Vol. 186, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
July 31, 1992
Pages
Protein transport in the permeabilized Schizosaccharomyces pombe Hideko Kambe-Honjoh,
838-845
cell of
Koji Yoda, and Makari Yamasaki
Departmentof Agricultural Chemistry, The University of Tokyo, Bunkyo-ku, Tokyo 113, Japan Received
June
14,
1992
We reconstituted a protein translocation-transport system composed of permeabilized spheroplasts(P-cells) of the fission yeast Schizosaccharomycespombe and the precursorof a sex pheromone, prepro-a-factor of the budding yeast Saccharomycescerevisiae. We found that P-cellspreparedfrom the spheroplastsformed in 0.7M KC1 asan osmotic stabilizer had the activity to transport pro-a-factor to the Golgi apparatus.Electron microscopic observations showed that membranes were preserved more intact in the P-cells prepared from the spheroplastsformed in 0.7M KC1 than in 0.7M sorbitol. A glycoprotein of S. pombe contains galactoseresidues,and we detected incorporation of radiolabeled galactoseresiduesinto the anti-prepro-a-factor immunoprecipitable fractions in this S. pombe system, but not in the S. cerevisiae system. This paper reports that a heterologoussystemof in vitro protein transport wasperformed, and prepro-a-factor hasthe signalsnecessaryfor early stepsof the transportin S. pombe. 0 1992Academic Press,Inc.
The fission yeast S. pombe is now an important model organism for investigation of propertiesof eukaryotic cells, because it resemblesmammaliancells in aspectsof its cell cycle, chromosomal DNA structure and intron splicing, comparing with the budding yeast S. cerevisiae (1). For the study of generally conservedmechanismsin eukaryotes, especially about protein secretion, S. pombe has a potentiality to offer an efficient system of protein transport betweenthe ER and the Golgi apparatus,sinceit has morphologically distinct Golgi stacks (2). However, there are few reports on the mechanisms of protein secretion in S. pombe. S. pombe produces acid phosphatase(3) and invertase (4)(5). Their molecular weights increaseby glycosylation as in S. cerevisiae except the carbohydrates additionally contain galactose(4)(5). Becauseprotein transport and modification with oligosaccharideis controlled by host- and protein-dependentparameters,it is interesting to seehow a secretory protein of S. cerevisiae is transported and glycosylated in S. pombe. In this report we carried out in vitro protein transport reaction with S. pombecomponentsfor the first time.
Abbreviations: P-cells, permeabilized spheroplasts;KCl-P-cells and sorbitol-P-cells, P-cells prepared from the spheroplastsformed in the presenceof KC1 or sorbitol; KCl-,or sorbitolbuffer, spheroplastformation buffer containing KC1 or sorbitol asan osmotic stabilizer; ConA, concanavalinA. 0006-291X/92 $4.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
838
Vol.
186,
No.
2, 1992
BIOCHEMICAL
MATERIALS
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
AND METHODS
Preparation of [35S]-Met-labeled prepro-a-factor. The plasmid pGEM2-a36 encoding prepro-a-factor was obtained from J. Rothblatt (6) and transcribed by SP6 RNA polymerase as described by Melton et al.(7). Translation was done as described by Tuite et al. (8) with the lysate of S. pombe JY333(h-, ade6, leul) obtained from M. Yamamoto (University of Tokyo), or with the lysate of Scerevisiae ABYS66 (a, pd. prb2, prcl, cpsl , ade) obtained from D. Wolf (Biochemisches Institut der Universitat Freiburg). Translation products were desaltedby gel filtration on a SephadexG-25 column equilibrated in reaction buffer (20 mM HEPES, pH 6.8, 150 mM K acetate, 250 mM sorbitol, 5 mM Mg acetate). Peak fractions were mixed and concentrated by Centricon 30 (Amicon) to a l/10 volume and stored a -80 OC. Preparation of P-cells and cytosol. Cells were treated as described (9) with minor modifications. Cells were harvested when they reached an optical density at 600 nm. of l-2. Spheroplastsof S. pombe were prepared by incubation with 0.1 mg/ml Zymolyase 100,000 (Seikagaku Kogyo Co.) and 0.05 mg/ml Novozym 234 (Novo Industries),in the caseswithout any indications. Spheroplastsof S. cerevisiae were prepared by incubation with 0.1 mg/ml Zymolyase 100,000. After regeneration and wash, spheroplastswere resuspendedin lysis buffer (9), and aliquotesof 300~1were transferredinto Eppendorf tubesand frozen in the vapor above liquid N,. Upon thawing at 24 OCthe spheroplasts,now referred as P-cells, acquired capacitiesto incorporateexogeneousmacromoleculessuchasprepro-a-factor. In vitro protein translocation and transport reaction. The reaction was performed