Vol. 157, No. 2JOURNAL OF BACTERIOLOGY, Feb. 1984, p. 375-3790021-9193/84/020375-05$02.00/0Copyright © 1984, American Society for Microbiology

Characterization and Comparison of a Neurospora crassa RNasePurified from Cultures Undergoing Each of Three Different States of

DerepressionRICHARD A. LINDBERG* AND HARVEY DRUCKER

Biology and Chemistry Department, Pacific Northwest Laboratory, Richland, Washington 99352

Received 2 August 1983/Accepted 4 November 1983

Extracellular RNase N4 from Neurospora crassa is derepressible by limitation of any of the three nutrientelements obtainable from RNA. We have purified and characterized the enzyme from cultures grown undereach of the three states of derepression. The purification procedure consisted of an ultrafiltration step,cation-exchange chromatography, and gel filtration. We found only one enzyme (N4) that hydrolyzed RNAat pH 7.5 in the presence of EDTA in culture filtrates from nitrogen-, phosphorus-, or carbon-limited cells.In all three cases, the enzymes were identical by polyacrylamide gel electrophoresis (Mr -9,500) and by gelfiltration (Mr -10,000). There were no differences in thermal stability or pH optimum; all three cross-

reacted with antibody to the nitrogen-derepressed enzyme in interfacial ring and in Ouchterlony tests.Digestion of homopolyribonucleotides indicated that N4 preferentially cleaved phosphodiester bondsadjacent to guanine residues. Results indicate that the enzymes are very similar or identical and areprobably products of the same gene. N4 appears to be homologous to guanine-specific RNases from otherfungal sources.

Microbial eucaryotes obtain nutrients similarly to procary-otic organisms; however, because of their larger genomesthey have evolved much more complex regulatory mecha-nisms for nutrient acquisition. Synthesis of some extracellu-lar hydrolases is controlled by more than one regulatorycircuit. In particular, extracellular proteases are derepressi-ble by limiting any of the three elemental nutrients obtain-able from protein: carbon, nitrogen, or sulfur. Neurosporacrassa alkaline and neutral proteases are synthesized andsecreted in response to C, N, or S derepression (4, 6). SomeN. crassa acid proteases are also derepressible by more thanone element (16). Similar regulatory phenomena for prote-ases has been reported for some species of Candida (19),Mucor (12), Aspergillus (2), and others. In the Aspergillusgenus, at least 10 different species secrete proteases inresponse to derepression for more than one nutrient (3).Evidence indicates that the same enzyme is produced underdifferent states of derepression, and models have beenpresented in which one gene is regulated by multiple regula-tory macromolecules (17, 18).We have found that RNase N4 is secreted from N. crassa

mycelia under three different states of derepression (14). Theenzyme probably functions by utilizing external RNA as asource of phosphorus, nitrogen, or carbon, since it is dere-pressed by limiting any of these three elements. Presumably,expression of a single structural gene is controlled byregulatory macromolecules from the three different nutrientacquisition circuits. In this communication, we report thepurification, characterization, and comparison of RNase N4synthesized under all three states of derepression.

MATERIALS AND METHODSGrowth of organism. Wild-type N. crassa 74-OR23-1A was

maintained and harvested as described by Turner and Mat-chett (21). Conidia were germinated and allowed to grow for12 h in Vogel's (24) minimal medium plus a carbon source, at

* Corresponding author.

375

which time cell density was -1.5 mg (dry cell weight) per ml.RNase assay. The assay was performed as described

previously (14). One unit of activity is defined as the amountof enzyme required to raise the non-acid-precipitable absor-bance at 260 nm (A260) of yeast RNA 1.0 U in 10 min at 37°C.

