Current Genetics
Current Genetics (1982) 5:109-117
© Springer-Verlag 1982
Isolation and Characterization of Acid Phosphatase Mutants in Schizosaccharomycespombe M. E. Schweingruber I , A. M. Schweingruber 1 , and M. E. Sch@bach 2 1 Institute for General Microbiology,Baltzer-Strasse4, CH-3012 Bern, Switzerland 2 Biologicaland Pharmaceutical Research Department, F. Hoffmann-LaRoche and Co. Ltd., CH4002 Basel, Switzerland
Summary. 72 mutants defective in the activity of the cell surface glycoprotein acid phosphatase were isolated and characterized. The mutants map in one genetic locus, phol. Many of them exhibit altered cell morphology. This characteristic cosegregates with acid phosphatase deficiency, implying that phol controls the activity of acid phosphatase and concomitantly cell morphology. Phol probably also influences growth rate and agglutination behaviour. By purifying acid phosphatase, two structurally related forms can be detected. One is inactive (form I) and one is the active acid phosphatase (form II). Mutant phol-270 and phol-277 lack the inactive form I. Mutant phol-38 exhibits mainly form I, form II being present only in minor amounts. Two other mutants examined differ only slightly from wildtype in their pattern of active and inactive forms. Tryptic peptide maps of the inactive and active acid phosphatase of the wildtype and the corresponding proteins of mutant phol304 reveal similar structural alterations for the two mutant proteins. The results show that phol controls the expression of the active and inactive acid phosphatase. We conclude that phol represents the structural gene of the two forms of acid phosphatase. Key words: Schizosaccharomyces pombe - Active acid phosphatase - Inactive acid phosphatase - Cell-surface glycoprotein
Introduction
To understand the genetic control and biological role of a specific cell surface glycoprotein we are investigating
Offprint requeststo: M. E. Schweingruber
acid phosphatase of yeast. Acid phosphatase of Saccharomyces cerevisiae is a mannoprotein and is mainly located in the cell wall (Linnemans et al. 1977). Minor amounts have also been found in close association with the plasmamembrane and other cell organelles (Van Rijn et al. 1975; Boer et al. 1975). We have recently shown that the enzyme can also exist in an inactive probably membrane-bound form (Schweingruber und Schweingmbet, in press). Little is known about the biological role of acid phosphatase. One function may be the hydrolysis of external phosphate esters which do not penetrate the protoplasmic membrane (G~ther and Kaffner 1968). We speculated that acid phosphatase is also involved in morphogenetic and behavioural differentiation of the yeast cell (Schweingruber and Schweingruber 1981). To elucidate its biological role we are investigating acid phosphatase of the fission yeast Schizosaccharomyces pornbe. Acid phosphatase of S. pombe is, like the corresponding enzyme of the budding yeast, a cell surface glycoprotein regulated by extracellular orthophosphate (Dibenedetto 1972). It has been purified to homogeneity and characterized (Dibenedetto and Cozzani 1975; Dibenedetto and Teller 1981). The active protein has a molecular weight of roughly 360,000 and can be dissociated into subunits with a molecular weight of about 90,000. Neutral sugars account for about 66% of the total molecular weight. Mannose and galactose are found as the main components of the carbohydrate moiety. Since its synthesis is tightly associated with cell elongation, it has been considered as a growth marker (Miyata and Miyata 1978). In this communication we report the isolation and characterization of mutants lacking acid posphatase activity. 0172-8083/0005/0109/$ O1.80
M.E. Schweingruber et al.: Mutants of Acid Phosphatase in Sehizosaccharomycespombe
110
Table 1. Properties of phol-mutants lacking acid phosphatase activity Class
I
Strain number
Morphology
Agglutination
Growth rate
38, 44, 86, 102 150,157,171,173 124,182,185,187 202,212,222,169
normal
normal
normal
PepticIe Mapping by High Pressure Liquid Chromatography (HPLC) Tryptic digests were prepared as follows: 1-2 mg of phosphatase protein were dissolved in 0.5 ml ammonium bicarbonate (2 mg/ml), heated for 5 rain at 100 °C and subsequerrtly digested with 2% (by weight of protein) trypsin (TPCK-Trypsin, Boehringer) for 16 h at 30 °C. Approximately 50 ug of the digest were applied to a 25 cm x 0.5 cm Spherisorb ODS 10 ~z-column and eluted with an orthophosphate-acetonitrile gradient by HPLC as described in detail elsewhere (Schweingruber and Schweingruber, in press). The peptides were monitored at 210 nm.
226,230,232,233 240,243,149,151 184,219,231,234 244,256,260,261 262,266,267,269 272,281,282,283
283,291,292,302 307,308,310 II
phatase (see Fig. 3 ) we observed an other protein peak in the last purification step. The appearence of this peak has not been reported by Dibenedetto and Cozzani (1975). All available data (see Results) indicate that this peak represents enzymatically inactive acid phosphatase.
