Properties of Repressible Alkaline Phosphatase from Wild Type and a Wall-less Mutant of Neurospora crassa”

SUMMARY The repressible alkaline phosphatase of Neurospora crassu was purified from both the mycelium of a wild type strain and from the medium in which cultures of the slime mutant (which lacks the normal cell wall) had been grown. The enzyme preparations from the two sources had similar amino acid compositions, immunological properties, specific activities, thermal stabilities, and kinetic constants, but differed in a number of other properties. Both enzyme preparations contained carbohydrate, but the carbohydrate content of the enzyme isolated from slime medium was almost double that of the enzyme from wild type mycelium (24 and 14 %, respec-tively). The molecular weight of the enzyme secreted by slime cells, estimated by gel filtration on Sephadex G-200 and polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate, was higher than that of the enzyme from wild type mycelium by an amount consistent with its increased carbohydrate content. Electrophoresis at pH 4.7 and 9.5 indicated that the enzyme isolated from slime is more anionic than the enzyme from wild type mycelium. analysis

The repressible alkaline phosphatase of Neurospora crassu was purified from both the mycelium of a wild type strain and from the medium in which cultures of the slime mutant (which lacks the normal cell wall) had been grown. The enzyme preparations from the two sources had similar amino acid compositions, immunological properties, specific activities, thermal stabilities, and kinetic constants, but differed in a number of other properties. Both enzyme preparations contained carbohydrate, but the carbohydrate content of the enzyme isolated from slime medium was almost double that of the enzyme from wild type mycelium (24 and 14 %, respectively). The molecular weight of the enzyme secreted by slime cells, estimated by gel filtration on Sephadex G-200 and polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate, was higher than that of the enzyme from wild type mycelium by an amount consistent with its increased carbohydrate content.
Electrophoresis at pH 4.7 and 9.5 indicated that the enzyme isolated from slime medium is more anionic than the enzyme from wild type mycelium. Chemical analysis revealed the presence of approximately 8 phosphate groups per enzyme molecule in the purified slime extracellular enzyme, whereas the wild type enzyme contained less than 0.5 phosphate molecule per enzyme molecule. The presence of phosphate in the slime extracellular enzyme, and the lack of significant amounts of phosphate in the wild type mycelial enzyme, was also demonstrated by determination of a2P associated with the enzymes isolated from the two sources following derepression in the presence of 82P04a-.
A significant portion of the repressible alkaline phosphatase produced by derepressed wild type N. crassa was found to be secreted into the growth medium. The electrophoretic mobility of the enzyme isolated from the wild type culture medium resembled that of the enzyme isolated from the slime culture medium rather than that of the enzyme isolated from wild type mycelium.
In Neurospora crassa a number of enzymes are derepressed by phosphorus starvation or by growth on a limiting phosphorus * This work was supported by National Institutes of Health Grant GM-08995.
$ Supported by a National Institutes of Health Career Development Award during most of the course of this research.
source. These include an alkaline phosphatase (1)) an acid phosphatase (Z), a phosphate permease which has a high affinity for phosphate at high pH (3,4), and one or more extracellular nucleases (5). The repressible alkaline phosphatase of N. ~ras~a is a particularly attractive subject for the study of control of protein synthesis in a eukaryotic organism since its range of activity from full repression to full derepression is over lOOO-fold (1,4). In addition, it is easily and specifically assayed even in crude extracts containing other phosphatases (l), it is readily purified in good yield (6), and the purified enzyme has been characterized by physical and chemical methods (7).
Recent studies have concentrated on the isolation and characterization of possible structural gene mutants' (8) and of regulatory mutants altered in the ability to repress or derepress (4,8) the repressible alkaline phosphatase and the physiologically related enzymes mentioned above. Further progress in understanding such mutants requires a knowledge of the physiological factors involved in derepression of the alkaline phosphatase and, ultimately, a knowledge of the actual mechanism of derepression. With this in mind we began a study of the kinetics of derepression in the slime mutant of N. crassa. This strain lacks the normal cell wall and grows as isolated multinucleate protoplasts rather than branching hyphae (9, 10). This fact makes it much easier to manipulate in kinetic studies than the wild type strain. It quickly became apparent that, whereas the wild type strain retains most of the enzyme in a cell-bound form, slime cultures secrete nearly all of their repressible alkaline phosphatase into the growth medium.
Preliminary examination of the enzyme secreted by slime cells revealed a clear difference between it and the enzyme prepared from wild type mycelia.
In the present paper we describe the purification of the repressible alkaline phosphatase secreted into the growth medium by slime cultures and compare some of its physical and chemical properties with those of the enzyme purified from wild type my-Celia. A number of distinct differences were found between the two purified enzyme preparations. jringens), alkaline phosphatase (type III, Escherichia co&), temperature program, 2" per min; detector temperature, 298"; m-amylase (type I-A, hog pancreas), phosphorylase a (rabbit injector temperature, 280"; carrier gas (helium), 115 cc per min; muscle), and mucin (type I, bovine submaxillary gland) were hydrogen (to detector), 35 cc per min; oxygen (to detector), 300 purchased from Sigma Chemical Co. Sephadex G-200, DEAE-cc per min; electrometer range, 10; electrometer attenuation, 16. Sephadex, and CM-Sephadex were obtained from Pharmacia Fine Samples containing approximately 1 mg of alkaline phosphatase Chemicals.
Reagents for acrylamide gel electrophoresis and AG in distilled water were prepared and hydrolyzed in 0.01 N HCl in 5OW-X8 cation exchange resin were purchased from Bio-Rad; the presence of Dowex 50-H+ as described by Lehnhardt and aniline blue-black from Canalco; p-nitrophenyl phosphate from Winzler (17). An internal standard was provided by the addition Calbiochem; disodium EDTA from Fisher Chemical Co.; n-galacof 0.10 ml of 2.1 mM n-arabinose and the neutral sugar fraction tose and n-arabinose from Pfanstiehl Laboratories; sodium dowas then isolated, reduced with NaBH4, and acetylated as dedecyl sulfate from British Drug House Chemicals Ltd.; deoxyriboscribed (17 arginine-l, crisp-i, aurescent, osmotic-l (jz, sg, arg-1, cr-1, aur, os-1) known as slime. The slime was obtained as a heterocaryon (FGSC No. 327) and the slime component of the heterocaryon was resolved as described by Emerson (9).