Purification procedure. Three liters of 12-h mycelia werefiltered and washed with one-half-strength Vogel's minimalmedium minus a nitrogen source. The mycelia were resus-pended in 6 liters of media which contained: Vogel's mediumminus a nitrogen source, 2% dextrose, 2 mM CaC12, and0.25% gelatin. These cultures were shaken at 30°C for 6 h,and the filtrate was collected. Sodium azide (to 0.02%) wasadded to the filtrate as a preservative. The remainder of theprocedure was carried out at 4°C.The culture filtrate was concentrated to -200 ml on a

H1P-5 ultrafiltration device (Amicon Corp.) by recirculatingflow, and the H1P-5 filtrate was reconcentrated to -200 mland pooled with the first concentrate. The 400 ml of concen-trate was dialyzed against 0.01 M citrate-0.02% NaN3 (pH5.8). This preparation was applied to a carboxymethylcellulose column (2.5 by 24 cm; Whatman CM 52) that hadbeen equilibrated with the dialysis buffer described above.The RNase activity was eluted with a 600-ml 0 to 0.12 M KClgradient. One peak of activity was eluted, and those frac-tions were pooled and concentrated on a Diaflo YM-5membrane (Amicon). The concentrate (-6 ml) was appliedto a Sephadex G-75 column (1.6 by 100 cm). One of severalprotein peaks contained all the RNase activity; those frac-tions were pooled and rechromatographed on the sameSephadex G-75 column. The resultant profile showed onepeak of protein, and corresponding fractions were pooled asRNase N4.The purification procedures for the carbon- and phospho-

rus-derepressed enzymes were the same as those describedabove, with the following modifications. For phosphorusderepression, KH2PO4 in Vogel's medium was replaced byan equal molar amount of KCl instead of omitting nitrogen.The culture filtrate (after addition of NaN3) received 8 mg of

on October 28, 2017 by guest

http://jb.asm.org/

Dow

nloaded from

376 LINDBERG AND DRUCKER

pure N. crassa alkaline protease (15) and was allowed toremain at room temperature for 2 days to digest the gelatininducer. For carbon limitation, the cells were derepressed inone-half-strength complete Vogel's medium without dex-trose. In addition, thermolysin at 75 ng/ml was added, aspreviously described for protease induction (5). The proce-dure for carbon-starved cultures also included carboxy-methyl cellulose chromatography as described above, butthe pH was 4.25, and elution was performed with a 0 to 0.2 MKCI gradient. This chromatographic step was performedbefore the pH 5.8 Whatman CM-52 chromatography.Enzyme characterization. Polyacrylamide slab gel electro-

phoresis in the presence of sodium dodecyl sulfate wascarried out as described previously (15), except that thedestaining solution was water-methanol-acetic acid (10:5:2).The cyanogen bromide cleavage was performed in 70%formic acid, as described by Lerch et al. (13), except that 2mg of protein were used. Enzyme concentration was esti-mated assuming that, at 280 nm, A1 cm 1 mg/ml = 1.0. The pHoptimum was determined by performing the standard assayat several different pH values. Buffers (at 0.1 M) employedwere citrate, from pH 5.5 to 6.5; phosphate, from pH 6.0 to7.5; and Tris from pH 7.0 to 9.0.Goat anti-N4 antibody was raised by injecting 0.5 mg of

purified N4 from N-derepressed cultures. Antigen was mixedwith Freund complete adjuvant and injected subscapularly.A booster was given 2 weeks later, after which serum wascollected and assayed for antibody with interfacial ring tests.The goat serum immunoglobulin G (IgG) was partially puri-fied by a 1.75 M ammonium sulfate precipitation; the precip-itated IgG was redissolved in and dialyzed against 0.075 Mphosphate-0.075 M NaCl-0.1% sodium azide at pH 7.2. Theamino acid analysis of the preparation purified from phos-phorus-limited cultures was performed by the AAA Labora-tory of Mercer Island, Washington.

Specificity toward homopolyribonucleotides was deter-mined by reacting pure N4 with substrate (Sigma ChemicalCo.) at 1 mg/ml (in 0.1 M Tris, 0.005 M EDTA [pH 7.5]).These preparations were sampled and precipitated, and A260was determined as described in the RNase assay reportedpreviously (14). A blank for each of the four homopolyribo-nucleotides that did not receive enzyme was subtracted fromthe respective experimental values.

