THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 249, No. 12, Issue of June 25, pp. 3884-3889, 1974

Printed in U.S.A.

Purification and Properties of Acid Deoxyribonucleases of

Human Gastric Mucosa and Cervix Uteri*

(Received for publication, August 6, 1973)

MASAYOSHI YAMANAKA, YORIAKI TSUBOTA, MOTOAKI ANAI, KOJI ISHIMATSU, MAKOTO OKUMURA, SHIBAN- OSUKE KATSUKI, AND YASUYUKI TAKAGI

From the Department of Biochemistry and the Second Department of Internal Medicine, Kyushu University School of Medicine, Fukuoka, 812, Japan

SUMMARY

The acid deoxyribonucleases in homogenates of human gastric mucosa and cervix uteri were purified extensively by a procedure including DEAE-cellulose and phosphocellulose chromatographies, gel filtration, and isoelectric focusing. The enzymes in the two preparations had similar properties. The active protein was resolved into at least two forms with different isoelectric points, as judged by the formation of acid-soluble fragments. The major component of the ac- tivity had an isoelectric point at pH 7.02, and the minor component at pH 6.86. However, other properties of these two forms were similar.

The final preparations of the gastric and uterine acid DNases were free of other types of DNase (DNases I, III, and IV), acid and alkaline phosphatases, acid and basic RNases, and nonspecific phosphodiesterase. Both enzymes had a molecular weight of approximately 38,000, as deter- mined by gel filtration on Sephadex G-100, and showed maximum activity at pH 5.2 to 5.5 in 33 mM acetate buffer. They did not require divalent cations for activity, and hydro- lyzed native, double-stranded DNA about 10 to 20 times faster than heat-denatured DNA. They did not act on RNA or calcium bis(p-nitrophenyl) phosphate, indicating that they did not contain intrinsic phosphodiesterase activity. The products formed from native DNA were 3’-phosphoryl- and 5’-hydroxyl-terminated oligonucleotides. The average chain length of the limit digests with these enzymes was approximately 10 and the predominant species were longer than hexanucleotides.

There has been considerable work on deoxyribonuclease, and various kinds of endo- and exonucleases from mammalian tissues and microorganisms have been purified and characterized (l-5). In man DNases have been studied in serum (6-9) and urine (10) and in the colon (II), uterus (12), leukocytes (13), and tumors (14), to examine the enzyme activities in various disease states. In these investigations however, unpurified or only partially

* This work was supported by research grants from the Ministry of Education of Japan.

purified preparations were used, so differences in the enzyme activities in various conditions were uncertain from the results1 To clarify the behavior of DNase under normal and abnorma. conditions and to understand the pathogenrtic significance of its change in the cells, precise analyses must be made using highly purified preparations.

Four types of DNase have so far been characterized in mam- malian tissues (l-4, 15, 16). One of these, acid DNase (DNase II), was purified from hog spleen by Bernardi and his co-workers (3, 4), and its properties were studied in detail. The enzyme is located in lysosomes (17) and recently Dulaney and Touster (18) also obtained a highly purified preparation from rat liver ly- sosomes. The lysosomal extract degrades DNA to nucleoside and inorganic phosphate, but the real function of this reaction in the cells is still obscure (19).

The present paper describes a procedure for purification of an acid DNase from human tissues. Gastric mucosa and cervix uteri, which differ in ontogenic origin, were used as enzyme sources. It was found that the enzymes from these two tissues had essentially similar properties: they existed in at least two forms with different isoelectric points and both degraded native DNA in an endonucleolytic manner into acid-soluble oligo- nucleotides terminating in 3’.phosphate. The following paper describes the hydrolytic actions of the two enzymes (20).

EXPERIMEh-TAL PROCEDURE

Materials

Escherichia coli [32P]DNA and its thermally denatured form were obtained as described previouslv (21). Ribosomal RNA containing [5-3H]uridine was prepared from E. coli by extraction with hot nhenol. as described bv Takeda and Yura (22). Alkaline phosphatase, snake venom phosphodiesterase, and 5’-nucleotidase were partially purified from commercial preparations from Worth- ington Biochemical Corp. as reported before (21). Hog spleen DNase II and spleen phosphodiesterase were also obtained from Worthington. Polynucleotide kinase was a gift from Drs. K. Shimizuand M. Sekiguchi (Faculty of Science, Kyushu Univer- sitv). It had been ureoared from E. coli infected bv bacterio- phage Tb by the method of Richardson (23). Crystalline bovine plasma albumin, Fraction V, was obtained from Armour, soybean trypsin inhibitor and egg albumin (twice crystallized) from Nu- tritional Biochemicals, and cytochrome c from Boehringer Mann- heim.