94,118,133,144 189,181,238,242 247,248,168,236 251,258,264,271 274,278,284,294 304,203
altered
normal
normal
III
270
altered
clumpy
normal
IV
277
altered
normal
altered
Isolation and Mapping of the phol Mutants Isolation. A stationary phase culture of wildtype 972, grown in
The mutants listed in the table were selected on the basis of their lack of acid phosphatase activity in colonies on plates. Quantitative measurements revealed, for all the mutants, less than 2% of the wildtype activity.Morphology, agglutination and growth rates were examined form cells grown at 30 °C in low phosphate medium. Morphology was examined under the microscope (see Fig. 1), agglutination was detected macroscopically (flocking cultures) and microscopically (see Fig. 1), g o w t h rates were determined as given in Fig. 2. Ceils exhibiting morphology, agglutination behaviour and growth rate only slightly different from wild-type were scored as normal
Materials a n d M e t h o d s
Growth Conditions and Purification of Acid Phosphatase Cells were grown in 21-Erlenmeyers containing 11YEPD-medium from which inorganic phosphate had been removed, referred to as low phosphate medium (Schweingruber and Schweingruber 1981), at 30 °C on a rotary shaker (Pilot Shake RC2) at a rotation speed of 120 rpm. At the end of the exponential phase, cells were harvested and washed in 0.1 M sodium acetate buffer. The yield was around 10 g wet cells/1 medium. Acid phosphatase was purified essentially according to the method of Dibenedetto and Cozzani (1975). This method has been shown to yield pure acid phosphatase as ascertained from ultracentrifugation, electrophoretic, chromatographic and endgroup determination data (Dibenedetto and Cozzani 1975; Dibenedetto and Teller 1981). We could confirm purity of the enzyme. By gelelectrophoresis under non denaturing conditions the purified enzyme yielded a single protein band that exhibited phosphatase activity as reported and by running the enzyme on SDS-gels only a single protein band was observed. In addition to peak II representing active acid phos-
yeast extract at 30°C, was centrifuged and washed twice in 0.9% NaC1. A suspension of 107 cells per ml was irradiated with 254 nm UV-light with a dose of 210,000 ergs x cm - 2 , resulting in 30-40% survival in the first experiments (isolates No. 1-90) and with a dose of 378,000 ergs x cm - 2 resulting in 1-9% survival for the other experiments (isolates Nr. 91-312). The irradiated cells were plated in an appropriate dilution in NaC1 solution on low phosphate medium and incubated at 25 °C. After 5 days incubation, the colonies were overlaid with 5 ml of the following mixture: 5 mg c~-naphthylphosphate (Na-Salt, SERVA), 50 mg Fast Blue B salt (SERVA) in 32 ml of 0.05 M sodium acetate buffer, pH 4.8, containing 1% Difco Bacto agar. Enzyme activity is indicated by a red staining of the colony. Plates were scored 5 to 10 rain after plating of the overlay. Before putting into stock, the isolates were restreaked and checked again for phosphatase activity. In addition, the absence of acid phosphatase activity in presumptive phol strains was confirmed by assaying cells that had been grown in low phosphate liquid medium, by the nitrophenylphosphate assay (Schweingruber and Schweingruber 1981).
Mapping. The isolated mutants, all of the h - mating type, were corssed into the h + mating type. For a genetic classification, a random analysis of ascospores, resulting from crosses phol h- x pho] h + on malt extract slants, was performed according to Leupold (1970). The spores were plated in appropriate dilutions on low phosphate medium plates, incubated at 25 °C and the colonies were then overlaid with c~-naphthylphosphate and Fast Blue B as described.
Results Mutants Lacking Acid Phosphatase Activity Exhibit Altered Morphology, Agglutination and Growth Rate Out o f roughly 100,000 screened colonies, we isolated 72 m u t a n t s w h i c h are deficient in acid p h o s p h a t a s e activity w h e n t e s t e d w i t h n i t r o p h e n y l p h o s p h a t e as substrate. The m u t a n t strains were e x a m i n e d for p r o p e r t i e s such as cell m o r p h o l o g y , agglutination b e h a v i o u r and
M. E. Schweingruber et al.: Mutants of Acid Phosphatase in Sehizosaccharomycespornbe
l 11
Fig. 1. A-D. Morphology of cells of wildtype 972 and 3 different phol-mutants, Strains were grown in low phosphate medium and examined under the microscope at a 400 fold magnification and photographed. The 3 phol-mutants are morphological aberrant. Phol-270 is, in addition, clumpy'
10 B
10
r
F
..a
growth rate. The results are summarized in Table 1. The mutants can be roughly divided into 4 classes. Apart from their lack of acid phosphatase activity, mutants of class I do not substantially differ from wild-type. Mutants of class II exhibit aberrant cell morphology, while the mutant of class III in addition clumps strongly. The class IV-mutant exhibits altered cell morphology and a drastically reduced growth rate. The picture of wildtype cells and of cells from a representative mutant of each of class II, III and IV are given in Fig. 1. The growth curve for the wildtype and forphol-277 is given in Fig. 2. It has to be stressed that the classification of
i=o
_l.i~
~-~
t (MIN)
1D~
~l Fig. 2. Growth curve of wildtype 972 and phol-277. Cells were grown in low phosphate medium. Growth was monitored by determining at intervals the titer of viable (colony forming) cells (=, e) and by measuring the turbidity of the culture at 530 nm (o, o). Both methods yielded roughly the same growth curve. The generation time of logarithmic phase cells is, under the given experimental conditions approximately 120 rain for the wildtype (open symbols) and 330 min for mutant phol-277 (closed symbols)
M.E. Schweingruber et al.: Mutants of Acid Phosphatase in Sehizosaeeharornyeespombe
112
Table 2. Tetrad analysis of different phol-mutants Strain number
Phenotyp (in addition to acid phosphatase deficiency)
Full tetrads analyzed
phol-187 phol-232 phol-244 phol-133 phol-247 phol-304 phol-270 phol-277
normal normal normal aberrant aberrant aberrant aberrant aberrant growing
13 4 10 6 11 8 7 10
cell morphology cell morphology cell morphology cell morphology, clumpy cell morphoiogy, slow
The phol-mutants (h-) listed in the table were crossed with the wildtype 975 h + and the progeny cells of the 4 spores of each tetrad were examined for acid phosphatase activity and if necessary for abberrant cell morphology agglutination behaviour and growth rate. In all cases a 2:2 segregation for wildtype and mutant phosphatase activity was observed. Aberrant cell morphology, agglutination behaviour and slow growth always cosegregated with acid phosphatase deficiency. Only the number of analyzed tetrads yielding 4 viable spores are given in the table. For mutant phol-277 spore germination was drastically reduced, less than 25% of the dissected asci yielded 4 colonies.
the mutants is arbitrary. Many mutants of class I do slightly differ form wildtype in morphology, agglutination and growth rate but the effects are rather slight while most mutants of class II, III and IV exhibit in addition to altered cell morphology, slight alterations in growth rate and agglutination behaviour. Many of the mutant strains of class II, III and IV are genetically unstable. This is indicated by the relatively high frequency of sectored colonies after staining for phosphatase activity. In many of the strains sporulation is impaired. Representative mutant strains of each class were crossed to the wildtype, and progeny cells of the spores of tetrads were examined for acid phosphatase acitivity, cell morphology, growth rate and agglutination. The strains and the number of full tetrads analyzed are given in Table 2. In all tetrads analyzed there was a 2 : 2 segregation for wildtype and acid phosphatase deficient phenotype. Aberrant cell morphology, agglutination behaviour and reduced growth rate cosegregated with the deficiency in phosphatase activity.
All the Mutants Deficient in Acid Phosphatase Activity Map Genetically at the Same Locus, phol All the mutants were crossed to 3 tester pho-mutants, which had been shown in preliminary experiments to belong to the same locus or at least to be closely linked. 5 0 - 5 0 0 meiotic progeny clones were examined for phosphatase activity. In no instance did we find colo-
nies containing phosphatase activity. This indicates that all mutants given in Table 1 map in the same genetic locus called phol.