RESULTSRNase N4 was purified from nitrogen-starved, phospho-

rus-starved, and carbon-starved cultures of N. crassa. Gela-tin (0.25%) was used as the exogenous protein required forenzyme production in all three cases, and thermolysin wasadded to C-derepressed cultures as described previously(14). Purified N. crassa alkaline protease (15) was added tothe culture filtrate from phosphorus-starved mycelia to di-gest the exogenous gelatin. This was accomplished withoutloss of RNase activity, indicating that N4 may be resistant toN. crassa protease. Three N. crassa proteases were coordi-nately produced in the N- and C-starved cultures and werenaturally present to digest the gelatin (16).By employing the RNase assay described previously (14),

we found only one polypeptide with RNase activity incultures grown for the isolation of extracellular RNase. Ineach of the three purification procedures, only one peak ofactivity appeared in the elution profiles from the first chro-matography; this activity remained a single peak in subse-quent chromatograms. The enzyme found in those cultureswas designated RNase N4. Recoveries for the procedureswere -1 to 3 mg from 6 liters of filtrate, representing 10 to

20% recovery. The specific activities of the final prepara-tions were -660,000, 620,000, and 570,000 U/mg for theenzymes from P-, N-, and C-starved cultures, respectively.The preparations appeared homogenous on polyacrylamideslab gels with 30-,ug samples (Fig. 1) but contained contami-nants when larger sample sizes were electrophoresed.The enzymes from the three different states of derepres-

sion were indistinguishable by sodium dodecyl sulfate-slabgel electrophoresis (Fig. 1). They eluted at the same positionwhen subjected to Sephadex G-75 gel filtration (Fig. 2).These results indicate that the three enzymes have the sameeffective size both under denaturing conditions and in nativeconformation. When electrophoresed and chromatographedwith protein standards, the position of N4 corresponded toMr of 9,500 and 10,000, respectively (Fig. 3).Thermal stability is generally regarded as a criterion in

determining slight difference in the primary structures of anenzyme. The stability of N4 from the three different cultureconditions was investigated, and no differences were ob-served (Fig. 4). The enzymes were stable at much highertemperatures with incubation periods shorter than 30 min.The pH optimum for hydrolysis of yeast RNA was near pH7.5 for N4. The pH profile generated was very similar to thatreported for the RNase investigated by Hasunuma et al. (8).Phosphate buffer was inhibitory, compared with Tris, andthe range of optimal activity was very narrow, with -20%decrease in activity only 0.5 pH unit on either side of pH 7.5.As with thermal stability, no differences were observedamong the three different enzyme preparations.

Inhibition by some cations has been reported for homolo-gous RNases (22); we therefore investigated inhibition for allthree enzymes, which were equally inhibited by the cationswe tested. Activity was reduced by -40% in the presence of0.01 M CaC12, 0.01 M MgC92, or 0.001 M HgCl2; 94 and 70%

_......s...r...s

A B C

FIG. 1. Polyacrylamide slab gel electrophoresis of RNase N4.Purified N4 from (A) phosphorus-, (B) nitrogen-, and (C) carbon-derepressed cultures were electrophoresed in 12.5% gels in thepresence of sodium dodecyl sulfate. Sample size was 30 ,ug.

J. BACTERIOL.

on October 28, 2017 by guest

http://jb.asm.org/

Dow

nloaded from

RNase FROM THREE STATES OF DEREPRESSION 377

0.28

0.24 F-

0.20EC

0

co

1- 0.16

w

0

z

t 0.120Cnm

0.08

0.04 _-

0

I

I

_

1

Il000.. -

70 80 90 100 110

FRACTION NUMBERFIG. 2. Sephadex gel filtration of RNase N4 from three culture

conditions. Shown is the elution profile of the final Sephadex G-75column in the purification procedure of N4 from carbon-limitedcultures (0); nitrogen-limited cultures (Y); and phosphorus-dere-pressed cultures (0). The profiles for the enzymes from N- and P-starved cultures are offset by 0.02 and 0.04 absorbance units,respectively.

inhibition was observed with 0.001 M ZnSO4 and CuSO4,respectively. N4 was also inactivated by iodoacetate underthe conditions described by Hashimoto et al. (7).The enzyme preparations were similar immunologically.