as describedby Baker et aL(9). Each reaction contained60 pl of P-cells, 2 11 of prepro-a-factor, 150 pg of additional cytosol , 50 pM GDP-mannose, 50 pM UDP-galactose, 1 mM ATP, 40 pM creatin phosphate,200 pg/ml creatin kinase, 1 mM cycloheximide, and the reaction buffer to bring the volume to 50 pl. Incubation for 45 min at 20 OCwas stoppedby addition of 50 pl of 2% SDS, and heated for 5 min at 95 OC. Analysis of the reaction products. For immunoprecipitation or ConA-precipitation, sampleswere diluted with 8OOplof IP buffer (9), mixed with 10 ~1 of anti-prepro-a-factor rabbit antiserumand 30 pl of 20 % (v/v) suspensionof protein A-Sepharose,or 30 pl of 20% (v/v) suspensionof ConA-Sepharose (Pharmacia), and rotated overnight at 4 OC. Then, Sepharosebeadswere washed as described (9) and heated at 95 OCin samplebuffer. The reaction products were analyzed by electrophoresison 11.25%SDS-polyacrylamide gel (10) or by scintilation counting. Radioactivity was alsodetectedby an imaging plate systemusingBioimage Analyzer BAS 2000 (FUJI PHOTOFILM CO., LTD.). Electron microscopy. Samples were fixed with 3% gfutaraldehyde in 0.05M phosphate buffer (pH 6.95) for 30 min, followed by 5 times washeswith the samebuffer and fixation in 1% 0~0, for 2hr. After washing with distilled water, sampleswere prestained with 0.5% uranyl acetate for 2 hr , and followed by agar-embeddingand dehydration with ethanol and acetone, and embeddedwith Spurr’sresin. Thin sectionswere stainedwith uranyl acetate and lead citrate, andobservedby a JEOL 200 CX electronmicroscopeat 100kV. Preparation of antibodies. Anti-prepro-o-factor antiserum was raised in rabbit against pgal-prepro-a-factor fusion protein coded on the plasmid pEX-a6 obtained from J.Rothblatt. This antiserum cross-reacts with synthetic a-factor peptide (Sigma). Anti-al ,6-mannose antiserumwasa kindly gift from A.Nakano.
RESULTS AND DISCUSSION Conditions
to make efficient
P-cells.
To study the mechanismsof protein secretion, we constructed in vitro translocation system derived from S. pombe. Permeabilized spheroplasts, P-cells, were used as membrane fractions, which have unbroken internal organellesbut porousplasmamembranes.P-cells were prepared by freezing and thawing of spheroplasts,and they were successfully used in an in vitro translocation and transport systemof S. cerevisiae (9). 839
Vol.
186,
No.
2, 1992
BIOCHEMICAL
;i 2 P e 2
AND
BIOPHYSICAL
RESEARCtI
COMMUNICATIONS
40 20
$ 0
Zymolyase
0.2
0.4
0.0
0.1
0.1
OWml)
Novo’zym
0.2
0.0
0.4
0.05
0.05
(mghnl)
Stabilizer
Sorbitol
KCI
Fig. 1. Comparison of translocation efficiency of P-cells derived from various preparation of spheroplasts. Spheroplasts were formed in buffers containing 1.2 M sorbitol (lane 1) or 0.7 M sorbitol (lanes 2-4), 0.75 x YP, 0.5% glucose, 10 mM Tris-HCl. (pH 7.5) (sorbitol-buffer), or in 0.7 M KC1 and 0.02 M KH2P0, (pH 5.8)(KCl-buffer)(lane 5) by treatments with Zymolyase and/or Novozym 234 at the concentrations indicated. Spheroplast formation was calculated from the decrease of the optical density at 600 nm after a I:50 dilution in H,O. The translocation activity by the P-cells is indicated as the percentages of the amounts of the ERform to the initial prepro-cc-factor added. Calculation of the amounts of the ER-form of pro-afactor were done by Bio-image analyzer,
In order to increase efficiency of translocation, we elaborated better conditions for P-cell preparation. Considering the conditions for spheroplast formation is the most critical point, we examined various combinations of osmotic stabilizer and enzyme concentration which were used during Zymolyase
and Novozym
treatment. After the enzyme treatment, the sorbitol-
containing medium or buffer were used for all samples as described in the Materials and Methods. Some examples of the results are shown in Fig. 1. Even though the efficiency
of
spheroplast formation was equally high in buffers containing 0.7 or 1.2 M sorbitol (sorbitolbuffer) or 0.7M KC1 (KCl-buffer), the translocation efficiency was twice higher in P-cells made from spheroplasts formed in 0.7 M KCl-buffer (KCl-P-cells) than those made from spheroplasts formed in 0.7 M or 1.2 M sorbitol-buffer (sorbitol-P-cells). This results suggested that membranes of intracellular organelle were maintained more intact in KCl-P-cells. To confirm this point, we observed spheroplasts and P-cells by electron microscope. Electron microscopic observations. Each of intact spheroplasts prepared in the buffer containing 0.7 M or 1.2M sorbitol or 0.7 M KC1 seemed to be similar on morphological
observation by electron microscope. However, 840
Vol.
186, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCtI COMMUNICATIONS
spheroplastpreparationin sorbitol buffer wasrevealedto be composedof a mixed population of intact spheroplastsand those with occasional breaks (Fig.2 ), while those in KCl-buffer was composed of uniformly intact spheroplasts.After permeabilization, clear morphological differences can be seenbetween the two P-cell preparations(Fig.3). The cell shapesand the extent of permeabilizationwere rather uniform in KCl-P-cells but heterogeneousin sorbitol-Pcell. Plasmamembranesand nuclei remainedintact in KCl-P-cells, but not in sorbitol-P-cells. Free ribosomeshad been washedout from both P-cells, though, membrane-boundribosomes remainedmore efficiently in KCl-P-cells. Theseobservationsshowedthat P-cellspreparedfrom spheroplastsformed in KCl-buffer kept their intracellular organellamembranemore intact and uninjured. This presumptionis coincident to the fact that KCl-P-cells possessed high efficiency of translocation. Since there are nascentcell wall on the cell division site, susceptibility of the cell wall to lytic enzymes should not be uniform over the whole surface area and may depend on kinds of osmotic stabilizer. If so, it is possiblethat somepart of the cell wall were lysed too much in sorbitol-buffer, and that is why most of sorbitol-P-cells had damaged nuclei and their translocation activity were lower than KCl-P-cells. It was reported that uniformly shaped spheroplastsof Spombe were difficult to make, and useof KC1 asan osmotic stabilizer gave good results for efficient formation of uniformly shaped spheroplasts(l1). This fact correspondsto our result that KCl-P-cells showed relatively uniform shapesand had high translocation efficiency. Spheroplastsof S. cerevisiae preparedin the stabilizing buffer of sorbitol are uniformly round, and P-cells madefrom them have high efficiency of translocation (9). It is a distinct difference between S. pombe and S. cerevisiae. In
vitro
transport
reaction.
We also detectedtransport activity in KCl-P-cells that had high efficiency of translocation.As shown in Fig.4, some heterogeneousmolecules of higher molecular weight (o-gpaF) were detectedabove the ER-form (c-gpaF) on SDS-PAGE. Their distribution pattern on SDS-PAGE was coincident to that of the highly glycosylated Golgi-form of pro-a-factor produced by the S. cerevisiae
P-cell system (9)(12). These heterogeneouslarger molecules were produced
posttranslationally in the presenceof cycloheximide. They could bind to anti-prepro-a-factor antiserumand ConA. In the presenceof 1 mM GTPyS, their formation was reducedto 20 % of the value in the absenceof GTPyS. GTPyS was known as an inhibitor of transport betweenthe ER and the Golgi in yeasts (9) and mammalian cells (13). These results indicated that the heterologouslarger moleculeswere the pro-o-factors further modified in the next compartment of the ER, i.e. the cis-Golgi compartment. In S. cerevisiae P-cells the outer chain glycosylation is started by the addition of a-1,6mannoseresiduesand the Golgi-form moleculescould be immunoprecipitatedwith anti-a- 1,6mannoseantiserum(9)(12). However it is not clear that what kind of sugarresidueswere added to the ER-form of pro-a-factor in the S. pombe P-cells, becausewe could not detected the Golgi-form molecules by the immunoprecipitation with anti-a- 1,6-mannoseantiserum. In S. pombe, the outer chain of a glycoprotein contains galactoseresiduesand UDP-galactosyl 841
Vol.
186,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH RESEARCtl
COMMUNICATIONS
Fig. 2. Morphology of Spheroplasts. Preparation of samples for electron microscopic observation were detailed in Materials and Methods.(A) Spheroplasts formed in 0.7M KC1 buffer;(B) spheroplasts formed in 0.7 M sorbitol buffer; Bars, 1 Frn.
842
Vol.
186,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Fig. 3. Morphology of P-cells. Sample preparation was described in Materials and Methods. (A) P-cells made from the spheroplasts formed in 0.7M KCI-buffer;(B) P-cells made from the spheroplasts formed in 0.7M sorbitol-buffer; Bars, 1 pm.
843
Vol.
186,
No.