Calcium bis(p-nitrophenyl) phosphate and p-nitrophenyl phos- phate were products of Daiichi Pure Chemical Co. (Tokvo, Japan). DEAE-cellulose and phosphocellulose were purchased from Brown

3884

by guest on January 30, 2020http://w

ww

.jbc.org/D

ownloaded from

Fraction

I. Crude extract. II. DEAE-cellulose.

III. Phosphocellulose. IV. Sephadex G-100. V. Electrofocusing A B

3885

TABLE I

Purification of acid DNases from human gastric mucosa and cervix uteri

Volume

ml mg 400 6,400 550 1,860

70 17.5 55 7.5

14) 300 31,460

5,720 4,125

units/mg % ml mg units units/mg %

2.2 100 400 2,400 6,000 2.5 100 16.9 220 520 430 7,200 16.7 120

326.8 40 95 22.5 2,280 101 38 550 29 46 7.2 1,800 250 30

4 0.12 833 6,940 6 4 0.08 312 3,700 5 6 0.14 1,920 13,714 13 5.5 0.09 850 9,430 14

Total protein

Human gastric DNase II

Total activity pecific activity Yield

I Human uterine DNase II

Volume Total protein Total activity

Specific activity

Co., DEAE-Sephadex A-25 and Sephadex G-100 from Pharmacia, and an ampholyte preparation, pH 6 to 8, from LKB Instrument’s, Inc., Stockholm, Sweden.

Methods

Assay of Acid DNase-Production of acid-soluble radioactivity from [32P]11NA was routinely determined in reaction mixture (0.3 ml) containing 10 pmoles of sodium acetate buffer, pH 5.4, 0.6 Mmole of EDTA, 10 nmoles of E’. coli [32P]DNA, and 0.5 to 5 units of enzyme. Twenty micrograms of bovine plasma albumin (in freshly prepared aqueous solution) (24) were also added to the medium, when the enzyme preparation had a specific act,ivity of more than 1000. The enzyme preparat,ion was diluted with TE solution.’ The mixture was incubated for 10 min at 35”, the acid-soluble fraction was then obtained by adding cold perchloric acid, and the radioactivity of an aliquot was determined as de- scribed previously (21). The acid-soluble fraction from control mixture without enzyme contained less than 0.5yo of the added radioactivity. One unit of acid DNase activity is defined as the amount of enzyme catalyzing the production of 1 nmole of acid- soluble 3*P in 10 min under the above conditions,

Assay of Other Enzyme Activities-I>Nase I activit,y was meas- ured in reaction mixture (0.3 ml) containing 30 pmoles of Tris-HCl buffer, pH 7.5, 1.5 Mmoles of MgClz, 10 nmoles of E. coli [32P]I>NA, and enzyme samples. Alkaline DNase III and IV were assayed by the procedure of Lindahl (15, 16) with minor modifications as follows. The reaction mixture (0.3 ml) contained 10 pmoles of Tris-HCl buffer, pH 8.5, 1 rmole of MgCl?, 3 pmoles of mercapto- ethanol, enzyme sample, and 10 nmoles of denatured E. coli [““PI- DNA for DNase III, or native E. coli [32P]DNA for DNase IV.

Acid RNase activity was assayed under similar conditions to acid DNase, except that the substrate was 100 nmoles of E. coli ribosomal [3H]RNA and an aliquot of the acid-soluble fraction obtained after incubation was added to Triton scintillation fluid for counting the radioactivity. Basic IlNase activity was assayed essentially like acid RNase, except that 20 /*moles of potassium phosphate buffer, pII 7.5, and 2 hmoles of MgClz were used in place of acetate buffer and EDTA.

Acid and alkaline phosphatase activities were determined with p-nitrophenyl phosphate as substrate under the experimental conditions used by Bernardi for measuring acid phosphatase (25) and those described in the Worthingt,on Corp. Catalogue for alka- line phosphatase. Phosphodiesterase activity was measured by the method of Burton and Petersen (26).