All three cross-reacted in interfacial ring tests with antibodyto the enzyme from N-starved cultures.. All three alsoreacted in a similar manner in the Ouchterlony test (Fig. 5).The cause of the broad precipitation lines in the Ouchterlonytest was unclear, but the reaction was the same for all threeenzymes. This reaction may be related to the fact thatRNases of similar properties have been reported to havevery weak immunogenicity (22).Amino acid analyses were performed on the enzyme

obtained from phosphorus-limited cultures. The results areshown and compared with the composition of N1 (an N.crassa-stationary-phase RNase) in Table 1. The number ofresidues was not extrapolated to integer values because thepreparation was not pure. Assuming 95% purity, most valuesapproximate the correct number of residues. The composi-tion of N4 was very similar to that of N1 and other guanine-specific RNases (25). N4 had fewer residues than N1, wheredifferences are apparent, commensurate with its smaller

size: N1 = 11,450 (7). N4 had two fewer cysteines than N1and only one methionine, compared with two for N1. Cyano-gen bromide cleavage of the nitrogen-derepressible enzymeyielded a fragment with an Mr of -500 to 1,000 less than theintact enzyme and a peptide fragment, which also indicatedone methionine residue.

Specificity was investigated by measuring activity relativeto homopolyribonucleotides. At low enzyme levels (i.e., 100ng/ml), only polyguanylic acid [poly(G)] was hydrolyzed atdetectable rates. As enzyme levels increased, polycytidylicacid [poly(C)], polyuridylic acid [poly(U)], and polyadenyl-ic acid [poly(A)] were hydrolyzed in that order. These dataindicate a specificity of G > C > U > A. All threepreparations hydrolyzed poly(G) equally at all enzyme levelstested. The nitrogen-derepressed enzyme preparationshowed higher activity with poly(C) than the other twopreparations. However, poly(C) was unstable in solution andwas not supplied with minimum-molecular-weight values.Therefore, activity required to increase acid-soluble prod-ucts was not comparable to the other three homopolyribonu-cleotides. It was apparent that G > U or A, but furtherexperimentation will be necessary to clarify the specificity ofN4.

DISCUSSIONWe have purified RNase N4 from cultures of N. crassa

grown under three different conditions. The enzyme isapparently homologous with the guanine-specific RNases.These enzymes, which include T1, F1, Ul, and N1 fromAspergillus oryzae, Fusarium moniliformis, Ustilagosphaerogena, and Neurospora crassa, respectively, possess

0

I-:wLI)0

FRACTION NUMBER (O-O)70 80 90 100

2 3 4 5 6RELATIVE MOBILITY (0-*)

FIG. 3. Molecular weight estimation of RNase N4 by gel filtra-tion (0) and sodium dodecyl sulfate-gel electrophoresis (0). Stan-dards used were: BSA, bovine serum albumin; OVAL, ovalbumin,THERM, thermolysin; TRYP, trypsinogen; and CYTC, cytochromeC. The arrows indicate N4, which corresponds to an Mr of 10,000and 9,500 for gel filtration and gel electrophoresis, respectively.

VOL. 157, 1984

on October 28, 2017 by guest

http://jb.asm.org/

Dow

nloaded from

378 LINDBERG AND DRUCKER

5;

-0.

801-

601-

40

201-

0

I I I

0

II40 50 60 70

TEMPERATURE (VC)FIG. 4. Thermal stability of N4 purified from the three states of

derepression. N4 produced from cultures limited for C (0), P (0),and N (A) were held at indicated temperatures for 30 min andassayed for remaining activity. Activity is expressed as percent ofthe 37°C sample.

similar enzymatic and physical properties. The propertiesof N4 resemble those of the enzymes listed, although themolecular weight of N4 is lower. The others have molecularweights ranging from 11,000 to 11,500 (7, 25). N1 wasreported as a stationary-phase enzyme (10, 20) and hassimilar properties to those reported here for N4 except fordifferences in size and amino acid composition (7, 22). N1has also been reported to be phosphate-derepressible (8), butfrom the evidence presented it was not clear whether theenzyme detected was N1 or N4. The relationship between N1and N4 requires elucidation; they are obviously very similarand may be differentially edited products of the same gene orthey may have resulted from the duplication of an ancestralgene.

Microbial hydrolases that are derepressible by limitationof more than one nutrient are believed to be products of thesame gene (17, 18). Genetic evidence is lacking, but charac-terizations of enzymes produced under different conditions

FIG. 5. Ouchterlony test of N4 with antibody raised against

nitrogen-derepressed enzyme. Wells 1, 2, and 3 contained N4 from

cultures derepressed for P, N and C, respectively; the center well

contained goat anti-N4. The Ouchterlony test was deproteinized byextensive dialysis and stained with Coomassie blue.