2, 1992
BIOCHEMICAL
o-gpaf c-gpaF
AND
I
2
34
0’ -
-
45’ -
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
-{ +
GTPyS
45’ +
--+
Fig. 4. In vitro transport reaction of prepro-a-factor by the KCI-P-cells of S. pombe. Prepro-a-factor synthesized in vitro was incubated with P-cells made from spheroplasts formed in KCl-buffer. Reaction conditions and product analyses were done as described in Materials and Methods. Reactions of lanes 4 and 7 were done in the presence of 1 mM GTPyS. Lanes l-4, anti-prepro-a-factor precipitations; lanes 5-7, ConA precipitations; ppaF, prepro-afactor; c-gpaF, core-glycosylated pro-a-factor (ER-form); o-gpaF, outer-chainglycosylated
pro-a-factor(Golgi-form).
transferaseis reported to locate in the Golgi apparatus(2). It is possiblethat galactoseresidues were also added to the pro-a-factor in this 5’. pombe P-cell system and prevented immunoprecipitation of outer chain glycosylated pro-a-factor by anti-a- 1,6-mannose antiserum.We conducteda translocation/transportreaction usingnonlabeledprepro-a-factor in the presenceof UDP-[14C]-galactose to test this possibility (Table 1). Only in the S.pombe system, we detected the remarkable [ 14C]-galactose incorporation into the anti-a-factor immunoprecipitablefractions after the reaction. This indicated that someof pro-a-factor derived proteins were modified alsowith galactosein this Xpombe system. In this report, the in vitro prepro-a-factor translocation/transport was performed by the cell constituents of S. pombe. This transport progressed post translationally and inhibited by
Table 1 [14C]-galactose incorporation into the anti-prepro-a-factor
antiserum precipitable fractions
Reaction time (min)
0
45
S.pombe
0
3500
Kcerevisiae
IO
600
Reactions using nonlabeled prepro-a-factor in the presence of 0.1 pCi UDP-galactose (325 Ci/mol, Amersham) instead of cold UDP-galactose were carried out as described in Materials and Methods, and products bound to ConA-sepharose were solubilized and reprecipitated with anti-prepro-a-factor antiserum. The 14C counts (cpm) of immunoprecipitated fractions were evaluated by scintilation counting. The values are the average counts of duplicate samples of one of the two experiments. the results were reproducible. 844
Vol.
186,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
GTPyS. Our results indicated that prepro-a-factor,
RESEARCtI
COMMUNICATIONS
a secretory protein of S. cerevisiae, was
accessible not only to the ER but also to an early compartment of the Golgi apparatus in S. pombe where mannoseand/or galactoseresiduesof outer chain glycosylation was further addedto the core-glycosylated pro-a-factor. However the efficiency of transport was rather low. There may be some different signalsfor efficient transport between the two yeasts, for example, the folding of pro-o-factor in the ER of S. pombe
is not suitable for further
transport. If the efficiency of transport is increasedwith someimprovement, this S.pombe systemis possibleto be an useful yeast in vitro systemfor adding terminal galactoseresidues to exogeneous proteins such as precursors of mammalian glycoproteins of which mature proteins contain galactoseiysidues. If a secretory protein of S. pombe itself can be usedas a precursor protein in this ir; vitro system, and secretion mutants of S. pombe
are isolated,
further information will be provided.
ACKNOWLEDGMENTS
We thank Drs. J. Rothblatt, M. Yamamoto, D. Wolf, K.Tachibana and A.Nakano for their kindly suppliesof materials, and A. Hirata for her help in electron microscopic observation. This study was supported by a Grant-in-Aid for Scientific Researchfrom the Ministry of Education, Science and Culture of Japanand a Grant for “Bio-design ResearchProgram” from the Institute of Physical and ChemicalResearch(Riken).
REFERENCES
1. Russell,P. and Nurse,P.(1986) Cell 45, 781-782. 2. Chappe1,T.G.and Warren,G.( 1989) J.Cell Biol. 109, 2693-2702. 3. Schweingruber,A.-M., Schoenholzer,F., Keller,L.,Schwaninger,R.,Trachsel,H. and Schweingruber,M.E.(1986) Eur.J. Biochem. 158, 133-140. 4. Moreno& Ruis,T., SBnchez,Y.. Villanueva,J.R. and.Rodriguez.L.( 1985) Arch.Microbiol. 142, 370-374. 5. Moreno& Sanchez,Y. and Rodrlguez,L.( 1990) Biochem. 5.267, 697-712. 6. Rothblatt,J.A. and Meyer,D.I. (1986) Cell 44, 619-628. 7. Melton,D.A., Kreig,P.A., Rebagliati,M.R., Maniatis,T.,Zinn,K. and Green,M.R.( 1984)Nucleic Acids Res. 12. 7035-7056. 8. Tuite,M.F., and Plesset,J.(1986) Yeast 2, 35-52. 9. Baker,D., Hicke,L., Reach,M., Schleyer,M. and Schekman,R.(l988) Cell 54, 335-344. 10. Laemmli,U.K.( 1970) Nature 227, 680-685. 11. Mann,W. and Jeffery,J.( 1986) BioscienceReports 6. 597-602. 12. Rouhola,H., Kabcenel1,A.K. and Ferro-Novick,S.(l988) J.Cell Biol. 107, 1465-1476. 13. Beckers,C.J.M. and Balch,W.E.(1989) J.Cell Biol. 108. 1245-1256.