Ide?ttification of Acid-soluble Digestion Products-The suscepti- bility of the digestion products to alkaline phosphatase and 5’- nncleotidase was studied by measuring the formation of 32P- labeled material not adsorbed on Norit, as previously described (21). Hydrolysis of the limit digest of DNA was performed with snake venom and spleen phosphodiesterases using the reaction systems described by Laskowski (27) and Bernardi (28), respec- tively. Labeling of the 5’.terminal of the limit digests with 32P was carried out by the action of polynucleotide kinase, as reported by Richardson (23). The digestion products of DNA were ana- lyzed by chromatography on a DEAE-Sephadex A-25 column (21).

1 The abbreviation used is: TE solution, 20 mM Tris-HCl buffer, pH 7.5, and 2 mM EDTA.

Yield

Other Ilfethods-The molecular weight of the purified enzyme was measured by the method of Andrews (29). Polyacrylamide gel electrophoresis was performed at pH 8.3 using the method of Davies (30). Protein concentration was determined by the method of Lowry et al. (31) with crystalline bovine plasma albu- min as a standard. Inorganic phosphorus was determined as described by Lowry and Lopez (32) 32P was counted in an Aloka model TDC-307 gas flow counter (Japan Radio Co., Tokyo). 3H was measured in a Beckman model LS-100 liquid scintillation counter.

RESULTS

Purijkation of Enzymes

All operations were conducted at O-4”, unless otherwise stated. The purification procedures and results of a typical preparation

are summariztd in Table I.

Crude &tracts-Stomachs were obtained by gastrectomy from patients with a duodenal ulcer or cholelithiasis, and uteri by total hysterectomy from patients with prolapse or myoma uteri. Microscopically none of the samples showed malignant changes. Soon after gastrectomy, the stomach was washed with running water and the mucosa was scraped off from the wall with a razor at the level of the muscularis submucosae. Im- mediately after hysterectomy the cervical portion of the uterus was washed with running tap water and a wedge of tissue was cut, including a large area of surface epithelium with a minimal amount of underlying layers. Acid DNase activity was scarcely

detectable in the muscular layers of the two organs. Sub- sequent procedures were the same for both materials.

The specimens were washed three times in ice-cold 0.5y0 NaCI-KC1 solution, wiped with filter paper, weighed, and cut into small pieces. Tissue was then homogenized for 1 min in a

Waring Rlendor with 10 volumes of cold TE solution. The homogenate was filtered through gauze and the filtrate was centrifuged at 8000 x g for 10 min. The precipitate (cell debris and nuclei) contained little acid DNase activity and was dis- carded while the supernatant fluid with enzyme activity (Frac- tion I) was stored in a deep freeze as crude extract.

DEAE-cellulose Fractionation-Fraction I (400 ml) was thawed

and applied to a column of DEAE-cellulose (3 x 30 cm) pre- viously equilibrated with TE solution. The column was eluted with the same buffer at a flow rate of 100 ml per hour. Frac- tions of 10 ml of the eluate were collected and assayed for DNase activity. Fractions containing more than 5 units of enzyme activity per ml were pooled (Fraction II).

Phosphocellulose Chromatography-Fraction II was dialyzed against 10 volumes of 20 mM collidine-HCl buffer, pH 6.5, con- taining 2 mM EDTA for 12 hours. It was then applied to a

by guest on January 30, 2020http://w

ww

.jbc.org/D

ownloaded from

3886

d 6

‘i 100

>; 4 C ., 5 0 I

L * 300 Uterine DNase al

Fraction number

E ; 100 Z 5 -

z 0

; 300

ii 01 = 200

E

1 i; 100 a

0 0 20 40 60 60 100

Fraction number

Fraction V-A

40

30

Fraction V-B

O 0 10 20 30 40 50 TOP m m of gel l%om

FIG. 1 (Zefl). Isoelectric focusing of fractions of gastric mucosa and cervix uteri from Sephadex. The experimental conditions are described in the text. The acid DNase activity (O--O) and pH (0) of each fraction were measured as described under “Methods.”

FIG. 2 (center). Isoelectric focusing of peaks a and b of gastric enzyme from phosphocellulose. The experimental conditions are described in the text. O----O, acid DNase; 0, pH.

FIG. 3 (right). Polyacrylamide gel electrophoresis of fra,ctions V-A and V-B of gastric enzyme. Equal volumes (0.1 ml) of 30%

column of phosphocellulose (1.5 X 50 cm) equilibrated with 1 liter of the same buffer. The column was washed with 100 ml of the same buffer and then eluted with 800 ml of a linear gra- dient of 0.2 to 0.6 M NaCl in the same buffer. The flow rate was 50 ml per hour, and fractions of 5 ml were collected. Acid DNase activity was eluted in a first, minor peak, a, and a second, major peak, b, immediately afterward, with approximately 0.35 M and 0.40 M NaCl, respectively. The ratio of activities in the two peaks was fairly constant. Fractions containing more than 5 units of enzyme activity per ml were pooled (Fraction III). Phosphodiesterase and acid phosphatase were not adsorbed on the column. and were eluted with the washing solution. Acid RNase was eluted just after Peak b and overlapped it.

