CLIN. CHEM. 39/3, 448-452 (1993)

448 CLINICAL CHEMISTRY, Vol.39,No.3,1993

Measurement of Deoxyribonuclease I Activity in Human Tissues and Body Fluids bya Single Radial Enzyme-Diffusion MethodDaita Nadano, Toshihiro Yasuda, and Koichiro KishI’

In the single radial enzyme-diffusion (SRED) method forassay of deoxyribonuclease I, a precisely measured vol-ume of the enzyme solution is dispensed into a circularwell in an agarose gel layer in which DNA and ethidiumbromide are uniformlydistributed. A circular dark zone isformed as the enzyme diffuses from the well radially intothe gel and digests substrate DNA. The diameter of thedark circleof hydrolyzedDNA increasesinsizewithtimeandcorrelates linearly with the amount ofenzyme appliedto the well.Thus,the SRED can be usedforquantitationofdeoxyribonuclease I witha limitof detection of 2 x 106unit.Thiscorresponds to 1 pg of purified urine deoxyribo-nucleaseI. We measuredthedeoxyribonuclease I activityof 17 different human tissues and body fluidsfromhealthydonors. Unne samples showed the greatest activity, 6.0 ±

2.2 kilo-units/g protein (mean ± SD). Serumdeoxyribo-nuclease I activity was 4.4 ± 1.8 units/L.

IndexIng Terms: nuclease assay - fluorescence ‘ isoen-zymes

DeoxyribonucleaseI (DNase I; EC 3.1.21.1) was thefirst enzyme to be recognized as specific for DNA, andbovine pancreatic DNase I has been extensively charac-terized in protein-chemicalstudies (1, 2). DNase I isgenerally considered mainly a digestive enzyme. How-ever, DNase I activity has been found in nondigestivetissues such as human kidney and liver (3, 4), whichsuggests that the biological role of DNase I has not beenfully clarified. Some reports have identifiedother func-tions for this enzyme: (a) DNase I binds specifically toG-actin and blocks actin polymerization (2, 5). (b) Ahuman urine-derived interleukin 1 inhibitor showsclose homology with DNase 1(6). (c) DNase I exhibitswell-defined specificity for certain kinds of DNA struc-tare (7). (d) A significant decrease of DNase I activityhas been observed in the urine of patients with xero-derma pigmentosum (8). (e) DNase I is effectiveinreducing the viscosity of sputum from patients withcystic fibrosis (9). Although DNase I activity has beenidentified in several tissues and body fluids (3-5, 10-14), the distribution of the enzyme in the human bodyhas not been systematically and quantitatively investi-gated. This is due, in part, to the lack of a simple andsensitive quantitative assay method.

In this report we describe a new microdeterminationmethod (single radial enzyme-diffusion, SEED) for

Department of Legal Medicine, Fukui Medical School, Mat-suoka-cho, Fukui 910-11, Japan.

1 Author for correspondence.Received July 7, 1992; accepted October 1, 1992.

DNase I and report on the use of this method for mea-suring the enzyme in human tissues and body fluids.

MaterIals and MethodsReagents

We purified human urine DNase I (1.8 x iO kilo-units/g protein) and DNase II (EC 3.1.22.1; 4.6 x iOkilo-units/g protein) as described previously (5, 15).Rabbit skeletal muscle G-actin was prepared by themethodof SpudichandWatt (16). We purchased snakevenom phosphodiesterase I and bovine spleen phospho-diesterase II (type 1-SA) from Boehringer Mannheim(Mannheim, Germany) and Sigma Chemical Co. (St.Louis, MO), respectively.

Biological Samples

Human tissue samples, including brain, heart, lung,liver, pancreas, kidney, spleen, and submandibulargland, from five to seven individuals (ages 9 days to 74years) were obtained at autopsy within 24 h postmortem.Thymus glands were obtained from eight individuals(ages newborn to 9 years) at autopsy within 60 h post-mortem. The deceased individuals were kept at roomtemperature until autopsy. Placentas were obtained atdelivery. All tissue samples were shown to be histologi-cally normal and were stored at -80 #{176}Cuntil used.Tissueextract preparations were done at 0-4 #{176}C;tissues wereminced, washed with saline to remove excess blood, andthen homogenized in 50 mmol/L potassium phosphatebuffer, pH 6.5. After centrifugation at 20 000 x g for 15mm, ainmonium sulfate was added to the supernate togive 85% saturation. The resulting precipitate was dis-solved in 20 mmol/L potassium phosphate buffer, pH 6.5,dialyzed in the same buffer, lyophilized, and stored at-20 #{176}Cuntil used. Serum samples were prepared fromvenous blood collected from healthy unrelated Japanesedonors(20 women, ages22-35 years; 81 men, ages 22-51years). Bloodwas centrifuged at 1500 x g for 10 miii, andthe supernate was stored at -80 #{176}Cuntil used. Erythro-cyte sampleswere obtained from healthy unrelated Jap-anese donors (five women and six men). Erythrocyteswere washed three times in isotonic saline with removalof as much of the supernate as possible,and the resultingpacked cells (-1 x iO’#{176}cells) were lysed by freezing andthawing. Leukocytes were prepared from heparinizedblood (10 mLfsubject) by gradient centrifugation onFicoll-Paque (Pharmacia LKB, Uppsala, Sweden) accord-ing to the manufacturer’s instructions and then washedthree times with 14.5 mmol/L TrisHCl, pH 7.6, contain-ing 0.126 molJLNaCl, 0.55 mmol/L D-glucose,5.0 pniol/LCaCl2, 98 MmolfL MgCl2, and 0.54 mmol/L KC1; cellnumbers were then determined. The leukocytes were