TABLE 1. Comparison of amino acid analysis of RNase N4preparation from phosphorus-starved N. crassa and that of

RNase N,'No. of residues

Amino acidN4 N1b

Ala 9.08 10Arg 3.15 3Asp 10.33 14Cys/2c 1.91 4Glu 5.43 4Gly 16.65 13His 2.47 3ISod 3.35 5Leu 3.81 4Lys 2.00 3Met 1.23 2Phe 3.87 5Pro 6.56 5Sere 12.12 14Thre 3.85 4Trpf 1.42 1Tyr 6.50 9Vald 3.06 4a Determined by assuming two lysine residues. Calculated by

averaging 24-, 48-, and 96-h hydrolysates, except as noted.b Reference 22.c Performic acid oxidized before acid hydrolysis. Calculated from

cysteic/aspartic acid ratio.d Maximum value.Extrapolated to zero time hydrolysis.

f Determined by the method of Hugli and Moore (9).

have yet to reveal any differences. Alkaline protease from N.crassa is perhaps the best-characterized triple-derepressibleenzyme. The proteases from C, N, or S derepression areidentical when tested by cellulose acetate electrophoresis,thermal inactivation, substrate specificity, several chro-matographic techniques, and immunological characteristics(4, 6). In addition, some genetic evidence exists to indicatethat one gene is responsible for the enzyme in all three cases.When electrophoresed, the alkaline protease from one strainof N. crassa migrated differently from that of all other strainstested; in this strain, the enzymes from all three states ofderepression behaved in that manner (6).RNase N4 obtained from each of the three different states

of derepression was also identical by several criteria. Thepreparation from N-starved cultures did show higher activitythan the other two with poly(C) as a substrate. Purerpreparations and further characterization will be necessaryto determine specificity and whether the enzymes are identi-cal in every respect. Multiple formns of the same RNase haverecently been reported for three different fungal organisms(11, 23, 25). In these cases, amino acid composition was thesame, but differences in isoelectric points, conformations, orspecific activities were observed. Whether multiple forms ofan enzyme are related to regulatory mechanisms or physio-logical status of the cell has not been investigated.The data presented in this paper seem to support current

models that describe multiple-derepression mechanisms asbeing composed of one gene controlled by three receptorsites, one for each of three different regulatory proteins (17,18). In the case of N4, both the nit-2 and nuc-J proteinsapparently act in a positive manner to initiate transcription(14). Multiple regulation has been studied at the level of thegene for yeast invertase by Carlson and Botsein (1). Theyhave found that two forms of invertase (each controlled by a

J. BACTERIOL.

on October 28, 2017 by guest

http://jb.asm.org/

Dow

nloaded from

RNase FROM THREE STATES OF DEREPRESSION 379

different regulatory mechanism) are transcribed from onestructural gene. Elucidation of regulatory mechanisms ofmultiple derepression and verification of the existence ofmultiple receptor sites for some microbial hydrolases awaitanalysis at the level of the gene.

ACKNOWLEDGMENTThis work was supported by the U.S. Department of Energy

under contract DE-AC06-76RLO-1830.

LITERATURE CITED1. Carlson, M., and D. Botsein. 1982. Two differentially regulatedmRNAs with different 5' ends encode secreted and intracellularforms of yeast invertase. Cell 28:145-154.

2. Cohen, B. L. 1973. The neutral and alkaline protease of Asper-gillus nidulans. J. Gen. Microbiol. 77:521-528.

3. Cohen, B. L. 1981. Regulation of protease production in Asper-gillus. Trans. Br. Mycol. Soc. 76:447-450.

4. Cohen, B. L., J. E. Morris, and H. Drucker. 1975. Regulation oftwo extracellular proteases of Neurospora crassa by inductionand by carbon- nitrogen- and sulfur-metabolite repression.Arch. Biochem. Biophys. 169:324-330.

5. Drucker, H. 1973. Regulation of exocellular proteases in Neu-rospora crassa: role of Neurospora proteases in induction. J.Bacteriol. 116:593-599.