Gel Filtration on Sephadex G-100-Fraction III was dialyzed against a solution of 30% polyethylene glycol in TE solution to reduce its volume and then against 0.1 M NaCl-0.02 M sodium acetate buffer, pH 5.5, for 6 hours. The solution obtained was centrifuged for 10 min at 20,000 x g to remove the considerable amount of insoluble material formed during the concentration procedure and dialysis. The supernatant solution (5 ml) was applied to a column of Sephadex G-100 (2 x 110 cm) previously washed with the same buffer. After allowing the enzyme solu- tion to enter the gel, the column was eluted with the same buffer at a flow rate of 7 ml per hour, and 3.5-ml fractions were col- lected. The peak of acid DNase activity emerged after passage of approximately 200 ml of the eluting solution, and fractions containing enzyme activity of more than 5 units per ml were pooled (Fraction IV). Acid RNase was eluted afterward and was completely separated from acid DNase.

Isoelectric Focusing-Fraction IV was dialyzed against a solu- tion of 30% polyethylene glycol in TE solution to reduce its volume to about 10 ml, and then against 2 liters of 1% glycine

sucrose and enzyme solution were layered onto two gels; 7.5% pore gels were used in Tris glycine (pH 8.3). Electrophoresis was run at 3 ma per tube at 4” for about 120 min. One gel was cut into approximately l-mm slices, and each slice was placed in 0.2 ml of TE solution. The enzyme was eluted over a period of 48 hours at O”, and each extract was assayed as described under “Methods,” except that the incubation period was 30 min. The other gel was stained with Coomassie brilliant blue and destained by the leaching method.

solution for 12 hours. The dialyzed fraction was subjected to electrophoresis in a llO-ml LKB electrofocusing column, as de- scribed by Vesterberg and Svensson (33). Carrier ampholytes were used to establish the pH gradient. The electrophoresis was run at 700 volts for 60 hours at 4”. Then fractions of 1 ml were collected and their DNase activity and pH were measured. Two active peaks were observed (Fractions V-A and V-B) (Fig. 1). The fractions in each peak were pooled and dialyzed against TE solution for 24 hours with several changes of the dialyzing solution. Then the preparations were stored at - 20”.

The final yields of human gastric and uterine DNase in the major fraction, Fraction V-B, were about 13 and 14% of the activities present in the crude extracts, representing purifica- tions of 6000- and 4000-fold, respectively.2

Multiple Forms of Enzymes and Homogeneities of Enzyme Preparations-During purification of the enzymes from stomach, chromatography on phosphocellulose yielded two active frac- tions, a and b, which overlapped. Moreover, when these had been combined and subjected to gel filtration isoelectric focusing again showed the presence of two forms of the enzyme (Fig. 1). When the two active peaks, a and b, from the phosphocellulose column were each subjected to isoelectric focusing the electro- phoretograms indicated that the protein in the first peak, a, was mainly located in Fraction A, while that in the second, major peak, b, was in Fraction B (Fig. 2). On gel electrophoresis, Fractions V-A and V-B both gave two bands (a! and B), ex-

2 During purification of the DNases the total activities of the DEAE-cellulose fractions were much greater than the initial ac- tivities in crude extracts. Preliminary experiments showed that a fraction inhibiting DNase activity was eluted when the DEAE- cellulose column was washed with 0.5 M NaCl. Accordingly, the increase in activity might be due to removal of some concomitant inhibitor.

by guest on January 30, 2020http://w

ww

.jbc.org/D

ownloaded from

3887

hibiting acid DNase activity with about one-third the mobility of that of bromphenol blue, the tracking dye (Fig. 3). The protein of the or-band was found more in Fraction V-A while P-band protein was found in Fraction V-B, although definite evidence for the correspondence of a-band protein to enzyme in Fraction V-A and P-band protein to that in Fraction V-B was not obtained, due to the limited amount of material available.

Fractions V-A and V-B were both about 80% homogeneous, and were contaminated with a different kind of protein possess- ing no enzyme activity. Similar results were obtained with Fractions V-A and V-B from uterus.