845
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
July 31, 1992
Pages
Protein transport in the permeabilized Schizosaccharomyces pombe Hideko Kambe-Honjoh,
838-845
cell of
Koji Yoda, and Makari Yamasaki
Departmentof Agricultural Chemistry, The University of Tokyo, Bunkyo-ku, Tokyo 113, Japan Received
June
14,
1992
We reconstituted a protein translocation-transport system composed of permeabilized spheroplasts(P-cells) of the fission yeast Schizosaccharomycespombe and the precursorof a sex pheromone, prepro-a-factor of the budding yeast Saccharomycescerevisiae. We found that P-cellspreparedfrom the spheroplastsformed in 0.7M KC1 asan osmotic stabilizer had the activity to transport pro-a-factor to the Golgi apparatus.Electron microscopic observations showed that membranes were preserved more intact in the P-cells prepared from the spheroplastsformed in 0.7M KC1 than in 0.7M sorbitol. A glycoprotein of S. pombe contains galactoseresidues,and we detected incorporation of radiolabeled galactoseresiduesinto the anti-prepro-a-factor immunoprecipitable fractions in this S. pombe system, but not in the S. cerevisiae system. This paper reports that a heterologoussystemof in vitro protein transport wasperformed, and prepro-a-factor hasthe signalsnecessaryfor early stepsof the transportin S. pombe. 0 1992Academic Press,Inc.
The fission yeast S. pombe is now an important model organism for investigation of propertiesof eukaryotic cells, because it resemblesmammaliancells in aspectsof its cell cycle, chromosomal DNA structure and intron splicing, comparing with the budding yeast S. cerevisiae (1). For the study of generally conservedmechanismsin eukaryotes, especially about protein secretion, S. pombe has a potentiality to offer an efficient system of protein transport betweenthe ER and the Golgi apparatus,sinceit has morphologically distinct Golgi stacks (2). However, there are few reports on the mechanisms of protein secretion in S. pombe. S. pombe produces acid phosphatase(3) and invertase (4)(5). Their molecular weights increaseby glycosylation as in S. cerevisiae except the carbohydrates additionally contain galactose(4)(5). Becauseprotein transport and modification with oligosaccharideis controlled by host- and protein-dependentparameters,it is interesting to seehow a secretory protein of S. cerevisiae is transported and glycosylated in S. pombe. In this report we carried out in vitro protein transport reaction with S. pombecomponentsfor the first time.
Abbreviations: P-cells, permeabilized spheroplasts;KCl-P-cells and sorbitol-P-cells, P-cells prepared from the spheroplastsformed in the presenceof KC1 or sorbitol; KCl-,or sorbitolbuffer, spheroplastformation buffer containing KC1 or sorbitol asan osmotic stabilizer; ConA, concanavalinA. 0006-291X/92 $4.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
838
Vol.
186,
No.
2, 1992
BIOCHEMICAL
MATERIALS
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
AND METHODS
Preparation of [35S]-Met-labeled prepro-a-factor. The plasmid pGEM2-a36 encoding prepro-a-factor was obtained from J. Rothblatt (6) and transcribed by SP6 RNA polymerase as described by Melton et al.(7). Translation was done as described by Tuite et al. (8) with the lysate of S. pombe JY333(h-, ade6, leul) obtained from M. Yamamoto (University of Tokyo), or with the lysate of Scerevisiae ABYS66 (a, pd. prb2, prcl, cpsl , ade) obtained from D. Wolf (Biochemisches Institut der Universitat Freiburg). Translation products were desaltedby gel filtration on a SephadexG-25 column equilibrated in reaction buffer (20 mM HEPES, pH 6.8, 150 mM K acetate, 250 mM sorbitol, 5 mM Mg acetate). Peak fractions were mixed and concentrated by Centricon 30 (Amicon) to a l/10 volume and stored a -80 OC. Preparation of P-cells and cytosol. Cells were treated as described (9) with minor modifications. Cells were harvested when they reached an optical density at 600 nm. of l-2. Spheroplastsof S. pombe were prepared by incubation with 0.1 mg/ml Zymolyase 100,000 (Seikagaku Kogyo Co.) and 0.05 mg/ml Novozym 234 (Novo Industries),in the caseswithout any indications. Spheroplastsof S. cerevisiae were prepared by incubation with 0.1 mg/ml Zymolyase 100,000. After regeneration and wash, spheroplastswere resuspendedin lysis buffer (9), and aliquotesof 300~1were transferredinto Eppendorf tubesand frozen in the vapor above liquid N,. Upon thawing at 24 OCthe spheroplasts,now referred as P-cells, acquired capacitiesto incorporateexogeneousmacromoleculessuchasprepro-a-factor. In vitro protein translocation and transport reaction. The reaction was performed as