Purification and Identification of an Inactive Form of Acid Phosphatase in Wildtype A method for the purification of acid phosphatase yielding a pure enzyme preparation has been published (Dibenedetto and Cozzani 1975; Dibenedetto and Teller 1981). Using this purification procedure we find that in the last step of the purification procedure - chromatography on a Sepharose CL-6B column - two peaks elute (Fig. 3). Peak II represents pure active acid phosphatase (see Material and Methods). The material of peak I is enzymatically inactive. For S. cerevisiae we have given evidence that the material of peak I represents inactive acid phosphatase (Schweingruber and Schweingruber, in press). A comparison of the tryptic peptide maps of active acid phosphatase and of the material eluting as peak I indicates that peak I of S. pombe also contains substantial amounts of inactive acid phosphatase. As shown in Fig. 4A and 4B the material of peak I reveals many peptides identical of those of active acid phosphatase (for details see legend to Fig. 4). We therefore call the material of peak I tentatively inactive acid phosphatase. Preliminary experiments reveal that the structure of the inactive form is more complex than that of the active form. In addition to carbohydrates it also contains lipids. This seems to impede its analysis on gels and we do have no proof yet that it is pure. The fingerprint suggests no major contaminations (see legend to Fig. 4), and the lack of this form in some mutants (see below) implies that if it is contaminated with non acid phosphatase proteins those contaminants have specifically to stick on inactive acid phosphatase,
Phol-mutants are Altered in the Expression of Active and Inactive Acid Phosphatase Mutationally inactivated acid phosphatase of 5 different
phol-mutants was purified by the same procedure as for the wildtype protein. Obviously, the enzymatic activity could not be followed to record recoveries. The purification was always started with roughly the same amount of cell material. The protein elution patterns from the Sepharose CL-6B column are given in Fig. 3. The lack of the inactive form (peak I) for mutant phol-270 and phol-277 is obvious. Mutant phol-38 exhibits an elution pattern almost reciprocal to that of the wildtype, low amounts of peak II - and relatively large amounts of peak I-material. For mutant phol-304 the protein
M. E. Schweingruber et al.: Mutants of Acid Phosphatase in Schizosaccharomyces pombe
i
ot
A : WT 972 h0.3
A/5 ~_i'0.2
-0.1
,
40
50
60
70
D : phol - 38
-0.2
.o.,
,~;_ 30
113
,
,
R0
00
. , L . . . . .
100
110
i
I;0
30
40
50
60
FRACTIONS
70
i
L
RO
90
100
120
]
FRACTIONS
3a
3d
f 0.3
E :phol-270
B :pho1"133
-0.3
.0.2
*0.2
.0.1
-0.1
e=
I
30
40
50
60
70
80
90
100
110
AiaTI=
1
120
1=
30
i
40
50
R0
FRACTIONS
ae
70
80
90
100
110
120
90
100
fl0
120
FRACTIONS
3b
3e
F : phol-277
C: p h o l - 3 0 4 -0,3
i0.3
=' -0.2
-0.2
30
3c
40
50
60
70
80
90
100
110
T20
30
FRACTIONS
40
50
60
70
80
FRACTIONS
3f
Fig. 3. A - F . Elution patterns of acid phosphatase protein from wildtype 972 and 5 different phol-mutants after Sepharose CL-6B cromatography. Starting with 20 g of wet cells harvested at the end of logarithmic growth, acid phosphatase was prepurified as described by Dibenedetto and Cozzani (1975). The partially purified enzyme was applied in a volume of 2 ml, to a Sepharose CL-6B column (1.5 c m x 90 cm) and eluted with 0.1 M sodium acetate buffer pH 4.0. Flow ra~e was adjusted to 3.6 ml/h. Fractions of 1,2 ml were collected and measured for absorbance at 280 nm. Peak I designates inactive and peak II represents active acid phosphatase of the wildtype 972 (see text). For tryptic peptide mapping the following fractions were pooled, dialyzed and lyophilized: 9 7 2 : 3 8 - 4 5 (I), 5 8 - 7 4 (II),phol-304:40-46 (I), 5 9 - 7 3 (II)
Fig. 4 A - D . Tryptic peptide maps of active and inactive acid phosphatase from wildtype 972 and mutant strain phoi-304. Acid phosphatase protein was digested with trypsin and the resulting peptides were separated by HPLC. The graphs represent the recorded elution profiles of peptides. The oblique line marks the phosphoric acid acetonitrile elution gradient (10%-60% acetonitrile). To facilitate comparison of the peptide pattern, the corresponding peptides or peptide groups have been designated with a number. Peptide maps A and B cannot be compared with C and D. They were generated by different column brands (A and B: Spherisorb ODS, 10 gm; C and D Lichrosorb RP18, 10 ~m). A and B: Tryptic peptide patterns of active (II) and inactive (I) acid phosphatase. The two proteins differ only in the yield of peptides 1-7. Slight differences, as observed for exampte for peptides 14 and 21 are due to run to run variations. There are no extra peptides visible in the peptide map of the inactive form that could be indicative for major contaminations
with non acid phosphatase protein. C and D: tryptic peptide patterns of active acid phosphatase from wildtype 972 and the mutants proteins of phol-304. The upper map in Fig. 4D represents the map of the peak II-protein and the lower map originates from the peak I-protein. The proteins of mutant phol-304 differ from the wildtype protein of the position indicated by an arrow. They lack in addition almost completly the peptides eluting in the hydrophoic region of the gradient. We have verified the indicated peptide changes by mixing digests of wildtype and mutant proteins (data not shown). The difference in the peptide group 1-3 for the two mutant proteins are also observed for the corresponding wildtype proteins (see A and B). Other small differences observed for some peptides are, according to our experience, not significant
116
M.E. Schweingruberet al.: Mutants of Acid Phosphatase in Sehizosaccharomycespombe
recovery seems to be reduced and for mutant pho1-133 the unusual shape of elution peak II is remarkable. To demonstrate that the isolated mutant proteins appearing in peak I and II really represent acid phosphatase protein and not some copurified contamination we fingerprinted the two proteins from mutant phol-304. As illustrated in Fig. 4 in the region between peptide 4 and 18 only acid phosphatase peptides can be detected. This is a strong indication that a least the major fraction of the isolated protein represents acid phosphatase protein. We also detect alterations in the peptide maps of mutants phol-304. In both forms peptide 9 and many of the peptides appearing in the hydrophobic region of the linear peptide map are missing. The peptide maps of the two protein forms of the mutant do as already observed for the wildtype proteins not differ significantly from each other except for the peptide group eluting in the hydrophilic region of the gradient (peptide 1-3).