6. Hanson, M. A., and G. A. Marzluf. 1975. Control of thesynthesis of a single enzyme by multiple regulatory circuits inNeurospora crassa. Proc. Natl. Acad. Sci. U.S.A. 72:1240-1244.

7. Hashimoto, J., T. Uchida, and F. Egami. 1971. Purification ofribonuclease Ul, and some properties of ribonuclease U1 andNl. J. Biochem. 70:903-911.

8. Hasunuma, K., A. Toh-E, and T. Ishikawa. 1976. Control of theformation of extracellular ribonuclease in Neurospora crassa.Biochim. Biophys. Acta 432:223-236.

9. Hugli, T. E., and S. Moore. 1972. Determination of the trypto-phan content of proteins by ion-exchange chromatography ofalkaline hydrolysates. J. Biol. Chem. 247:2828-2834.

10. Kasai, K., T. Uchida, F. Egami, K. Yoshida, and M. Nomoto.1969. Purification and crystallization of ribonuclease N, fromNeurospora crassa. J. Biochem. 66:389-396.

11. Kanaya, S., and T. Uchida. 1981. Purification of ribonuclease T,

by affinity chromatography. J. Biochem. 89:591-597.12. Lasure, L. L. 1980. Regulation of extracellular acid protease in

Mucor miehei. Mycologia 72:483-493.13. Lerch, L., C. Longoni, and E. Jordi. 1982. Primary structure of

tyrosinase from Neurospora crassa I. Purification and aminoacid sequence of the cyanogen bromide fragments. J. Biol.Chem. 257:6408-6413.

14. Lindberg, R. A., and H. Drucker. 1983. Regulation of a Neuros-pora crassa extracellular RNase by phosphorus, nitrogen, andcarbon derepressions. 157:380-384.

15. Lindberg, R. A., L. D. Eirich, J. S. Price, L. Wolfinbarger, Jr.,and H. Drucker. 1981. Alkaline protease from Neurosporacrassa: purification and characterization. J. Biol. Chem.256:811-814.

16. Lindberg, R. A., W. G. Rhodes, L. D. Eirich, and H. Drucker.1982. Extracellular acid proteases from Neurospora crassa. J.Bacteriol. 150:1103-1108.

17. Marzluf, G. A. 1977. Regulation of gene expression in fungi, p.196-242. In J. C. Copeland and G. A. Marzluf (ed.), Regulatorybiology. Ohio State University Press, Columbus, Ohio.

18. Metzenberg, R. L. 1979. Implications of some genetic controlmechanisms in Neurospora. Microbiol. Rev. 43:361-383.

19. Ogrydziak, D., A. Demain, and S. R. Tannenbaum. 1977. Regu-lation of extracellular protease production in Candida lipolytica.Biochim. Biophys. Acta 497:525-538.

20. Takai, N., T. Uchida, and F. Egami. 1966. Purification andproperties of ribonuclease N1, an extracellular ribonuclease ofNeurospora crassa. Biochim. Biophys. Acta 128:218-220.

21. Turner, J. R., and W. H. Matchett. 1968. Alteration of trypto-phan-mediated regulation in Neurospora crassa by indoleglyc-erol phosphate. J. Bacteriol. 95:1608-1614.

22. Uchida, T., and F. Egami. 1971. Microbial ribonucleases, p.205-250. In P. D. Boyer (ed.), The enzymes, vol. 4. AcademicPress, Inc., New York.

23. Uchida, T., and Y. Shibata. 1981. An affinity adsorbent, 5'-adenylate-aminohexyl-sepharose. I. Purification and propertiesof two forms of RNase U2. J. Biochem. 90:465-471.

24. Vogel, H. V. 1964. Distribution of lysine pathways among fungi:evolutionary implications. Am. Nat. 98:435-446.

25. Yoshida, H., I. Fukuda, and M. Hasheguchi. 1980. Purificationby affinity chromatography and physiochemical properties ofthe guanine-specific ribonuclease of Fusarium moniliform. J.Biochem. 88:1813-1818.

VOL. 157, 1984

on October 28, 2017 by guest

http://jb.asm.org/

Dow

nloaded from