Properties of Enzymes

The properties of the enzymes in Fractions V-A and V-B of the two tissues were studied, and the results obtained with the latter fractions, are mainly described below. Fractions V-A had less activity but except for a difference in their isoelectric point their properties were similar.

St&My-The final preparations of the two enzymes, as well as less purified preparations and crude extracts, were stable at 0” for at least 2 months, and could be stored at -20” for more than 1 year without appreciable loss of activity. The heat sensitivities of the enzymes were studied by preincubating the preparations at various temperature for 20 min. The enzymes were rather stable up to 55”, but lost 50% of their activity at 60” and were inactivated completely at 70”.

Eflect of pH and Temperature on the DNase Reaction-The acid DNases from stomach and uterus exhibited maximum ac- tivity at pH 5.2 to 5.5 in reaction mixture containing 33 mM sodium acetate buffer, and at around pH 6.4 in N-tris(hydroxy- methyl)methyl-2.aminoethanesulfonic acid (TES buffer) of the same concentration. The pH curves indicated rather slow de- crease of the reaction rate on the acidic side of the optimal range.

The optimal temperature for the two DNase fractions from the two tissues, measured from the initial reaction rates, was about 45”, wit.h a gradient decrease of t.he reaction rate at lower tem- peratures.

Inhibition and Activation-The activity of acid DNase is known to vary with t.he ionic species present in the reaction medium and the ionic strength. Hydrolysis of native DNA by the purified gastric and uterine acid DNases at the optimal pH was inhibited by approximately 507, when MgZ+ (1 mM), MnZf (1 mM), Ca2+ (1 mM), or CL?+ (1 mM) was added to the reaction medium in place of EDTA. Fe3+ (0.1 InM) inhibited the reac- tion completely, while Na+ (1 mM) did not cause any reduction of the hydrolytic rate. On the other hand, EDTA did not in- fluence the velocity of the reaction, indicating that divalent cations were not required for activity. As reported with hog spleen DNase II (34), POd3- (1 mnf) and SO,2 (1 mM) were also slightly inhibitory to gastric and uterine DNases (about 20yo loss of act,ivity). 2-Mercaptoethanol, at a concentration of 2%, did not affect the reaction velocity of either enzyme, but bovine plasma albumin, at a level of 0.3 mg per ml, was stimulatory. The effect of the latter may be to stabilize the enzyme protein.

Isoelectric Point-The acid DNases from stomach and uterus have the same isoelectric points: Fractions V-A at pH 6.86 and Fractions V-B at pH 7.02 (Fig. 1).

Molecular Weight of Enzyme-Although the DNase activity from gastric mucosa was found in two fractions on phospho- cellulose chromatography, isoelectric focusing, or disc gel elec- trophoresis, the activities in these fractions were eluted as a single peak from Sephadex G-100, indicating that they have a

similar molecular weight. By comparison of the relationship between the elution volumes and the logarithms of the mo- lecular weights of standard proteins (egg albumin, soybean tryp- sin inhibitor, and cytochrome c) and the elution volumes of the final preparations (Fractions V-A and V-B), it was estimated that the gastric acid DNase is a mixture of molecules with a mo- lecular weight of around 38,000. The same results were ob- tained with hog spleen DNase II and the two forms of DNase from cervix uteri.

Effect of Enzyme Concentration on DNase Reaction-With purified enzyme preparations the reaction rate was proportional to the amount of protein employed up to a level of enzyme which permitted rapid degradation of approximately 30 to 40% of the DNA substrate.

Substrate Specijkity of Enzyme-The reactions of the DNases from the two tissues with native and denatured DNAs were followed (Fig. 4). Both enzymes initially degraded native DNA about 10 to 20 times faster than thermally denatured DNA. Hydrolysis of native DNA proceeded quickly until ap- proximately 30 to 60% of the DNA had been broken down and then more slowly and finally the reaction stopped when 70 to 90% of the DNA had been degraded. E. coli ribosomal RNA was not hydrolyzed by the purified enzyme preparations, as described below.

Identijication of Acid-soluble Digestion Products-The reaction products of the enzymes from the two tissues were assayed by measuring the susceptibility of the products to E. coli alkaline phosphatase at different stages during the digestion of native DNA. At the point where hydrolysis of the DNA reached a limit, 10% of the total acid-soluble phosphate was sensitive to the action of alkaline phosphatase. Thus the average chain length of the product was calculated as approximately 9 to 10.