Discussion
The results presented in this paper demonstrate that all the selected mutants defective in acid phosphatase activity map in one genetic locus, phol. The finding that only one locus affects acid phosphatase activity is rather surprising. In other organisms such as S. cerevisiae (Toh-e et al. 1981) and N. crassa (Nelson et al. 1976) several loci have been found to influence acid phosphatase activity. The reason for our results may be due to our selection procedure. We selected the mutants on a rich medium containing low amounts of inorganic phosphate but still containing organic phosphate whereas the mutants of S. cerevisiae (Toh-e et al. 1973) and N. crassa (Gleason and Metzberg 1974) were selected on a synthetic minimal medium containing no organic phosphate and limiting amounts of inorganic phosphate. A third of the mutants being defective in acid phosphatase activity exhibits in addition aberrant cell morphology. Aberrant cell morphology cosegregates with acid phosphatase activity defiency in all of the 42 analyzed tetrads. This is a strong indication that phol not only controls acid phosphatase activity but also cell morphology. Two mutants exhibit in addition to abnormal cell morphology also aberrant agglutination and reduced growth. In the 7 respectively 10 full tetrads analyzed both characteristics cosegregate with acid phosphatase dediciency suggesting that phol or a locus closely linked to phol also controls agglutination and growth rate. The fact that many of the mutants scored as being normal or only aberrant in cell morphology in Table 1 are also slightly altered in agglutination and growth rate is an indication that phol itself is responsible for the phenotype ofphol270 and phol-277. These two mutants probably repre-
sent two rather rare alleles of the phol-locus. To deduce the nature of the phol-locus we first examined in preliminary experiments the activity of alkaline phosphatase. For S. cerevisiae (Toh-e et al. 1981) it has been shown that some regulatory genes not only affect acid phosphatase activity but also alkaline phosphatase activity. In all tested phol-mutants however alkaline phosphatase activity was not affected (our unpublished results). The possibility that phol could be responsible for glycosylation and therefore altered expression of acid phosphatase activity was examined by measuring the activity of an other cell surface glycoprotein, invertase. Again in all tested phol-mutants invertase activity was normal (our unpublished results). Subsequently we purified acid phosphatase from the wildtype 972 and 5 different phol-mutants. The purification procedure yields for the wildtype as for S. cerevisiae (Schweingruber and Schweingruber, in press) at least two structurally related forms as judged by tryptic peptide mgpping. One is enzymatically active and represents pure activ acid phosphatase. The other is enzymatically inactive and tentatively called inactive acid phosphatase. We do have no proof yet that the inactive form is pure. All the available data so far do not suggest major contaminations. The physiological role of the inactive form is not clear yet. The three facts currently under investigation in our laboratory namely that this form is mainly membrane-bound in contrast to the active form which is soluble, that it differs from active acid phosphatase in lipid and carbohydrate content and that its amount drastically varies under different physiological conditions suggest to us that it is not a purification artefact. The elution patterns from the Sepharose CL-6B column reveal that in phol-304 the amount of both tile active and inactive form is reduced whereas in two other mutants (phol270 and phol-277) preferentially the inactive form and in a fourth mutant mainly the active form is reduced. These results together with the fingerprint data indicating similar if not identical alterations of the peptide patterns of the active and inactive form of mutant phol-304 strongly suggest that the expression of the active as well as the inactive form of acid phosphatase is under the control of phol. Our interpretation of the data is that phol codes for the protein moiety of the two forms. The different alleles probably cause structural alterations and thereby differential expression and modification of the acid phosphatase molecules. Examples how structural alterations of the protein moiety of a glycoprotein can affect its expression and modification have been given recently for 1 g M p chains (Sidman et al. 1982). This interpretation has to be verified by comparing the relevant protein or DNA-sequences of active and inactive forms of the wildtype and mutant proteins. The given interpretation implies that acid phosphatase would also be involved in the process of
M. E. Schweingruber et al.: Mutants of Acid Phosphatase in Sehizosaccharomyces pombe
morphogenesis and possibly cell agglutination and regulation of growth rate. This role for acid phosphatase has not yet been reported for yeast but is not unexpected. Acid phosphatase is one o f the glycoproteins that cover the yeast cell surface and recently it has been shown that mutants defective in the carbohydrate moiety of cell surface mannoproteins are slow growing, clumpy and altered in cell morphology (Ballou et al. 1980).
Acknowledgements. This work was supported by the Swiss National Science Foundation. We thank Miss V. Kindlimann for carefully typing the manuscript.