Stock cultures, except for the slime strain, were maintained on slants of Fries' medium (11) sunnlemented with 1.5% sucrose and ~ solidified with 1.5yo Difco agar. --The slime culture w&maintained for the duration of this study by daily transfers on Fries' medium supplemented with 1.5yo sucrose and 1 mM arginine.
A 0.2-ml aliquot of a 24-hour culture was taken each day to inoculate 100 ml of fresh medium in a 50%ml Erlenmeyer flask. Large cultures for enzyme purification were grown in 3-liter Fernbach flasks containing 800 ml of medium each. The cultures were incubated at 30" in a rotary shaker-incubator with shaking at 90 rpm. Cultures, except slime, for enzyme purification or derepression studies were grown by inoculation of fresh medium with washed conidia suspended in distilled water. When derepressed cultures were desired, the KHzPOd (7.35 mM) normally included in Fries medium was replaced by an equivalent amount of KC1 and KHzPOa, if desired, was added to give the molarity indicated. Enzyme Assay-The repressible alkaline phosphatase was assaved essentiallv as described bv Nvc et al. (1) bv incubating the enzyme at 37""with a 5 mM solution-of p-nitrophenylphosphate in 0.3 M glycine buffer, pH 9.0, containing 1 mM EDTA.
The final volume was 1.0 ml. The reaction was stopped by the addition of 0.5 ml of 2 N KOH in 90% ethanol (12) and the p-nitrophenol released was measured by its absorbance at 405 nm. One unit of enzyme activity is the quantity of enzyme which releases 1 pmole of p-nitrophenol per min under the specified conditions. Specific activity is defined as enzyme units per mg of protein.
Protein was routinely measured by the method of Lowry et al. (13) using bovine serum albumin as the standard.
Protein in solutions of purified enzyme was also determined by its absorbance at 280 nm. One unit of protein (absorbance unit) based on this parameter was the amount of pure enzyme in 1 ml of a solution with an absorbance of 1 at 280 nm (l-cm light path). Amino Acid Composition-Amino acid analyses were performed with the Beckman model 120 automatic amino acid analyzer using a single column procedure.
Samples of purified alkaline phosphatase were dialyzed against four changes, 400 volumes each, of distilled water for 48 hours. Aliquots containing approximately 0.5 mg of protein (0.5 A) were distributed in Pyrex tubes and evap--orated over solid KOH.in an evacuated desiccator. The samples were hvdrolvzed at 110" in 0.75 ml of 6 N HCl (triple distilled) in evacuaied, sealed tubes. The hydrolysate was ekaporated over solid KOH in vacua. Treatment of samples with performic acid, prior to hydrolysis, for the determination of cysteine and cystine as cysteic acid was performed as described by Moore (14). Tryptophan was estimated as described by Bencze and Schmid (15) from the ultraviolet absorption spectrum of purified enzyme in 0.1 N NaOH.
Carbohydrate Composition-Neutral sugars were analyzed by gas-liquid chromatography of the alditol acetates. A Nuclear Chicago gas chromatograph equipped with flame ionization detector was used. A column packed with 1% ECNSS-M on 60/80 mesh Gas-chrom Q (16) was used. The conditions for chromatography were: column temperature, 135" initial and 175" final; (18).
Electrophoresis-Disc electrophoresis was carried out using either the alkaline buffer system (pH 9.5) of Tamura and Ui (19), or an acidic buffer system similar to that described by Reisfeld et al. (20). The latter system was modified by raising the pH of the running gel, stacking gel, and electrode buffer 0. 4 unit each (i.e. 4.7, 7.2, and 4.9, respectively) and substituting r-aminobutyric acid (pK, = 4.05) for p-alanine (pK, = 3.6) in the electrode buffer. Running gels containing 5yo acrylamide were used with either buffer system and electrophoresis was carried out at 4' using a constant current of 2.5 ma per tube for 2.5 hours.
Bands of enzyme activitv were detected following electrophoresis by incubating the gels fn a 0.05yo (w/v) solution of 5-bromo-4chloro-3-indolyl phosphate in 0.10 M barbital buffer, pH 8.5, which also contained 0.01 M EDTA.
Staining was terminated by washing with several changes of distilled water and the washed gels were stored in distilled water at 4". Gels were stained for protein by immersing them for 45 min in a solution containing 0.0597e (w/v) of both aniline blue-black and Coomassie blue in methanol-acetic acid-H20 (10: 10:80). Excess dye was removed with the methanol-acetic acid solvent. Samples were prepared for sodium dodecyl sulfate disc electrophoresis by mixing-with "sample buffer" consisting of 10 mM sodium ohosnhate buffer (nH 7.2). 0.14 M 2-mercaotoethanol. 0.25 1 1 M sucrose, and 0.1% (w&) sodium dodecyl sulfate, and 'then heating 2 min in a boiling water bath. The rest of the electrophoretic procedure was exactly as described by Laemmli (21). The acrylamide concentration used was 10%.