The products formed from native DNA by the enzymes were separated according to their chain length using a column of DEAE-Sephadex A-25 eluted with 7 M urea-O.02 M Tris-HCI buffer, pH 7.5, containing a gradient of increasing NaCl con- centration. Table II shows the proportions of small oligo- nucleotides of from mono- to pentanucleotides after exhaustive digestiou of DNA. The two enzymes produced similar amounts of oligonucleotides and no mononucleotides. The amount of dinucleotides was small, and most of t.he digest consisted of larger oligonucleotides which could not be separated by this technique. These results suggest that the enzymes act endo- nucleolytically.

The limit digests of DNA by gastric and uterine DNases were converted to mononucleotides and mononucleoside diphosphates by treatment with excess snake venom phosphodiesterase. Of the 32P-labeled products, 80 and 98% were susceptible to the actions of purified 5’nucleotidase and E. coli alkaline phos- phatase, respectively. When the limit digests were treated with spleen phosphodiesterase, approximately 95% of the total 32P became susceptible to alkaline phosphatase. In addition, the limit digests of DNA were labeled with [Y-~~P]ATP by me- diation of polynucleotide kinase to the same extent, with and without previous incubation with alkaline phosphatase. There- fore, the reaction products were concluded to contain terminal 3’-phosphoryl and 5’-hydroxyl groups.

Absence of Other Enzyme Activities-Less than I nmole of p-nitrophenol was liberated when 10 pmoles of p-nitrophenyl phosphate and calcium bis(p-nitrophenyl) phosphate were in- cubated with 10 units of the final preparations of the two en- zymes at pH 5.4 or 7.4 for 2 hours, or at pH 5.4 for 24 hours. In addition, when 100 nmoles of native DNA were incubated

by guest on January 30, 2020http://w

ww

.jbc.org/D

ownloaded from

3888

60

0 60 N W-J

20

0 30 60 90 120 150

Time in minutes

FIG. 4. Effect of the structure of substrate DNA. Two series of reaction mixtures were set up as described under “Methods,” with native and denatured DNA as substrates and 4 units of the purified DNases. Separate tubes were incubated at 35” for the indicated times, and then assayed for acid-soluble radioactivity. After incubation for 65 min, 4 units more enzyme were added to each reaction mixture and incubations were continued for totals of 90, 120, and 150 min. O-0, native DNA; O-0, de- natured DNA substrate.

TABLE II

ZdentiJication of digestion products of native DNA by gastric and uterine DNases

The reaction mixtures (0.6 ml) contained 40 pmoles of sodium acetate buffer, pH 5.3, 1.2 @moles of EDTA, 100 nmoles of [“zP]- DNA, and 20 units of each enzyme. The mixtures were incubated at 35”, and after 120 min, 20 units more of enzyme were added, and incubation was continued for a total of 360 min. Then 0.5 ml (2 X lo6 cpm) of each digest was applied to a DEAE-Sephadex A-25 column.

Products

Mononucleotides.. Dinucleotides. Trinucleotides. Tetranucleotides., Pentanucleotides., >Pentanucleotides

Total 12P &ted

Gastric DNase Uterine DNase

%

0 0 0.8 0.5 2.2 2.5 3.4 4.6 3.2 4.2

90.4 88.2

with 20 units of the enzymes for 6 hours at pH 5.3, less than 0.01 nmole of the mononucleotide was released (Table II).

When 10 units of the final preparations of the two enzymes were incubated with 100 nmoles of E. coli [3H]RNA for 2 hours at pH 5.4 or 7.5, less than 1% of the substrate was converted to acid-soluble products. These results indicate that the purified preparations were free from detectable nonspecific phospho- diesterase, acid or alkaline phosphatase or acid or basic RNase activities, and were real DNases capable of hydrolyzing native DNA preferentially.

In crude extracts of the two tissues the activities of DNase

I and IV were too weak to be detected, but there was a con- siderable amount of DNase III activity. However, this was removed from t.he acid DNase fraction by DEAE-cellulose chro- matography. When 10 nmoles of native or denatured DNA were incubated with 5 units of the final preparations of the two enzymes for 1 hour at pH 7.5 or 8.5, less than 0.01 nmole of acid- soluble nucleotides was formed, indicating that the preparations were not contaminated with these DNases.

DISCUSSION

Highly purified preparations of acid DNase were obtained from homogenates of human gastric mucosa and cervix uteri as described in this paper. These preparations appeared about 80% homogeneous by disc electrophoresis, but were free from other nucleolytic enzymes.