References Ballou L, Cohen RE, Ballou CE (1980) J Biol Chem 255:59865991 Boer P, van Rijn HI, Reinking AEP (1975) Biochim Biophys Acta 377:331-342 Dibenedetto G (1972) Biochim Biophys Acta 286:363-374 Dibenedetto G, Cozzani I (1975) Biochemistry 14:2847 2852 Dibenedetto G, Teller DC (1981) J Biol Chem 256:3926-3930
117
Gleason MK, Metzenberg RL (1974) Genetics 78:645-659 Gtinther T, Kattner W (1968) Z Naturforsch 236:77-80 Leupold U (1970) Genetical methods for Schizosaccharomyces pombe. In: Prescott DM (ed) Methods in cell physiology, vol 4. Academic Press Inc, New York, pp 169-177 Linnemans WA, Boer P, Elber PF (1977) J Bacteriol 131: 638-644 Miyata M, Miyata H (1978) J Bacteriol 136:558-563 Nelson RE, Lehman JF, Metzenberg R (1976) Genetics 84: 183 192 Van Rijn HJM, Linnemans WAM, Boer P (1975) J Bacteriol 123:1144-1149 Schweingruber ME, Schweingruber AM (1981) Differentiation 19:68-70 Schweingruber AM, Schweingruber ME (1982) Biochim Biophys Acta (in press) Sidman C, Potash MJ, K6hler G (1981) J Binl Chem 256: 13180-13187 Toh-e A, Inouye S, Oshima Y (1981) J Bacteriol 145:221-232 Toh-c A, Ueda Y, Kakimoto S, Oshima Y (1973) J Bacteriol 113:727-738
Communicated by F. Kaudewitz Received March 22, 1982
Current Genetics (1982) 5:109-117
© Springer-Verlag 1982
Isolation and Characterization of Acid Phosphatase Mutants in Schizosaccharomycespombe M. E. Schweingruber I , A. M. Schweingruber 1 , and M. E. Sch@bach 2 1 Institute for General Microbiology,Baltzer-Strasse4, CH-3012 Bern, Switzerland 2 Biologicaland Pharmaceutical Research Department, F. Hoffmann-LaRoche and Co. Ltd., CH4002 Basel, Switzerland
Summary. 72 mutants defective in the activity of the cell surface glycoprotein acid phosphatase were isolated and characterized. The mutants map in one genetic locus, phol. Many of them exhibit altered cell morphology. This characteristic cosegregates with acid phosphatase deficiency, implying that phol controls the activity of acid phosphatase and concomitantly cell morphology. Phol probably also influences growth rate and agglutination behaviour. By purifying acid phosphatase, two structurally related forms can be detected. One is inactive (form I) and one is the active acid phosphatase (form II). Mutant phol-270 and phol-277 lack the inactive form I. Mutant phol-38 exhibits mainly form I, form II being present only in minor amounts. Two other mutants examined differ only slightly from wildtype in their pattern of active and inactive forms. Tryptic peptide maps of the inactive and active acid phosphatase of the wildtype and the corresponding proteins of mutant phol304 reveal similar structural alterations for the two mutant proteins. The results show that phol controls the expression of the active and inactive acid phosphatase. We conclude that phol represents the structural gene of the two forms of acid phosphatase. Key words: Schizosaccharomyces pombe - Active acid phosphatase - Inactive acid phosphatase - Cell-surface glycoprotein
Introduction
To understand the genetic control and biological role of a specific cell surface glycoprotein we are investigating
Offprint requeststo: M. E. Schweingruber
acid phosphatase of yeast. Acid phosphatase of Saccharomyces cerevisiae is a mannoprotein and is mainly located in the cell wall (Linnemans et al. 1977). Minor amounts have also been found in close association with the plasmamembrane and other cell organelles (Van Rijn et al. 1975; Boer et al. 1975). We have recently shown that the enzyme can also exist in an inactive probably membrane-bound form (Schweingruber und Schweingmbet, in press). Little is known about the biological role of acid phosphatase. One function may be the hydrolysis of external phosphate esters which do not penetrate the protoplasmic membrane (G~ther and Kaffner 1968). We speculated that acid phosphatase is also involved in morphogenetic and behavioural differentiation of the yeast cell (Schweingruber and Schweingruber 1981). To elucidate its biological role we are investigating acid phosphatase of the fission yeast Schizosaccharomyces pornbe. Acid phosphatase of S. pombe is, like the corresponding enzyme of the budding yeast, a cell surface glycoprotein regulated by extracellular orthophosphate (Dibenedetto 1972). It has been purified to homogeneity and characterized (Dibenedetto and Cozzani 1975; Dibenedetto and Teller 1981). The active protein has a molecular weight of roughly 360,000 and can be dissociated into subunits with a molecular weight of about 90,000. Neutral sugars account for about 66% of the total molecular weight. Mannose and galactose are found as the main components of the carbohydrate moiety. Since its synthesis is tightly associated with cell elongation, it has been considered as a growth marker (Miyata and Miyata 1978). In this communication we report the isolation and characterization of mutants lacking acid posphatase activity. 0172-8083/0005/0109/$ O1.80
M.E. Schweingruber et al.: Mutants of Acid Phosphatase in Sehizosaccharomycespombe
110
Table 1. Properties of phol-mutants lacking acid phosphatase activity Class
I
Strain number
Morphology
Agglutination
Growth rate
38, 44, 86, 102 150,157,171,173 124,182,185,187 202,212,222,169
normal
normal
normal
PepticIe Mapping by High Pressure Liquid Chromatography (HPLC) Tryptic digests were prepared as follows: 1-2 mg of phosphatase protein were dissolved in 0.5 ml ammonium bicarbonate (2 mg/ml), heated for 5 rain at 100 °C and subsequerrtly digested with 2% (by weight of protein) trypsin (TPCK-Trypsin, Boehringer) for 16 h at 30 °C. Approximately 50 ug of the digest were applied to a 25 cm x 0.5 cm Spherisorb ODS 10 ~z-column and eluted with an orthophosphate-acetonitrile gradient by HPLC as described in detail elsewhere (Schweingruber and Schweingruber, in press). The peptides were monitored at 210 nm.
226,230,232,233 240,243,149,151 184,219,231,234 244,256,260,261 262,266,267,269 272,281,282,283
283,291,292,302 307,308,310 II
phatase (see Fig. 3 ) we observed an other protein peak in the last purification step. The appearence of this peak has not been reported by Dibenedetto and Cozzani (1975). All available data (see Results) indicate that this peak represents enzymatically inactive acid phosphatase.