Estimation oj Molecular Weight by SDK-Acrylamide
Gel Electrophoresis-SDS-acrylamide gel electrophoresis was carried out as described above. The desired N. crassa alkaline phosphatase was prepared in sample buffer and the denatured protein was subjected to electrophoresis along with marker peptides derived from phosphorylase a (mol wt = 94,000), bovine serum albumin (mol wt = 67,000), a-amylase (mol wt = 48,000), and E. coli alkaline phosphatase (mol wt = 40,000). The gels were stained for protein as described earlier and the mobility of each band was calculated as suggested by Weber and Osborn (22). Estimation of Molecular Weight by Gel Filtration-Analytical gel filtration of the purified alkaline phosphatases was carried out as suggested by Andrews (23) using a column (1.5 X 82 cm) of Sephadex G-200 equilibrated with 0.05 M Tris-HCl buffer, pH 8.3, containing 0.1 M NaCl.
Elution was conducted with the same buffer. Samples were applied in a volume of 1.5 ml and fractions of 3.0 f 0.05 ml were collected at a flow rate of approximately 12 ml per hour. Bovine serum albumin (mol wt = 67,000), serum albumin dimer (mol wt = 134.000). and yeast alcohol dehydrogenase (mol wt -= 150,000) were mixed with the desired alkaline phosphatase sample prior to each run to provide internal standards for molecular weight estimation.

Immunological
Methods-One milligram of purified wild type alkaline phosphatase was mixed with Freund's complete adjuvant and injected subcutaneously (subscapular region) into an adult rabbit.
Injection of the same mixture was repeated three times at weekly intervals.
Two weeks after the last injection the rabbit was bled and serum was prepared from the collected blood. The crude serum was fractionated with solid (NHd)2SOd. The fraction precipitating between 20 and 50% saturation which contained all *The abbreviation used is: SDS, sodium dodecyl sulfate.
of the detectable antibody activity, was dissolved in one-half the original volume of 0.01 M potassium phosphate buffer, pH 7.2, and dialyzed against 30 volumes of the same buffer containing 0.1 M NaCl. The dialyzed fraction was stored frozen at -15" until needed. Double diffusion studies of enzyme and antibody in agar gels were carried out as suggested by Ouchterlony (24) using microscope slides (1 X 3 inches) covered with a 0.85yo agarose gel prepared in a buffer containing 6 mM sodium borate (pH 8.3), 0.85% (w/v) NaCl. and 0.05% (w/v) sodium azide. Precipitin lines were allowed to develop for 72 hours at room temperature.
Nonprecipitated protein was removed by washing with several changes of borate-saline (0.85yo NaCl solution) buffer and the gels were then stained for protein by immersing for 1 min in a solution of 1% (w/v) aniline blue-black in 7.5oj, (v/v) acetic acid. Alternatively, the gels were stained for enzyme activity using a 0.05yo (w/v) solution of 5-bromo-4-chloro-3-indolyl phosphate in 0. At the time of harvest more than 99% of the radioactivity was inside the cells of either culture. The wild type mycelia were harvested on a Millipore filter, washed with distilled water, and then homogenized in 10 volumes (4.5 ml) of 0.2 M Tris buffer (pH 7.4) by grinding with alumina powder (12). The homogenate was centrifuged 10 min at 1000 X g. The pellet was discarded and the supernatant solution was centrifuged 60 min at 105,000 X g. The pellet was discarded.
The supernatant solution served as the source of wild type enzyme. The slime cells were harvested by centrifugation and discarded. The cellfree medium was concentrated to 4 ml on a Diaflo ultrafilter (UM-10 membrane) (Amicon Corp., Lexington, Md.) and used as the source of slime enzyme.
The wild type 105,000 X g supernatant and the concentrated slime medium (containing 1.3 units per ml and 0.75 unit per ml of repressible alkaline phosphatase, respect,ively) were treated as follows: 50 pg of pancreatic ribonuclease, 2.5 rg of TZ ribonuclease, and 100 &of deoxyribonuclease were added to each, and the preparations were incubated for 30 min at 37". The enzyme solutions were then fractionated by the addition of solid (NHd)&Oh; the fraction precipitating between 65 and 95yo saturation was collected by centrifugation, dissolved in 4 ml of 0.05 M potassium phosphate buffer, pH 7.2, and dialyzed against 1 liter of the same buffer for 24 hours.
The repressible alkaline phosphatase was precipitated from the dialyzed solutions by the addition of 0.4 ml of partially purified antiserum.
The tubes were incubated 45 min at room temperature and then centrifuged 5 min at 1500 X g to sediment the antibody-enzyme precipitate.
The precipitates were washed once with 0.5 ml of 0.05 M phosphate buffer.3 The washed precipit'ates were mixed with 1.0 ml of the sample buffer for SDS-acrylamide gel electrophoresis (described above) for each 1.9 units of enzyme activity present in the respective sample before precipitation with antiserum.
The suspensions were heated for 2 min at 100". Aliquots (0.1 ml) of the-resulting solutions were subjected to electrophoresis on SDS-acrvlamide eels. Followine: electroDhoresis the gels were cut into l.'i5-mm sections. Each iection was digested in a scintillation vial with 30yo hydrogen peroxide (25) and the saP content of each slice was determined by liquid scintillation counting (26).

Purification of Repressible
Alkaline Phosphatase from Wild Type N. crassa-The wild type strain, 74-OR&la, was grown in 20 liters of Fries' low phosphate (0.25 mM) medium for 3 days at 25". The mycelium (246 g wet weight) was mixed with crushed Dry Ice and the mixture was ground to a fine powder (approximately 10 min) in a Waring Blendor operating at high speed. The powder was stored at -15" for 3 days during which time the Dry Ice evaporated.
The powdered mycelium-was extracted with 800 ml of ice-cold distilled water bv brief homoeenization in the Waring Blendor.
The homogenati was centriruged 15 min at 5,000 X g and the supernatant solution was saved. The residue was extracted with 200 ml of distilled water as above and the supernatant solution after centrifugation was combined with the first.
The subsequent purification of the repressible alkaline phosphatase from the above extract was carried out exactly as described by Kadner el al. (6) except that the second CM-Sephadex chromatography step was omitted.