describedby Baker et aL(9). Each reaction contained60 pl of P-cells, 2 11 of prepro-a-factor, 150 pg of additional cytosol , 50 pM GDP-mannose, 50 pM UDP-galactose, 1 mM ATP, 40 pM creatin phosphate,200 pg/ml creatin kinase, 1 mM cycloheximide, and the reaction buffer to bring the volume to 50 pl. Incubation for 45 min at 20 OCwas stoppedby addition of 50 pl of 2% SDS, and heated for 5 min at 95 OC. Analysis of the reaction products. For immunoprecipitation or ConA-precipitation, sampleswere diluted with 8OOplof IP buffer (9), mixed with 10 ~1 of anti-prepro-a-factor rabbit antiserumand 30 pl of 20 % (v/v) suspensionof protein A-Sepharose,or 30 pl of 20% (v/v) suspensionof ConA-Sepharose (Pharmacia), and rotated overnight at 4 OC. Then, Sepharosebeadswere washed as described (9) and heated at 95 OCin samplebuffer. The reaction products were analyzed by electrophoresison 11.25%SDS-polyacrylamide gel (10) or by scintilation counting. Radioactivity was alsodetectedby an imaging plate systemusingBioimage Analyzer BAS 2000 (FUJI PHOTOFILM CO., LTD.). Electron microscopy. Samples were fixed with 3% gfutaraldehyde in 0.05M phosphate buffer (pH 6.95) for 30 min, followed by 5 times washeswith the samebuffer and fixation in 1% 0~0, for 2hr. After washing with distilled water, sampleswere prestained with 0.5% uranyl acetate for 2 hr , and followed by agar-embeddingand dehydration with ethanol and acetone, and embeddedwith Spurr’sresin. Thin sectionswere stainedwith uranyl acetate and lead citrate, andobservedby a JEOL 200 CX electronmicroscopeat 100kV. Preparation of antibodies. Anti-prepro-o-factor antiserum was raised in rabbit against pgal-prepro-a-factor fusion protein coded on the plasmid pEX-a6 obtained from J.Rothblatt. This antiserum cross-reacts with synthetic a-factor peptide (Sigma). Anti-al ,6-mannose antiserumwasa kindly gift from A.Nakano.
RESULTS AND DISCUSSION Conditions
to make efficient
P-cells.
To study the mechanismsof protein secretion, we constructed in vitro translocation system derived from S. pombe. Permeabilized spheroplasts, P-cells, were used as membrane fractions, which have unbroken internal organellesbut porousplasmamembranes.P-cells were prepared by freezing and thawing of spheroplasts,and they were successfully used in an in vitro translocation and transport systemof S. cerevisiae (9). 839
Vol.
186,
No.
2, 1992
BIOCHEMICAL
;i 2 P e 2
AND
BIOPHYSICAL
RESEARCtI
COMMUNICATIONS
40 20
$ 0
Zymolyase
0.2
0.4
0.0
0.1
0.1
OWml)
Novo’zym
0.2
0.0
0.4
0.05
0.05
(mghnl)
Stabilizer
Sorbitol
KCI
Fig. 1. Comparison of translocation efficiency of P-cells derived from various preparation of spheroplasts. Spheroplasts were formed in buffers containing 1.2 M sorbitol (lane 1) or 0.7 M sorbitol (lanes 2-4), 0.75 x YP, 0.5% glucose, 10 mM Tris-HCl. (pH 7.5) (sorbitol-buffer), or in 0.7 M KC1 and 0.02 M KH2P0, (pH 5.8)(KCl-buffer)(lane 5) by treatments with Zymolyase and/or Novozym 234 at the concentrations indicated. Spheroplast formation was calculated from the decrease of the optical density at 600 nm after a I:50 dilution in H,O. The translocation activity by the P-cells is indicated as the percentages of the amounts of the ERform to the initial prepro-cc-factor added. Calculation of the amounts of the ER-form of pro-afactor were done by Bio-image analyzer,
In order to increase efficiency of translocation, we elaborated better conditions for P-cell preparation. Considering the conditions for spheroplast formation is the most critical point, we examined various combinations of osmotic stabilizer and enzyme concentration which were used during Zymolyase
and Novozym
treatment. After the enzyme treatment, the sorbitol-
containing medium or buffer were used for all samples as described in the Materials and Methods. Some examples of the results are shown in Fig. 1. Even though the efficiency
of
spheroplast formation was equally high in buffers containing 0.7 or 1.2 M sorbitol (sorbitolbuffer) or 0.7M KC1 (KCl-buffer), the translocation efficiency was twice higher in P-cells made from spheroplasts formed in 0.7 M KCl-buffer (KCl-P-cells) than those made from spheroplasts formed in 0.7 M or 1.2 M sorbitol-buffer (sorbitol-P-cells). This results suggested that membranes of intracellular organelle were maintained more intact in KCl-P-cells. To confirm this point, we observed spheroplasts and P-cells by electron microscope. Electron microscopic observations. Each of intact spheroplasts prepared in the buffer containing 0.7 M or 1.2M sorbitol or 0.7 M KC1 seemed to be similar on morphological
observation by electron microscope. However, 840
Vol.
186, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCtI COMMUNICATIONS
spheroplastpreparationin sorbitol buffer wasrevealedto be composedof a mixed population of intact spheroplastsand those with occasional breaks (Fig.2 ), while those in KCl-buffer was composed of uniformly intact spheroplasts.After permeabilization, clear morphological differences can be seenbetween the two P-cell preparations(Fig.3). The cell shapesand the extent of permeabilizationwere rather uniform in KCl-P-cells but heterogeneousin sorbitol-Pcell. Plasmamembranesand nuclei remainedintact in KCl-P-cells, but not in sorbitol-P-cells. Free ribosomeshad been washedout from both P-cells, though, membrane-boundribosomes remainedmore efficiently in KCl-P-cells. Theseobservationsshowedthat P-cellspreparedfrom spheroplastsformed in KCl-buffer kept their intracellular organellamembranemore intact and uninjured. This presumptionis coincident to the fact that KCl-P-cells possessed high efficiency of translocation. Since there are nascentcell wall on the cell division site, susceptibility of the cell wall to lytic enzymes should not be uniform over the whole surface area and may depend on kinds of osmotic stabilizer. If so, it is possiblethat somepart of the cell wall were lysed too much in sorbitol-buffer, and that is why most of sorbitol-P-cells had damaged nuclei and their translocation activity were lower than KCl-P-cells. It was reported that uniformly shaped spheroplastsof Spombe were difficult to make, and useof KC1 asan osmotic stabilizer gave good results for efficient formation of uniformly shaped spheroplasts(l1). This fact correspondsto our result that KCl-P-cells showed relatively uniform shapesand had high translocation efficiency. Spheroplastsof S. cerevisiae preparedin the stabilizing buffer of sorbitol are uniformly round, and P-cells madefrom them have high efficiency of translocation (9). It is a distinct difference between S. pombe and S. cerevisiae. In
vitro
transport
reaction.
We also detectedtransport activity in KCl-P-cells that had high efficiency of translocation.As shown in Fig.4, some heterogeneousmolecules of higher molecular weight (o-gpaF) were detectedabove the ER-form (c-gpaF) on SDS-PAGE. Their distribution pattern on SDS-PAGE was coincident to that of the highly glycosylated Golgi-form of pro-a-factor produced by the S. cerevisiae
P-cell system (9)(12). These heterogeneouslarger molecules were produced
posttranslationally in the presenceof cycloheximide. They could bind to anti-prepro-a-factor antiserumand ConA. In the presenceof 1 mM GTPyS, their formation was reducedto 20 % of the value in the absenceof GTPyS. GTPyS was known as an inhibitor of transport betweenthe ER and the Golgi in yeasts (9) and mammalian cells (13). These results indicated that the heterologouslarger moleculeswere the pro-o-factors further modified in the next compartment of the ER, i.e. the cis-Golgi compartment. In S. cerevisiae P-cells the outer chain glycosylation is started by the addition of a-1,6mannoseresiduesand the Golgi-form moleculescould be immunoprecipitatedwith anti-a- 1,6mannoseantiserum(9)(12). However it is not clear that what kind of sugarresidueswere added to the ER-form of pro-a-factor in the S. pombe P-cells, becausewe could not detected the Golgi-form molecules by the immunoprecipitation with anti-a- 1,6-mannoseantiserum. In S. pombe, the outer chain of a glycoprotein contains galactoseresiduesand UDP-galactosyl 841
Vol.
186,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH RESEARCtl
COMMUNICATIONS
Fig. 2. Morphology of Spheroplasts. Preparation of samples for electron microscopic observation were detailed in Materials and Methods.(A) Spheroplasts formed in 0.7M KC1 buffer;(B) spheroplasts formed in 0.7 M sorbitol buffer; Bars, 1 Frn.
842
Vol.
186,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Fig. 3. Morphology of P-cells. Sample preparation was described in Materials and Methods. (A) P-cells made from the spheroplasts formed in 0.7M KCI-buffer;(B) P-cells made from the spheroplasts formed in 0.7M sorbitol-buffer; Bars, 1 pm.
843
Vol.
186,
No.