All of t.he acid DNases so far found in mammalian tissues appear to have similar properties (1, 3, 4, 18). The acid DNases reported here are the first enzymes to be purified from human materials of different embryonic origins, and they also had es- sentially the same properties.

Bernardi and Griffe (34) observed that a highly purified prep- aration of hog spleen DNase II catalyzed the slow hydrolysis of p-nitrophenol from calcium bis(p-nitrophenyl) phosphate and suggested that the phosphodiesterase activity was an intrinsic property of the acid DNase molecule. Hodes and his co-workers (35) and Slor (36) succeeded in separating these two activities, but Sicard et al. (37) failed to confirm their results. Recently, Dulaney and Touster (18) obtained a preparation of acid DNase from rat liver lysosomes which appeared homogeneous on elec- trophoresis, and found that it did not show nonspecific phos- phodiesterase activity. In this work, during purification of human acid DNases, phosphodiesterase was clearly separated from acid DNase fractions by phosphocellulose chromatography, and the final preparations of DNases were completely free of phosphodiesterase activity. Thus, our results indicate that the two activities belong to different protein molecules.

The human DNases from gastric mucosa and cervix uteri were both resolved into at least two forms, which had different isoelectric points but otherwise similar properties. However, it is possible that formation of active degradation products during the chromatographic procedure may lead to multiplicity of enzyme forms. Multiple forms of acid DNase have been ob- tained from hog spleen (38), rat liver (18)) mouse liver (39), hog liver (40), and beef brain (41). Bernardi et al. (38) prepared two active components of DNase with similar properties from hog spleen by chromatography on CM-Sephadex. They sug- gested that the acid extraction involved in the purification pro- cedure resulted in the appearance of a minor component, which was formed from the major component by acid hydrolysis and which could be separated from the latter by chromatography.

In the procedure described here for purification of human acid DNases, drastic treatments were avoided and preparations were kept close to neutrality to assure the chemical integrities of the molecules throughout the purification. Each peak of activity from the phosphocellulose column gave a single peak at the identical position when rechromatographed on the same column, and the enzymes in the two final fractions (Fractions V-A, V-B) were recovered in the same positions on repeated electrofocusing. Moreover, varying the time of treatment in a Waring Blendor from 30 s to 15 min and repeated freezing and thawing of crude extracts did not affect the pattern of DNase activity at all. Thus it seems unlikely that proteolytic enzymes

by guest on January 30, 2020http://w

ww

.jbc.org/D

ownloaded from

3889

in the materials cause any hydrolysis of the DNase molecule 20. during the purification procedure. However, the properties of 21, these two active forms were very similar. Therefore further investigations are needed on whether the multiple forms of the 22. enzyme in human materials are due to proteolytic modification 23. of the molecules, either at the ends of the chain or by intra- molecular cleavage, or whether the enzyme activities reside in 24.

different proteins. 25.

REFERENCES

1. 2.

3. 4.

5.

6.

7. 8.

9. 10.

11. 12.

13. 14. 15.

16.

17.

18.

19.

BERNARDI, G. (1966) in Procedures in Nucleic Acid Research (CANTONI, G. L., AND DAVIES, D. It., eds) pp. 236-244, Harper & Row, New York

LASKOWSKI, M., SR. (1967) Advan. Enzymol. 29, 165-220 26. BURTON, K., AND PETERSEN, G. B. (1960) Biochem. J. 76, LASKOWSKI. M.. SR. (1971) in The Enzumes (BOYXR. P. D.. 17-27

ed) Vol. 4, pp.‘289-311, Academic Press, New York ’ ’ 27. BERNARDI, G. (1968) Advan. Enzymol. 31, l-49 BERNARDI, G. (1971) in The Enzymes (BOYF,R, P. D., ed) Vol.

4, pp. 271-287, Academic Press, New York 28. LEHMAN, I. It. (1971) in The Enzymes (BOYER, P. D., ed) Vol.

4, pp. 251-270, Academic Press, New York WROBLEWSKI, F., AND BODANSKY, 0. (1950) Proc. Sot. Exp. 29.

Biol. Med. 74, 443-445 30. KURNICK, N. B. (1953) Arch. Rio&em. Biophys. 43, 97-107 31.

HERRIOT, IL M., CONNOLY, J. H., AND GUPTA, S. (1961) Nature 189, 817-820 32.

DOCTOR, V. M. (1971) PTOC. Sot. Exp. Biol. Med. 138, 241-243 Z~LLNF,R. E. J.. MUELLER. W. E. G.. ZAHN. It. K.. BREHM. 33.