94,118,133,144 189,181,238,242 247,248,168,236 251,258,264,271 274,278,284,294 304,203
altered
normal
normal
III
270
altered
clumpy
normal
IV
277
altered
normal
altered
Isolation and Mapping of the phol Mutants Isolation. A stationary phase culture of wildtype 972, grown in
The mutants listed in the table were selected on the basis of their lack of acid phosphatase activity in colonies on plates. Quantitative measurements revealed, for all the mutants, less than 2% of the wildtype activity.Morphology, agglutination and growth rates were examined form cells grown at 30 °C in low phosphate medium. Morphology was examined under the microscope (see Fig. 1), agglutination was detected macroscopically (flocking cultures) and microscopically (see Fig. 1), g o w t h rates were determined as given in Fig. 2. Ceils exhibiting morphology, agglutination behaviour and growth rate only slightly different from wild-type were scored as normal
Materials a n d M e t h o d s
Growth Conditions and Purification of Acid Phosphatase Cells were grown in 21-Erlenmeyers containing 11YEPD-medium from which inorganic phosphate had been removed, referred to as low phosphate medium (Schweingruber and Schweingruber 1981), at 30 °C on a rotary shaker (Pilot Shake RC2) at a rotation speed of 120 rpm. At the end of the exponential phase, cells were harvested and washed in 0.1 M sodium acetate buffer. The yield was around 10 g wet cells/1 medium. Acid phosphatase was purified essentially according to the method of Dibenedetto and Cozzani (1975). This method has been shown to yield pure acid phosphatase as ascertained from ultracentrifugation, electrophoretic, chromatographic and endgroup determination data (Dibenedetto and Cozzani 1975; Dibenedetto and Teller 1981). We could confirm purity of the enzyme. By gelelectrophoresis under non denaturing conditions the purified enzyme yielded a single protein band that exhibited phosphatase activity as reported and by running the enzyme on SDS-gels only a single protein band was observed. In addition to peak II representing active acid phos-
yeast extract at 30°C, was centrifuged and washed twice in 0.9% NaC1. A suspension of 107 cells per ml was irradiated with 254 nm UV-light with a dose of 210,000 ergs x cm - 2 , resulting in 30-40% survival in the first experiments (isolates No. 1-90) and with a dose of 378,000 ergs x cm - 2 resulting in 1-9% survival for the other experiments (isolates Nr. 91-312). The irradiated cells were plated in an appropriate dilution in NaC1 solution on low phosphate medium and incubated at 25 °C. After 5 days incubation, the colonies were overlaid with 5 ml of the following mixture: 5 mg c~-naphthylphosphate (Na-Salt, SERVA), 50 mg Fast Blue B salt (SERVA) in 32 ml of 0.05 M sodium acetate buffer, pH 4.8, containing 1% Difco Bacto agar. Enzyme activity is indicated by a red staining of the colony. Plates were scored 5 to 10 rain after plating of the overlay. Before putting into stock, the isolates were restreaked and checked again for phosphatase activity. In addition, the absence of acid phosphatase activity in presumptive phol strains was confirmed by assaying cells that had been grown in low phosphate liquid medium, by the nitrophenylphosphate assay (Schweingruber and Schweingruber 1981).
Mapping. The isolated mutants, all of the h - mating type, were corssed into the h + mating type. For a genetic classification, a random analysis of ascospores, resulting from crosses phol h- x pho] h + on malt extract slants, was performed according to Leupold (1970). The spores were plated in appropriate dilutions on low phosphate medium plates, incubated at 25 °C and the colonies were then overlaid with c~-naphthylphosphate and Fast Blue B as described.
Results Mutants Lacking Acid Phosphatase Activity Exhibit Altered Morphology, Agglutination and Growth Rate Out o f roughly 100,000 screened colonies, we isolated 72 m u t a n t s w h i c h are deficient in acid p h o s p h a t a s e activity w h e n t e s t e d w i t h n i t r o p h e n y l p h o s p h a t e as substrate. The m u t a n t strains were e x a m i n e d for p r o p e r t i e s such as cell m o r p h o l o g y , agglutination b e h a v i o u r and
M. E. Schweingruber et al.: Mutants of Acid Phosphatase in Sehizosaccharomycespornbe
l 11
Fig. 1. A-D. Morphology of cells of wildtype 972 and 3 different phol-mutants, Strains were grown in low phosphate medium and examined under the microscope at a 400 fold magnification and photographed. The 3 phol-mutants are morphological aberrant. Phol-270 is, in addition, clumpy'
10 B
10
r
F
..a
growth rate. The results are summarized in Table 1. The mutants can be roughly divided into 4 classes. Apart from their lack of acid phosphatase activity, mutants of class I do not substantially differ from wild-type. Mutants of class II exhibit aberrant cell morphology, while the mutant of class III in addition clumps strongly. The class IV-mutant exhibits altered cell morphology and a drastically reduced growth rate. The picture of wildtype cells and of cells from a representative mutant of each of class II, III and IV are given in Fig. 1. The growth curve for the wildtype and forphol-277 is given in Fig. 2. It has to be stressed that the classification of
i=o
_l.i~
~-~
t (MIN)
1D~
~l Fig. 2. Growth curve of wildtype 972 and phol-277. Cells were grown in low phosphate medium. Growth was monitored by determining at intervals the titer of viable (colony forming) cells (=, e) and by measuring the turbidity of the culture at 530 nm (o, o). Both methods yielded roughly the same growth curve. The generation time of logarithmic phase cells is, under the given experimental conditions approximately 120 rain for the wildtype (open symbols) and 330 min for mutant phol-277 (closed symbols)
M.E. Schweingruber et al.: Mutants of Acid Phosphatase in Sehizosaeeharornyeespombe
112
Table 2. Tetrad analysis of different phol-mutants Strain number
Phenotyp (in addition to acid phosphatase deficiency)
Full tetrads analyzed
phol-187 phol-232 phol-244 phol-133 phol-247 phol-304 phol-270 phol-277
normal normal normal aberrant aberrant aberrant aberrant aberrant growing
13 4 10 6 11 8 7 10
cell morphology cell morphology cell morphology cell morphology, clumpy cell morphoiogy, slow
The phol-mutants (h-) listed in the table were crossed with the wildtype 975 h + and the progeny cells of the 4 spores of each tetrad were examined for acid phosphatase activity and if necessary for abberrant cell morphology agglutination behaviour and growth rate. In all cases a 2:2 segregation for wildtype and mutant phosphatase activity was observed. Aberrant cell morphology, agglutination behaviour and slow growth always cosegregated with acid phosphatase deficiency. Only the number of analyzed tetrads yielding 4 viable spores are given in the table. For mutant phol-277 spore germination was drastically reduced, less than 25% of the dissected asci yielded 4 colonies.
the mutants is arbitrary. Many mutants of class I do slightly differ form wildtype in morphology, agglutination and growth rate but the effects are rather slight while most mutants of class II, III and IV exhibit in addition to altered cell morphology, slight alterations in growth rate and agglutination behaviour. Many of the mutant strains of class II, III and IV are genetically unstable. This is indicated by the relatively high frequency of sectored colonies after staining for phosphatase activity. In many of the strains sporulation is impaired. Representative mutant strains of each class were crossed to the wildtype, and progeny cells of the spores of tetrads were examined for acid phosphatase acitivity, cell morphology, growth rate and agglutination. The strains and the number of full tetrads analyzed are given in Table 2. In all tetrads analyzed there was a 2 : 2 segregation for wildtype and acid phosphatase deficient phenotype. Aberrant cell morphology, agglutination behaviour and reduced growth rate cosegregated with the deficiency in phosphatase activity.
All the Mutants Deficient in Acid Phosphatase Activity Map Genetically at the Same Locus, phol All the mutants were crossed to 3 tester pho-mutants, which had been shown in preliminary experiments to belong to the same locus or at least to be closely linked. 5 0 - 5 0 0 meiotic progeny clones were examined for phosphatase activity. In no instance did we find colo-
nies containing phosphatase activity. This indicates that all mutants given in Table 1 map in the same genetic locus called phol.