Purification of Revressible Alkaline Phosvhatase from Slime Stra&'of N. crassa-kultures of slime (800 mi per flask, 16 flasks) were grown for 18 hours (late log phase) in standard Fries' medium and were then harvested by centrifugation (5 min at 500 X g), washed once with phosphate-free Fries' medium (35 original culture volume), and then suspended in the original volume of fresh Dhosnhate-free Fries' medium and incubated 6 hours at 30". Cells \ere-removed by centrifugation for 5 min at 600 X g and the spent medium, which contained more than 90% of the total repressible alkaline phosphatase, was further clarified by filt,ering through a 1.2.crrn MilliDore filter.
The filtered medium was concentrated approximately 30-fold using a hollow fiber ultrafilter (Bio-Fiber 80 Beaker. Bio-Rad Laboratories, Richmond, Cal.). 'All further steps weke carried out at (t5". The concentrated protein solution from the medium was brought to 60% saturation by the addition of 370 mg of solid (NHh)2SOa per ml. The mixture was stirred 30 min and then centrifuged 15 min at 13,000 X g. The precipitate was discarded and the supernatant solution was brought to 95yo saturation by the addition of 257 mg of solid (NHd)zSOn per ml. The mixture was stirred 60 min and then centrifuged as above. The supernatant solution was discarded and the precipitate was dissolved in 10 ml of 0.01 M Tris buffer, pH 7.8, and dialyzed for 24 hours against 2 successive 2-liter volumes of the same buffer.
The dialyzed enzyme preparation was applied to a column (4.5 X 50 cm) of DEAE-Sephadex A-50 which had been equilibrated with the 0.01 M Tris buffer, pH 7.8, and the column was washed free of unbound protein with 800 ml of the same buffer. The phosphatase was then eluted by application of a linear gradient consisting of 2000 ml of 0.01 M Tris buffer, pH 7.8, in the mixing chamber and 2C00 ml of the same buffer containing 0.25 M NaCl in the reservoir.
Fractions (19 ml each) were collected and assayed for phosphatase activity.
Fractions 83 to 125 (1590 to 2380 ml of the gradient) contained the bulk of the activity (8170); these were pooled and concentrated to 18 ml by use of a Diaflo ultrafilter equipped with a UM-10 filter and then further concentrated to 2.9 ml using a collodion bag ultrafilter (Schleicher and Schuell Co., Keene, N. H.).
The concentrated enzyme fraction was applied to a column (1.5 X 80 cm) of Sephadex G-200 which had been equilibrated with 0.05 M Tris buffer, pH 8.3, containing 0.10 M NaCl. The column was eluted with the same buffer at a flow rate of 15 ml per hour and fractions (4.4 ml) were collected and analyzed for protein and phosphatase activity.
Fractions 14 to 18 contained the bulk of the phosphatase activity (80%) and had a constant ratio of enzyme activitv to protein content (absorbance at 280 nm). These fractions-were poo!ed and concentrated from 22 to 8.6 ml by use of the Schleicher and Schuell ultrafilter and then stored frozen at -15".

Enzyme
Putification- Tables  I and II summarize the purification of the repressible alkaline phosphatase from wild type mycelia and slime culture media, respectively.
The enzyme from B The antibody-enzyme complex formed under these conditions the two sources had similar solubiiity properties in ammonium retains approximately 50yo of the original enzyme activity. Pre-sulfate solutions and similar elution profiles on Sephadex G-200 liminary experiments revealed that essentially all of this residual columns, but differed in their behavior on ion exchangers. The activity is recovered in the pellet after centrifugation as above. In the experiments using enzyme prepared from a$P-labeled cul-enzyme from wild type mycelia is bound to CM-Sephadex at pH tures described above, less than 2yo of the residual enzyme ac-8.3 but is not bound to DEAE-Sephadex at pH 7.8; the phostivity was lost in the supernatant fluid and wash solution after phatase from slime culture media behaves in the opposite manner centrifugation of the antibody-enzyme complex. on the two types of ion exchange media, indicating a more acidic nature for the enzyme from this source. The final specific activity of both enzyme preparations was essentially identical and was very similar to the value reported earlier by Kadner et al. (6). Polyacrylamide Gel Electrophoresis-The results of gel electrophoresis of the purified phosphatase at pH 4.7 and 9.5 are shown in Fig. 1. A single, somewhat broad, protein band was observed for both enzyme preparations at both pH values. The slime enzyme migrated less rapidly toward the cathode at pH 4.7 and more rapidly toward the anode at pH 9.5 than did the wild type enzyme. The differences in the electrophoretic behavior of the two enzymes are consistent with their different chromatographic behavior on ion exchange media as noted above, with the slime enzyme behaving as the more acidic species in all cases. Duplicate gels stained for alkaline phosphatase activity (not shown) exhibited exactly the same banding pattern as the gels stained for protein (see Fig. 1).
The purified enzymes were further compared by electrophoresis at pH 8.8 in the presence of 0.1% sodium dodecyl sulfate (21), a procedure generally assumed to separate proteins on the basis of size rather than charge differences. Only a single band of protein was observed with either preparation (Fig. 2) and the mobility of the wild type enzyme was slightly greater than that of the slime enzyme.
Zmmunodi$usion-The immunological similarity of the repressible alkaline phosphatases purified from the two sources ABCD A B FIG. 1 (left). Polyacrylamide gel electrophoresis of purified wild type and slime repressible alkaline phosphatases.
Electrophoresis was carried out at either pH 4.7 (tubes A and B) or at pH 9.5 (tubes C and D) as described under "Experimental Procedures." Each sample contained 5 pg of protein.
