impaired ingrowth on peroxisome-inducing carbon sources. Bio-sci. Biotech. Biochem. 59, 869–875.

Sakai, Y., Rogi, T., Takeuchi, R., Kato, N. and Tani, Y.(1995b). Expression of Saccharomyces adenylatekinase gene in Candida boidinii under the regulation ofits alcohol oxidase promoter. Appl. Microbiol. Bio-technol. 42, 860–864.

Sakai, Y., Marshall, P. A., Saiganji, A., et al. (1995c).The Candida boidinii peroxisomal membrane proteinPmp30 has a role in peroxisomal proliferation and isfunctionally homologous to Pmp27 from Saccharomy-ces cerevisiae. J. Bacteriol. 177, 6773–6781.

Sakai, Y., Goh, T. K. and Tani, Y. (1993). High-frequency transformation of a methylotropic yeast,Candida boidinii, with autonomously replicating plas-mids which are also functional in Saccharomycescerevisiae. J. Bacteriol. 175, 3556–3562.

Sakai, Y. and Tani, Y. (1992). Cloning and sequencingof the alcohol oxidase-encoding gene (AOD1) fromthe formaldehyde-producing asporogenous methylo-trophic yeast, Candida boidinii S2. Gene 114, 67–73.


Sakai, Y., Kazarimoto, T. and Tani, Y. (1991). Trans-formation system for an asporogenous methylo-trophic yeast, Candida boidinii: cloning of theorotidine-5*-phosphate decarboxylase gene (URA3),isolation of uracil auxotrophic mutants, and use of themutants for integrative transformation. J. Bacteriol.173, 7458–7463.

Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989).Molecular Cloning: A Laboratory Manual, 2nd edn.Cold Spring Harbor Laboratory Press, Cold SpringHarbor, New York.

Shimizu, S., Yasui, K., Tani, Y. and Yamada, H. (1979).Acyl-CoA oxidase from Candida tropicalis. Biochem.Biophys. Res. Commun. 91, 108–113.

Tani, Y., Sakai, Y. and Yamada, H. (1985a). Isolationand characterization of a mutant of a methanol yeast,Candida boidinii S2, with higher formaldehydeproductivity. Agric. Biol. Chem. 49, 2699–2706.

Tani, Y., Sakai, Y. and Yamada, H. (1985b). Produc-tion of formaldehyde by a mutant of methanol yeast,Candida boidinii S2. J. Ferment. Technol. 63, 443–449.

Towbin, H., Staehelin, T. and Gordon, J. (1979). Elec-trophoretic transfer of proteins from polyacrylamidegels to nitrocellulose sheets: procedure and someapplications. Proc. Natl. Acad. Sci. USA 76, 4350–4354.

Tuttle, D. L. and Dunn, W. A. (1995). Divergent modesof autophagy in the methylotrophic yeast Pichiapastoris. J. Cell Sci. 108, 25–35.

Tuttle, D. L., Lewin, A. S. and Dunn, W. A. J. (1993).Selective autophagy of peroxisomes in methylotrophicyeasts. Eur. J. Cell Biol. 60, 283–290.

van den Bosch, H., Schutgens, R. B. H., Wanders,R. J. A. and Tager, J. M. (1992). Biochemistry ofperoxisomes. Annu. Rev. Biochem. 61, 157–197.

Veenhuis, M. and Goodman, J. M. (1990). Peroxisomalassembly: membrane proliferation precedes theinduction of the abundant matrix proteins in themethylotrophic yeast Candida boidinii. J. Cell Sci. 96,583–590.

Waterham, H. R., Keizer-Gunnink, I., Goodman, J. M.,Harder, W. and Veenhuis, M. (1992). Development ofmultipurpose peroxisomes in Candida boidinii grownin oleic acid-methanol limited continuous cultures. J.Bacteriol. 174, 4057–4063.

Yeast 14, 1175–1187 (1998)

regulation of peroxisomal proteins and organelle proliferation by multiple carbon sources in the...

Documents