Purification and Identification of an Inactive Form of Acid Phosphatase in Wildtype A method for the purification of acid phosphatase yielding a pure enzyme preparation has been published (Dibenedetto and Cozzani 1975; Dibenedetto and Teller 1981). Using this purification procedure we find that in the last step of the purification procedure - chromatography on a Sepharose CL-6B column - two peaks elute (Fig. 3). Peak II represents pure active acid phosphatase (see Material and Methods). The material of peak I is enzymatically inactive. For S. cerevisiae we have given evidence that the material of peak I represents inactive acid phosphatase (Schweingruber and Schweingruber, in press). A comparison of the tryptic peptide maps of active acid phosphatase and of the material eluting as peak I indicates that peak I of S. pombe also contains substantial amounts of inactive acid phosphatase. As shown in Fig. 4A and 4B the material of peak I reveals many peptides identical of those of active acid phosphatase (for details see legend to Fig. 4). We therefore call the material of peak I tentatively inactive acid phosphatase. Preliminary experiments reveal that the structure of the inactive form is more complex than that of the active form. In addition to carbohydrates it also contains lipids. This seems to impede its analysis on gels and we do have no proof yet that it is pure. The fingerprint suggests no major contaminations (see legend to Fig. 4), and the lack of this form in some mutants (see below) implies that if it is contaminated with non acid phosphatase proteins those contaminants have specifically to stick on inactive acid phosphatase,
Phol-mutants are Altered in the Expression of Active and Inactive Acid Phosphatase Mutationally inactivated acid phosphatase of 5 different
phol-mutants was purified by the same procedure as for the wildtype protein. Obviously, the enzymatic activity could not be followed to record recoveries. The purification was always started with roughly the same amount of cell material. The protein elution patterns from the Sepharose CL-6B column are given in Fig. 3. The lack of the inactive form (peak I) for mutant phol-270 and phol-277 is obvious. Mutant phol-38 exhibits an elution pattern almost reciprocal to that of the wildtype, low amounts of peak II - and relatively large amounts of peak I-material. For mutant phol-304 the protein
M. E. Schweingruber et al.: Mutants of Acid Phosphatase in Schizosaccharomyces pombe
i
ot
A : WT 972 h0.3
A/5 ~_i'0.2
-0.1
,
40
50
60
70
D : phol - 38
-0.2
.o.,
,~;_ 30
113
,
,
R0
00
. , L . . . . .
100
110
i
I;0
30
40
50
60
FRACTIONS
70
i
L
RO
90
100
120
]
FRACTIONS
3a
3d
f 0.3
E :phol-270
B :pho1"133
-0.3
.0.2
*0.2
.0.1
-0.1
e=
I
30
40
50
60
70
80
90
100
110
AiaTI=
1
120
1=
30
i
40
50
R0
FRACTIONS
ae
70
80
90
100
110
120
90
100
fl0
120
FRACTIONS
3b
3e
F : phol-277
C: p h o l - 3 0 4 -0,3
i0.3
=' -0.2
-0.2
30
3c
40
50
60
70
80
90
100
110
T20
30
FRACTIONS
40
50
60
70
80
FRACTIONS
3f
Fig. 3. A - F . Elution patterns of acid phosphatase protein from wildtype 972 and 5 different phol-mutants after Sepharose CL-6B cromatography. Starting with 20 g of wet cells harvested at the end of logarithmic growth, acid phosphatase was prepurified as described by Dibenedetto and Cozzani (1975). The partially purified enzyme was applied in a volume of 2 ml, to a Sepharose CL-6B column (1.5 c m x 90 cm) and eluted with 0.1 M sodium acetate buffer pH 4.0. Flow ra~e was adjusted to 3.6 ml/h. Fractions of 1,2 ml were collected and measured for absorbance at 280 nm. Peak I designates inactive and peak II represents active acid phosphatase of the wildtype 972 (see text). For tryptic peptide mapping the following fractions were pooled, dialyzed and lyophilized: 9 7 2 : 3 8 - 4 5 (I), 5 8 - 7 4 (II),phol-304:40-46 (I), 5 9 - 7 3 (II)
Fig. 4 A - D . Tryptic peptide maps of active and inactive acid phosphatase from wildtype 972 and mutant strain phoi-304. Acid phosphatase protein was digested with trypsin and the resulting peptides were separated by HPLC. The graphs represent the recorded elution profiles of peptides. The oblique line marks the phosphoric acid acetonitrile elution gradient (10%-60% acetonitrile). To facilitate comparison of the peptide pattern, the corresponding peptides or peptide groups have been designated with a number. Peptide maps A and B cannot be compared with C and D. They were generated by different column brands (A and B: Spherisorb ODS, 10 gm; C and D Lichrosorb RP18, 10 ~m). A and B: Tryptic peptide patterns of active (II) and inactive (I) acid phosphatase. The two proteins differ only in the yield of peptides 1-7. Slight differences, as observed for exampte for peptides 14 and 21 are due to run to run variations. There are no extra peptides visible in the peptide map of the inactive form that could be indicative for major contaminations
with non acid phosphatase protein. C and D: tryptic peptide patterns of active acid phosphatase from wildtype 972 and the mutants proteins of phol-304. The upper map in Fig. 4D represents the map of the peak II-protein and the lower map originates from the peak I-protein. The proteins of mutant phol-304 differ from the wildtype protein of the position indicated by an arrow. They lack in addition almost completly the peptides eluting in the hydrophoic region of the gradient. We have verified the indicated peptide changes by mixing digests of wildtype and mutant proteins (data not shown). The difference in the peptide group 1-3 for the two mutant proteins are also observed for the corresponding wildtype proteins (see A and B). Other small differences observed for some peptides are, according to our experience, not significant
116
M.E. Schweingruberet al.: Mutants of Acid Phosphatase in Sehizosaccharomycespombe
recovery seems to be reduced and for mutant pho1-133 the unusual shape of elution peak II is remarkable. To demonstrate that the isolated mutant proteins appearing in peak I and II really represent acid phosphatase protein and not some copurified contamination we fingerprinted the two proteins from mutant phol-304. As illustrated in Fig. 4 in the region between peptide 4 and 18 only acid phosphatase peptides can be detected. This is a strong indication that a least the major fraction of the isolated protein represents acid phosphatase protein. We also detect alterations in the peptide maps of mutants phol-304. In both forms peptide 9 and many of the peptides appearing in the hydrophobic region of the linear peptide map are missing. The peptide maps of the two protein forms of the mutant do as already observed for the wildtype proteins not differ significantly from each other except for the peptide group eluting in the hydrophilic region of the gradient (peptide 1-3).