Tubes A and C, wild type enzyme; tubes B and D, slime enzyme. The direction of electrophoretic migration was from top to bottom, toward the cathode for tubes A and B and toward the anode for tubes C and D. The gels were all stained for protein. FIG. 2 (center). SDS-polyacrylamide gel electrophoresis of purified wild type and slime repressible alkaline phosphatases. Electrophoresis was carried out as described under "Experimental Procedures." The sample for tube A contained 3 pg of wild type enzyme. The sample for tube B contained 3 pg of slime enzyme. The gels were stained for protein. FIG. 3 (right). Double diffusion analysis in agar gels of purified repressible alkaline phosphatases and antiserum prepared against the purified wild type enzyme. The center well contained 20 ~1 of ammonium sulfate fractionated antiserum.
Peripheral wells marked A contained approximately 4 pg of wild type enzyme. Peripheral wells marked B contained approximately 4 rg of slime enzyme. Precipitin lines in 1 were stained for protein with aniline blue-black.
was tested by double diffusion in agar gels against antiserum prepared against the wild type mycelial enzyme. A sharp, fully connecting precipitin line was observed on gels stained for either protein or enzyme activity (Fig. 3). In addition a second, much weaker, precipitin line for the slime enzyme was observed on gels stained for protein.
Amino Acid Composition-Amino acid analysis of the wild type and slime repressible alkaline phosphatases gave the results shown in Table III. The initial results are expressed on the basis of the amount of each amino acid corresponding to 1 A of enzyme at 280 nm as defined under "Experimental Procedures." These initial results have been used to calculate the number of residues of each amino acid per 136,000 g of protein. This value is the molecular weight of the protein moiety of the native wild type enzyme, as calculated from the data of Kadner et al. (6). The observed amino acid composition of the two enzyme preparations do not appear to differ significantly; the small differences for some of the amino acids are probably close to or within the experimental error.
Carbohydrate Content-Quantitative determination of the neutral and amino sugars present in the purified enzymes was carried out on samples hydrolyzed under mild conditions as described under "Experimental Procedures." The results are incorporated into the calculations for the composition of the two enzymes, and are shown in Table III. The values observed for the wild type enzyme in this study are fairly similar to those reported by Kadner et aZ. (6), but the values observed for the slime en-zyme differ markedly, the content of mannose and galactose being much higher in the latter enzyme. From the data in Table III the carbohydrate content of the wild type and slime phosphatases are calculated to be 13.7 and 23.8%, respectively, of the total weight of the enzymes.
Phosphate Content-Samples of the purified enzyme preparations were hydrolyzed in 6 N HCl and the inorganic phosphate content of the hydrolysates was measured by the method of Ames and Dubin (27) as modified by Bloch and Schlesinger (28). This method indicated less than 1 phosphate group present per native enzyme molecule of the wild type phosphatase (Table   IV), whereas the slime phosphatase was found to contain approximately 8 phosphate groups per enzyme molecule. The occurrence of phosphate in the repressible alkaline phos- n Based on calorimetric determination (13). * Calculated assuming a molecular weight of 136,000 for the protein moiety of the repressible alkaline phosphatase, based on the data of Kadner et al. (6). phatases from the two sources was further examined by isolation of enzyme from cultures derepressed in the presence of a2P04*-. The isolation procedure, described in detail under "Experimental Procedures," involved precipitation of enzyme from crude preparations with specific antiserum, solubilization of the antibodyenzyme complex in sodium dodecyl sulfate solution, and polyacrylamide gel electrophoresis in the presence of the detergent. Standards consisting of the respective purified alkaline phosphatases were carried through the same procedure.
The results of staining one set of gels for protein are shown in Fig. 4. The distribution of a2P on the gels, and a diagram of the position of the protein bands of the respective purified alkaline phosphatases on duplicate gels, are shown in Fig. 5. In agreement with the results of direct chemical analysis of the purified enzymes, the enzyme isolated from wild type mycelia has only very low levels of azP ( <0.5 phosphate per enzyme molecule) associated with it, whereas the enzyme isolated from slime media coincides with a large peak of azP activity. The preparation of enzyme samples and the procedure for electrophoresis are described in detail under "Experimental Procedures." Preparations were as follows: A, 3.5 pg (0.214 unit) of purified wild type mycelial enzyme; B, O.l-ml aliquot (equivalent to 0.210 unit) of enzyme isolated from an extract of wild type mycelium by precipitation with antiserum; C, OJ-ml aliquot (equivalent to 0.210 unit) of enzyme isolated from slime medium by precipitation with antiserum; D, 3.5 pg (0.220 unit) of purified enzyme from slime medium.
The very strong "extraneous" bands in B and C are denaturated and reduced chains of immunoglobulin. Enzyme from cultures of the wild type (X--X) or slime (O--O) strains of Neurospora crassa derepressed in medium containing 0.10 mM KHtZ2POa (40 pCi) was isolated by precipitation with antiserum and subjected to SDS-polyacrylamide gel electrophoresis as described under "Experimental Procedures." Duplicates of Gels B and C shown in Fig. 4 were cut into 1.75.mm slices and the radioactivity in each slice was determined as described.
A diagrammatic representation of the position of bands of the wild type mycelial enzyme (A) and the slime medium enzyme (B) on gels stained for protein ( Fig.  4) is shown for comparison.
in the repressible alkaline phosphatases purified in this study was tested by incubating 15 pg of each purified phosphatase with 50 pg of sialidase (N-acetylneuraminate glycohydrolase) for 24 hours at 30" in 0.12 ml of 0.1 M potassium acetate buffer, pH 5.0 (33). This amount of sialidase released 22 pg of sialic acid from 500 c(g of bovine submaxillary mucin in parallel incubations uuder the same conditions.
The recovery of alkaline phosphatase activity was greater than 90% for both the wild type arid slime enzymes at the cud of the sialidase treatment.