2, 1992
BIOCHEMICAL
o-gpaf c-gpaF
AND
I
2
34
0’ -
-
45’ -
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
-{ +
GTPyS
45’ +
--+
Fig. 4. In vitro transport reaction of prepro-a-factor by the KCI-P-cells of S. pombe. Prepro-a-factor synthesized in vitro was incubated with P-cells made from spheroplasts formed in KCl-buffer. Reaction conditions and product analyses were done as described in Materials and Methods. Reactions of lanes 4 and 7 were done in the presence of 1 mM GTPyS. Lanes l-4, anti-prepro-a-factor precipitations; lanes 5-7, ConA precipitations; ppaF, prepro-afactor; c-gpaF, core-glycosylated pro-a-factor (ER-form); o-gpaF, outer-chainglycosylated
pro-a-factor(Golgi-form).
transferaseis reported to locate in the Golgi apparatus(2). It is possiblethat galactoseresidues were also added to the pro-a-factor in this 5’. pombe P-cell system and prevented immunoprecipitation of outer chain glycosylated pro-a-factor by anti-a- 1,6-mannose antiserum.We conducteda translocation/transportreaction usingnonlabeledprepro-a-factor in the presenceof UDP-[14C]-galactose to test this possibility (Table 1). Only in the S.pombe system, we detected the remarkable [ 14C]-galactose incorporation into the anti-a-factor immunoprecipitablefractions after the reaction. This indicated that someof pro-a-factor derived proteins were modified alsowith galactosein this Xpombe system. In this report, the in vitro prepro-a-factor translocation/transport was performed by the cell constituents of S. pombe. This transport progressed post translationally and inhibited by
Table 1 [14C]-galactose incorporation into the anti-prepro-a-factor
antiserum precipitable fractions
Reaction time (min)
0
45
S.pombe
0
3500
Kcerevisiae
IO
600
Reactions using nonlabeled prepro-a-factor in the presence of 0.1 pCi UDP-galactose (325 Ci/mol, Amersham) instead of cold UDP-galactose were carried out as described in Materials and Methods, and products bound to ConA-sepharose were solubilized and reprecipitated with anti-prepro-a-factor antiserum. The 14C counts (cpm) of immunoprecipitated fractions were evaluated by scintilation counting. The values are the average counts of duplicate samples of one of the two experiments. the results were reproducible. 844
Vol.
186,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
GTPyS. Our results indicated that prepro-a-factor,
RESEARCtI
COMMUNICATIONS
a secretory protein of S. cerevisiae, was
accessible not only to the ER but also to an early compartment of the Golgi apparatus in S. pombe where mannoseand/or galactoseresiduesof outer chain glycosylation was further addedto the core-glycosylated pro-a-factor. However the efficiency of transport was rather low. There may be some different signalsfor efficient transport between the two yeasts, for example, the folding of pro-o-factor in the ER of S. pombe
is not suitable for further
transport. If the efficiency of transport is increasedwith someimprovement, this S.pombe systemis possibleto be an useful yeast in vitro systemfor adding terminal galactoseresidues to exogeneous proteins such as precursors of mammalian glycoproteins of which mature proteins contain galactoseiysidues. If a secretory protein of S. pombe itself can be usedas a precursor protein in this ir; vitro system, and secretion mutants of S. pombe
are isolated,
further information will be provided.
ACKNOWLEDGMENTS
We thank Drs. J. Rothblatt, M. Yamamoto, D. Wolf, K.Tachibana and A.Nakano for their kindly suppliesof materials, and A. Hirata for her help in electron microscopic observation. This study was supported by a Grant-in-Aid for Scientific Researchfrom the Ministry of Education, Science and Culture of Japanand a Grant for “Bio-design ResearchProgram” from the Institute of Physical and ChemicalResearch(Riken).
REFERENCES
1. Russell,P. and Nurse,P.(1986) Cell 45, 781-782. 2. Chappe1,T.G.and Warren,G.( 1989) J.Cell Biol. 109, 2693-2702. 3. Schweingruber,A.-M., Schoenholzer,F., Keller,L.,Schwaninger,R.,Trachsel,H. and Schweingruber,M.E.(1986) Eur.J. Biochem. 158, 133-140. 4. Moreno& Ruis,T., SBnchez,Y.. Villanueva,J.R. and.Rodriguez.L.( 1985) Arch.Microbiol. 142, 370-374. 5. Moreno& Sanchez,Y. and Rodrlguez,L.( 1990) Biochem. 5.267, 697-712. 6. Rothblatt,J.A. and Meyer,D.I. (1986) Cell 44, 619-628. 7. Melton,D.A., Kreig,P.A., Rebagliati,M.R., Maniatis,T.,Zinn,K. and Green,M.R.( 1984)Nucleic Acids Res. 12. 7035-7056. 8. Tuite,M.F., and Plesset,J.(1986) Yeast 2, 35-52. 9. Baker,D., Hicke,L., Reach,M., Schleyer,M. and Schekman,R.(l988) Cell 54, 335-344. 10. Laemmli,U.K.( 1970) Nature 227, 680-685. 11. Mann,W. and Jeffery,J.( 1986) BioscienceReports 6. 597-602. 12. Rouhola,H., Kabcenel1,A.K. and Ferro-Novick,S.(l988) J.Cell Biol. 107, 1465-1476. 13. Beckers,C.J.M. and Balch,W.E.(1989) J.Cell Biol. 108. 1245-1256.
845