Discussion
The results presented in this paper demonstrate that all the selected mutants defective in acid phosphatase activity map in one genetic locus, phol. The finding that only one locus affects acid phosphatase activity is rather surprising. In other organisms such as S. cerevisiae (Toh-e et al. 1981) and N. crassa (Nelson et al. 1976) several loci have been found to influence acid phosphatase activity. The reason for our results may be due to our selection procedure. We selected the mutants on a rich medium containing low amounts of inorganic phosphate but still containing organic phosphate whereas the mutants of S. cerevisiae (Toh-e et al. 1973) and N. crassa (Gleason and Metzberg 1974) were selected on a synthetic minimal medium containing no organic phosphate and limiting amounts of inorganic phosphate. A third of the mutants being defective in acid phosphatase activity exhibits in addition aberrant cell morphology. Aberrant cell morphology cosegregates with acid phosphatase activity defiency in all of the 42 analyzed tetrads. This is a strong indication that phol not only controls acid phosphatase activity but also cell morphology. Two mutants exhibit in addition to abnormal cell morphology also aberrant agglutination and reduced growth. In the 7 respectively 10 full tetrads analyzed both characteristics cosegregate with acid phosphatase dediciency suggesting that phol or a locus closely linked to phol also controls agglutination and growth rate. The fact that many of the mutants scored as being normal or only aberrant in cell morphology in Table 1 are also slightly altered in agglutination and growth rate is an indication that phol itself is responsible for the phenotype ofphol270 and phol-277. These two mutants probably repre-
sent two rather rare alleles of the phol-locus. To deduce the nature of the phol-locus we first examined in preliminary experiments the activity of alkaline phosphatase. For S. cerevisiae (Toh-e et al. 1981) it has been shown that some regulatory genes not only affect acid phosphatase activity but also alkaline phosphatase activity. In all tested phol-mutants however alkaline phosphatase activity was not affected (our unpublished results). The possibility that phol could be responsible for glycosylation and therefore altered expression of acid phosphatase activity was examined by measuring the activity of an other cell surface glycoprotein, invertase. Again in all tested phol-mutants invertase activity was normal (our unpublished results). Subsequently we purified acid phosphatase from the wildtype 972 and 5 different phol-mutants. The purification procedure yields for the wildtype as for S. cerevisiae (Schweingruber and Schweingruber, in press) at least two structurally related forms as judged by tryptic peptide mgpping. One is enzymatically active and represents pure activ acid phosphatase. The other is enzymatically inactive and tentatively called inactive acid phosphatase. We do have no proof yet that the inactive form is pure. All the available data so far do not suggest major contaminations. The physiological role of the inactive form is not clear yet. The three facts currently under investigation in our laboratory namely that this form is mainly membrane-bound in contrast to the active form which is soluble, that it differs from active acid phosphatase in lipid and carbohydrate content and that its amount drastically varies under different physiological conditions suggest to us that it is not a purification artefact. The elution patterns from the Sepharose CL-6B column reveal that in phol-304 the amount of both tile active and inactive form is reduced whereas in two other mutants (phol270 and phol-277) preferentially the inactive form and in a fourth mutant mainly the active form is reduced. These results together with the fingerprint data indicating similar if not identical alterations of the peptide patterns of the active and inactive form of mutant phol-304 strongly suggest that the expression of the active as well as the inactive form of acid phosphatase is under the control of phol. Our interpretation of the data is that phol codes for the protein moiety of the two forms. The different alleles probably cause structural alterations and thereby differential expression and modification of the acid phosphatase molecules. Examples how structural alterations of the protein moiety of a glycoprotein can affect its expression and modification have been given recently for 1 g M p chains (Sidman et al. 1982). This interpretation has to be verified by comparing the relevant protein or DNA-sequences of active and inactive forms of the wildtype and mutant proteins. The given interpretation implies that acid phosphatase would also be involved in the process of
M. E. Schweingruber et al.: Mutants of Acid Phosphatase in Sehizosaccharomyces pombe
morphogenesis and possibly cell agglutination and regulation of growth rate. This role for acid phosphatase has not yet been reported for yeast but is not unexpected. Acid phosphatase is one o f the glycoproteins that cover the yeast cell surface and recently it has been shown that mutants defective in the carbohydrate moiety of cell surface mannoproteins are slow growing, clumpy and altered in cell morphology (Ballou et al. 1980).
Acknowledgements. This work was supported by the Swiss National Science Foundation. We thank Miss V. Kindlimann for carefully typing the manuscript.
References Ballou L, Cohen RE, Ballou CE (1980) J Biol Chem 255:59865991 Boer P, van Rijn HI, Reinking AEP (1975) Biochim Biophys Acta 377:331-342 Dibenedetto G (1972) Biochim Biophys Acta 286:363-374 Dibenedetto G, Cozzani I (1975) Biochemistry 14:2847 2852 Dibenedetto G, Teller DC (1981) J Biol Chem 256:3926-3930
117
Gleason MK, Metzenberg RL (1974) Genetics 78:645-659 Gtinther T, Kattner W (1968) Z Naturforsch 236:77-80 Leupold U (1970) Genetical methods for Schizosaccharomyces pombe. In: Prescott DM (ed) Methods in cell physiology, vol 4. Academic Press Inc, New York, pp 169-177 Linnemans WA, Boer P, Elber PF (1977) J Bacteriol 131: 638-644 Miyata M, Miyata H (1978) J Bacteriol 136:558-563 Nelson RE, Lehman JF, Metzenberg R (1976) Genetics 84: 183 192 Van Rijn HJM, Linnemans WAM, Boer P (1975) J Bacteriol 123:1144-1149 Schweingruber ME, Schweingruber AM (1981) Differentiation 19:68-70 Schweingruber AM, Schweingruber ME (1982) Biochim Biophys Acta (in press) Sidman C, Potash MJ, K6hler G (1981) J Binl Chem 256: 13180-13187 Toh-e A, Inouye S, Oshima Y (1981) J Bacteriol 145:221-232 Toh-c A, Ueda Y, Kakimoto S, Oshima Y (1973) J Bacteriol 113:727-738
Communicated by F. Kaudewitz Received March 22, 1982