Disc gel electrophoresis of the treated alkaline phosphatases and of untreated controls was carried out at pH 4.7 and 9.5. The electrophoretic mobility of the treated samples was identical with that of the respective controls indicatiug the absence of terminal sialic acid residues in both forms of the enzyme. Jfolecular U'eight-The molecular weights of the wild type and slime enzymes were estimated by gel filtration, along with protein standards of known molecular weight, on a Sephades G-200 column as described under "Experimental Procedures." The peak elutiou volume of the slime alkaline phosphat,ase (Fig. 6) was less that1 that of the wild type etizyme, in agreement with the larger size expected of the former enzyme due t,o its increased carbohydrate content (Table III). The appareut molecular weights, estimated from the liue formed by plotting elution volume against log molecular weight of the staudards, were 165,000 for the wild type enzyme and 178,000 for the slime cuzyme.
The native form of the N. crassa repressible alkaline phosphatase is a dimer (6). The molecular weight of the subuuits was estimated by gel elcctrophoresis of the N. crassa alkaliue phosphatases and standards of kuown molecular weight in the preseuce of sodium dodecyl sulfate as described under "Esperimeutal Procedures." The results are showu in Fig. 7. The molecular weights of the subunits, estimated from the standard curve, were 85,000 for the wild type enzyme and 90,000 for the slime enzyme.
Thermal Stability-The rate of loss of activity during heat denaturation of the two enzyme preparations was measured at three pH values. The enzymes were prepared iu the indicated  The purified Neurospora crassa repressible alkaline phosphatases were subjected to SDS-polyacrylamide gel electrophoresis, as described under "Experimental Procedures," along with marker proteins of known subunit molecular weight. The gels were stained for protein and the mobility of the standard proteins and the N. crassa phosphatases were calculat,ed as suggested by Weber and Osborn (22). buffer by dialysis against a large csccss of the buffer, followed by dilutiou to a coucentratiou of 2 c(g per ml in the same buffer. Aliquots of the enzyme solutious were heated at the iudicated temperature for appropriate lengths of time, cooled rapidly, and surviving enzyme activity was determined in the standard enzyme assay. The half-life of each euzyme under the iudicated conditious of temperature aud p1-I were derived from the linear plots of iucubatiou time against the logarithm of the surviving enzyme activity.
As shown iu Table V, the half-lives of the wild type and slime enzymes did uot differ significantly under the conditions tested.
It is interesting to note however that, whereas a precipitate formed in solutions of the wild type enzyme heated above the denaturation temperature, no coagulation or precipitation was observed in similar solutions of the slime enzyme even after heating at 100" for several hours.
Kinetic Properties-K, values for p-nitrophenylphosphate, using the standard 0.3 M glycinate buffer (pH 9)-l mM EDTA assay system, were calculated from Lineweaver-Burk plots (34). The values obtained were 1.72 x lo+ M for the wild type enzyme and 1.56 x lo-* M for the slime enzyme. The observed Ki for phosphate under the same conditions was 1.31 x 10m4 M for the wild type enzyme and 1.23 x 10e4 M for the slime enzyme.
Incubation of Enzyme in Highly Concentrated Solution-The possibility that the phosphatase secreted by slime cells might be capable of a self-modification process in which neighboring enzyme molecules would liberate phosphate from each other was tested by prolonged incubation of highly concentrated enzyme solutions.
Samples of enzyme, 670 pg each, were prepared by dialysis against distilled water and then dried by lyophilization in conical tubes. The dried material was dissolved by the addition of 5 ~1 of Fries' salts (minus phosphate).
The concentrated enzyme solutions were incubated 24 hours at 37" in a closed container over moistened filter paper and then diluted to a concentration of 2 mg per ml with 0.2 M Tris-HCl buffer, pH 7.4. Enzyme assays indicated a recovery of greater than 95% of the enzyme activity.
Disc electrophoresis at pH 4.7 revealed no observable change in the electrophoretic mobility of either enzyme.
Distribution of Repressible Alkaline Phosphatase between Cells and Media of Wild Type and Several Mutant &rains of N. crassa-The repressible alkaline phosphatase of N. crassa has been considered to be primarily, if not exclusively, cell-bound (1,2) rather than excreted into the media as is the repressible acid phosphatase of the same organism (2,35). The finding that in cultures of the slime strain the repressible alkaline phosphatase is primarily free in the medium, prompted us to do a quantitative examination of the distribution of this enzyme in cultures of wild type and of several mutant strains affected in cell wall synthesis (36). Cultures were grown for 24 hours in 100 ml of phosphatefree Fries' medium supplemented with either 0.05 mM KH,POd or 2 mM 0-phosphorylethanolamine; this latter phosphorus source allows essentially maximal rates of growth of the organism, but at the same time gives quite a high degree of derepression of alkaline phosphatase (37). Cells were harvested and extracts were prepared by homogenization with alumina powder (12) in 0.05 M sodium acetate buffer, pH 5.0 (2). Alkaline phosphatase determinations on both the cell extracts and the cell-free culture media from each culture were carried out and the results are summarized in Table VI. As already mentioned, the repressible alkaline phosphatase activity in slime culture was found almost exclusively in the culture medium.
However, even in wild type a significant fraction of this enzyme appeared to be liberated into  the medium (20 to 30% under the conditions tested). The proportion of the phosphatase released into the medium by the cell wall mutants (OS-~; OS, cr; and OS, cr, aur) was intermediate between the value observed for the wild type and slime cultures.
Electrophoretic Mobility of Repressible Alkaline Phosphatase Isolated from Wild Type Culture ilfedium-Medium from a derepressed culture of wild type N. crassa (grown for 20 hours on 0.05 mM KHzPOl medium) was concentrated approximately 30fold with a Diaflo ultrafilter (UM-10 membrane). A further lo-fold concentration was achieved by the addition of Lyphogel (Gelman Instrument Co., Ann Arbor, Mich.).
The final concentrated solution (0.3 ml) contained 0.10 unit of repressible alkaline phosphatase activity (an extract of the mycelium harvested from this medium contained 0.90 unit of enzyme).
This solution was dialyzed overnight against 0.01 M Tris-HCl buffer, pH 7.4, and was then subjected to electrophoresis in the pH 9.5 buffer system. Standards consisting of the purified wild type mycelial and slime medium enzymes were run at the same time.