G., AND KORT~NG, G. W: (1971) Klin. W&r. 49, 1290-1294 ’ DALE, R. (1965) Clin. Chim. Acta 11, 547-556 34.

GOLDBERG, D. M., AND PITTS, J. F. (196G) Brit. J. Cancer 20, 729-742 35.

ESCHENBACH, C. (1971) Klin. Wochenschr. 49, 949-957 DAOUST, R., AND AMANO, H. (1963) Cancer Res. 23, 131-134 36.

LINDAHL, T., GALLY, J. A., AND EDELMAN, G. M. (1969) J. Biol. Chem. 244, 5014-5019 37.

LINDAHI,, T., GALLY, J. A., AND EDELMAN, G. M. (1969) Proc. Nat. Acad. Sci. U. S. A. 62, 597-603 38.

DE DUVF,, C. (1969) in Lysoiomes in Biology and Pathology (DINGLE, J. T., AND F&L, H. B., eds) -Vol. 1, pp. 3-40, 39. HODES, M. IX., YIP, L. C., AND SANTOS, F. R. (1967) Enzymo- North-Holland Publishine Co.. Amsterdam Zogia 32, 241-254

DULANF,Y, J. T., AND TOUS;ER, 6. (1972) J. Biol. Chem. 247, 40. CORDONNIER, C., AND BERNARDI, G. (1969) Can. J. Biochem.

1424-1432 46, 989-995 ARSENIS, C., GORDON, J. S., AND TOUSTER, 0. (1970) J. Biol. 41. ROSENBLUTH, IL, AND SUNG, S.-C. (1969) Can. J. Biochem. 47,

Chem. 246, 205-211 1081-1088

TSUBOTA, Y., YAMANAKA, M., AND TAKAGI, Y. (1974) J. Biol. Chem. 249,389&3894

ANAI, M., HIRAHASHI, T., AND TAKAGI, Y. (1970) J. Biol. Chem. 246, 767-774

TAKE:DA, Y., AND YURA, T. (1968) Virology 36, 55-62 RICHARDSON, C. C. (1965) Proc. Nat. Acad. Sci. U. S. A. 64,

158-165 ANAI, M., HARaGUCHI, H., AND TAKAGI, Y. (1972) J. Biol.

Chem. 247, 193-198

LASKO~SKI, M., SR. (196G) in Procedures in Nucleic Acid Re- search (CANTONI, G. L., AND DAVIES, D. R., eds) pp. 154- 187, Harper & Row, New York

BERNARDI, G. (1966) in Procedures in Nucleic Acid Research (CANTONI, G. L., AND DAVIES, D. It., eds) pp. 144-153, Harper & Row, New York

ANDRI~WS, P. (1964) Biochem. J. 91, 222-233 DAVIXS, B. J. (1964) Ann. N. Y. Acad. Sci. 121, 404-427 LOWRY, 0. H., ROSE~ROUGH. N. J.. FARR. A. L.. AND RAN-

DALL; It. J. i1951) J. Biol. ‘Chem. i93, 26k275 ’ LOWRY, 0. H., AND LOPEZ, J. A. (1946) J. Biol. Chem. 162,

421-428 VESTERBERG, O., AND SVENSSON, II. (1966) Acta Chem. &and.

20, 820-834 BERN.~RDI, G., AND GRIFFE, M. (1964) Biochemistry 3, 1419-

1426 SWENSON, M. Ii., AND HODES, M. E. (1969) J. Biol. Chem. 244,

1803-1807 SLOR, H. (1970) Biochem. Biophys. Res. Commun. 38, 1084-

1090 SICARD, P. J., OBRENOVITCH, A., AND AUBEL-SADRON, G.

(1972) Biochim. Biophys. Acta 268, 468-479 BERNARDI, G., BERNARDI, A., AND CIIF,RSI, A. (1966) Biochim.

Biophys. Acta 129, l-11

by guest on January 30, 2020http://w

ww

.jbc.org/D

ownloaded from

Okumura, Shibanosuke Katsuki and Yasuyuki TakagiMasayoshi Yamanaka, Yoriaki Tsubota, Motoaki Anai, Koji Ishimatsu, Makoto

and Cervix UteriPurification and Properties of Acid Deoxyribonucleases of Human Gastric Mucosa

1974, 249:3884-3889.J. Biol. Chem. 

  http://www.jbc.org/content/249/12/3884Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/249/12/3884.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on January 30, 2020http://w

ww

.jbc.org/D

ownloaded from