The bands of enzyme activity observed following electrophoresis are shown in Fig. 8. The repressible alkaline phosphatase isolated from wild type culture medium migrates at a rate more similar to that of the slime enzyme than that of the enzyme purified from wild type mycelium.
Properties of the Residual Repressible Alkaline Phosphatase in Washed Slime Cells-Although the bulk of the repressible alkaline phosphatase in slime cultures is found free in the medium (Table  VI), approximately 5% of the total activity is found still associated with the cells even after they have been washed with fresh medium.
Attempts to purify the enzyme from extracts of washed slime cells by ammonium sulfate fractionation revealed that a considerable portion (15 to 75y0 depending on how the cells were washed) of the activity was precipitated in 657, saturated solutions of the salt. Less than 5% of the enzyme isolated from slime medium or wild type mycelium is precipitated under the same conditions.
The electrophoretic mobility at pH 9.5 of this "65% ammonium sulfate precipitable" alkaline phosphatase is less than that of the enzyme isolated from either slime medium or wild type mycelium as shown in Fig. 9. The structure of this less soluble, more basic form of the enzyme was shown to be immunologically similar to the other repressible alkaline phosphatase species by precipitation of the expected number of enzyme units with graded amounts of the antiserum prepared against the wild type mycelial enzyme (Table VII). ABCABC FIG. 8 (left). Polyacrylamide gel electrophoresis of the repressible alkaline phosphatase isolated from wild type culture medium. Repressible alkaline phosphatase present in the medium of a derepressed culture of wild type Neurospora crassa was isolated as described in the text and subjected to polyacrylamide gel electrophoresis at pH 9.5 (lube B). The results of electrophoresis of purified wild type mycelial enzyme (tube A) and purified slime medium enzyme (tube C) are shown for comparison.
Bands of enzyme activity on the gels were located by staining with B-bromo-4-chloro-3-indolyl phosphate. FIG. 9 (right). Polyacrylamide gel electrophoresis of cell-bound repressible alkaline phosphatase from slime.
Cells from derepressed slime cultures were harvested by centrifugation and washed once with fresh medium.
Washed cells from 200 ml of medium were suspended in 5 ml of 0.05 M sodium acetate buffer, pH 5.0, and homogenized with alumina powder (12). The extract obtained after centrifugation at 27,000 X g for 15 min was fractionated with solid ammonium sulfate. The fraction, precipitating between 0 to 65% saturation, was dissolved in 1 ml of 0.01 M Tris-HCl, pH 7.4, and dialyzed overnight against the same buffer (100 ml). Samples containing approximately 0.02 enzyme unit were subjected to electrophoresis at pH 9.5, along with samples of purified wild type mycelial enzyme and purified slime medium enzyme. Following electrophoresis, enzyme activity was located on the gels by staining with 5-bromo-4-chloro-3-indolyl phosphate. The samples used are (A) slime medium enzyme, (B) wild type mycelial enzyme, (C) slime cell-bound, 0 to 65% ammonium sulfate-precipitable enzyme.

DISCUSSION
In contrast to the similarities noted above are the striking differences observed between the slime and wild type enzymes with respect to electrophoretic mobility, phosphate content, and carbohydrate content.
The repressible alkaline phosphatase secreted by the sZime The repressible alkaline phosphatase of N. crassa was shown mutant was purified to apparent homogeneity as judged by poly-by Kadner et al. (6) to be a glycoprotein. In their study the acrylamide gel electrophoresis at pH 8.8 in the presence of sodium carbohydrate content of the purified enzyme was found to be dodecyl sulfate, and at pH 4.7 and pH 9.5 in the absence of the approximately 11.5% of the total weight. In the present study detergent.
The enzyme was also purified from the mycelium of the carbohydrate content of the purified wild type enzyme was a wild type strain by the procedure of Kadner et al. (6). The found to be approximately 13.7%. The relative amounts of two purified enzyme preparations exhibited essentially identical glucosamine, mannosc, and galactose found in the wild type specific activities (Tables I and II) based on protein determina-enzyme are very similar to those found by Kadner et al. (6). tion by either the calorimetric procedure of Lowry et al. (13) or The somewhat higher total amounts found in this study may be by absorbance at 280 nm. However, the specific activity of the due to differences in the wild type strains used in the two studies. phosphatase in crude concentrated medium from slime cultures The slime enzyme purified in this study was found to contain a The indicated volume of 20-fold diluted antiserum was added to O.lO-ml aliquots of either the purified wild type mycelial enzyme (diluted to 8.7 pg per ml) or the 0 to 65% ammonium sulfate precipitable fraction of an extradt of washed, derepressed slime cells. The total volume was adjusted to 0.20 ml by addition of borate-saline buffer. Tubes were incubated 60 min at room temperature, centrifuged 15 min at 1500 X g, and repressible alkaline phosphatase remaining unprecipitated was determined in the standard assay system using 0.025-ml aliquots of the supernatant solutions. was la-fold higher than that in the crude extract of wild type mycelium.
In fact, the results of the purification procedure indicate that the enzyme represents about 25% of the total protein present in the cell-free medium of derepressed slime cultures. This selective appearance of the enzyme in the culture medium clearly indicates active secretion by the slime cells rather than liberation by lysis of cells during starvation for phosphate.
The repressible alkaline phosphatases purified from the two sources were very similar in a number of properties. Amino acid analysis revealed no significant differences; the small differences for some of the amino acids (Table III) are in the range of experiment,al error. More detailed studies, e.g. peptide mapping, will be required, however, to rule out a possible difference of one or a few residues in the primary structure of the two forms of the enzyme. Further evidence of the similarity of the slime and wild type enzymes is provided by the following: (a) apparent identity in immunodiffusion studies (Fig. 3) ; (b) similar heat stability at three pH values (Table V) ; (c) nearly identical molecular weight (after correction for carbohydrate content) both in the native form (Fig. 5) and after dissociation into subunits (Fig. 6) ; and (d) similar affinities for the substrate, p-nitrophenylphosphate, and the inhibitory product, orthophosphate. much larger per cent of carbohydrate than the wild type enzyme, and the relative amounts of glucosamine, mannose, and galactose were also quite different.
The glucosamine content of the slime enzyme was almost identical with that of the wild type enzyme, but the mannose content was approximately 80% higher and t.he galactose content was over 3-fold higher.
No other carbohydrate, other than probably insignificant traces of glucose, was found in either enzyme preparation.
The electrophoretic mobility of a protein in polyacrylamide gel is determined by both the size and the net charge on the protein. The fact that the slime phosphatase is the faster species when migration is toward the anode (pH 9.5) and is the slower species when migration is toward the cathode (pH 4.7) indicates that it is separated from the wild type enzyme primarily due to a difference in charge. The behavior of the two enzyme preparations on DEAE-and CM-Sephadex during purification supports this conclusion.
The possibility that the difference between the slime and wild type enzymes could be at least partially due to a difference in amino acid composition cannot be completely ruled out at this time. As pointed out above, amino acid analysis is not sensitive enough to show conclusively a difference of only one, or a few, residues in such a large molecule. However, wild type N. crassa also secretes a significant amount of repressible alkaline phosphatase into the growth medium (Table VI), and the electrophoretic mobility of this secreted wild type enzyme more nearly resembles that of the secreted slime enzyme than that of the enzyme retained by wild type mycelia (Fig. 8). This secreted wild type enzyme has not been obtained in sufficient quantity to allow its purification and chemical analysis, but its similarity to the secreted slime phosphatase strongly suggests that the secreted wild type enzyme and the mycelial wild type enzyme differ in degree of post-translational modification rather than differing in their primary amino acid structure. There is precedent for such a phenomenon: the exo-1 (38) and T9 (39) mutants of N. crassa appear to be simultaneously affected in cell wall synthesis, and they synthesize glucoamylase with altered gel filtration (38) or isoelectric focusing (39) properties.
Sialic acid, which is known to account for the electrophoretic heterogeneity of a number of enzymes (29-32), is probably not present in either the wild type or slime repressible alkaline phosphatases. Treatment of the purified enzymes with a large excess of sialidase failed to alter their respective electrophoretic mobilities, a result which would have been expected if terminal sialic acid residues had been present.
The purified slime-phosphatase was found to contain approximately 8 phosphate groups per enzyme molecule (Table IV) whereas the wild type enzyme contained less than 0.5 phosphate group per enzyme molecule.
This difference in phosphate content would seem to provide a sufficient explanation for the electrophoretic difference between the two forms of the enzyme. Bloch and Schlesinger (28) have shown that the purified, native E. coli alkaline phosphatase contains 1.6 to 2.1 moles of tightly bound inorganic phosphate per mole of enzyme. This phosphate is removed by dialysis against nitrilotriacetic acid, a procedure which removes tightly bound zinc ions from the enzyme. The alkaline phosphatase isolated from slime cells labeled with 82P048retained most of its bound azP during an isolation procedure which included dialysis against phosphate buffer, heating at 100" in phosphate buffer containing 0.1 To sodium dodecyl sulfate, and SDS-polyacrylamide gel electrophoresis (Fig. 4). Retention of the radioactive phosphate under these conditions indicates that in this case it is probably present as covalently bound organic phosphate rather than tightly bound inorganic phosphate.
The subcellular localization of the repressible alkaline phosphatase of N. crassa has not been studied, although the enzyme has been considered to be primarily cell-bound rather than secreted into the medium (1,2). The analogous enzyme in several bacteria is known to be located in the periplasmic space between the cell membrane and cell wall (40)(41)(42)(43).
The fact that the slime mutant, which lacks the normal cell wall, secretes approximately 95% of the repressible alkaline phosphatase into the medium strongly suggests that this enzyme also is normally located in the periplasmic space. A similar distribution of enzyme between cells and growth medium in slime cultures has recently been observed for invertase (lo), an enzyme previously shown to be primarily located in a position external to the plasma membrane (441. If the similar electrophoretic mobility of the phosphatasc secreted into the medium by both wild type mycelia and slime cells (Fig. 8) is a reflection of similar structures, then the slime enzyme may represent a natural stage in maturation rather than an "abnormal" form resulting from the mutant phenotype. It is even possible that this is the form of the enzyme as it initially enters the periplasmic space. According to this hypothesis the bulk of this enzyme, which is retained in the periplasmic space, would then be further modified, perhaps by alkaline phosphatase itself, to produce the "wild type" form of the enzyme. The enzyme secreted by slime would escape this modification due to its rapid dilution into the surrounding medium. Incubation of the purified slime phosphatase at a concentration (approximately 50%, w/v) which might occur if all of the enzyme were localized in the periplasmic space did not result in "self-modification" as judged by lack of change in electrophoretic mobility. Further studies will be required to determine whether enzyme(s) capable of converting either the cell-bound or secreted form of the enzyme to the alternative form can be isolated from N. crassa.
A small amount of repressible alkaline phosphatase activity remains associated with the slime cells even after they have been washed with fresh medium.
Yluch of this activity is indistinguishable from the enzyme found secreted into the medium. However, a significant proportion (15 to 750/, depending on the thoroughness with which the cells are washed) of this residual cellbound enzyme clearly differs in its solubility and electrophoretic properties ( Fig. 9) from both the purified enzyme from slime medium and the purified enzyme from wild type mycelium.
The structure of this less soluble, more basic form o! the enzyme is clearly related to that of the other repressible alkaline phosphatase species since it is precipitated by antiserum prepared against the purified wild type mycelial enzyme (Table VII).
It seems possible that this form of the phosphatase may represent enzyme which has not yet been modified by the attachment of the carbohydrate residues.