CLINICAL CHEMISTRY, Vol. 39, No.3, 1993 449

then suspended in 10 mmol/L Tris-maleate, pH 6.5, lysedby somcation, and stored at -80 #{176}C.Semen and breastmilk were obtained from healthy volunteers. Urine sam-ples (24-h urine) were freeze-dried after ultrafiltration,concentration, and dialysis (17), and stored at -20#{176}Cuntil used. Saliva samples were collectedfrom 10 healthyindividuals without any stimulation, dialyzed againstdistilled water, lyophilized, and stored at -20 #{176}C.Weobtained informed consent from all of the donorsfollow-ing procedures in accordance with the Helsinki Declara-tion of 1975, as revised in 1983.

ONase I Assay by the SRED MethodThis method isbased on a previously reported method

for purification of human kidney DNase I (4). Wedissolved 10 gIL salmon testis DNA (type III, SigmaChemical Co.) in distilled water, stirring for 2-3 h atroom temperature. The solution was stored at 4#{176}C.Ethidium bromide (Boehringer Mannheim) was dis-solved in distilled water at 5 g/L. DNase I reactionbuffer was 0.1 mol/L sodium cacodylate, pH 6.5, with 20mmol/L MgCl2 and 2 mmol/L CaC12 (4, 5). Each assaygel plate was prepared as follows: (a) 5.1 mL of reactionbuffer was placed in a test tube; (b) solutions of 0.27 mLof DNA and 54 pL ofethidium bromide were added.Thissolution was mixed well and maintained at 50#{176}Cfor-10 mmn. A wide-bore pipet was used to transfer thehighly viscous DNA solution. (c) 5.4 mL of 20 g/L moltenagarose GP-36 (Nacalai Tesque, Kyoto, Japan) in dis-tilled water at 50#{176}Cwas added and immediately pouredonto a horizontal Agafix MSL sheet (Wako Pure Chem-ical Industries, Osaka, Japan) or agarose-coatedpolyes-ter GelBond film (Pharmacia LKB). The final volume(10.8 mL) was sufficient to prepare a gel plate measur-ing 7.5 x 13.5 cm and 2 mm thick. After congelation atroom temperature, a row of cylindrical sample wells(radius [r0] = 1.0 mm) with centers 15 mm apart waspunched in the gel. A 7.5 x 13.5 cm gel plate accommo-dated 32(4 x 8) samples. Unknown samples (2 izL) andserial dilutions of purified urine DNase I of knownactivity were placed in the wells with a micropipet. The2-pL sample solutions filled the wells almost to thebrim. The gel plate was incubated ina moist chamber at37#{176}C,and DNase I activity was observed periodicallywith an ultraviolet transilluminator at 312 mu (ModelTF.1OM; Vilber Lourmat, Marne La Vall#{233}e,France). Acircular dark zone formed as the DNase I diffusedradially from the well and digested the DNA substrate.Incubation was continued until test samples gave well-defined dark circles with radii (re) of 2.5-8.5 mm. Dif-ferences in length of 0.1 mm were determined with avernier micrometer. A standard curve was constructedby plotting log10DNase I activity against the diffusionradius (r - r0). The assay gels could be preserved for atleast 1 month after drying at 50#{176}C.

We examined the relation between the amount ofDNase I applied to the gel plate in the SRED method andthe size of the dark circle produced by the diffusedenzyme as follows: We prepared the gel plate as describedabove, except that cylindrical wells with two different

radii (r0 = 1.0 and 1.5 mm) were punched in the gel.Known units of purified human urine DNase I in differ-ent volumes (u8 = 2.0 to 6.0 1zL) of 10 mmol/L Tris-maleate, pH 6.5, containing 5 mmol/L CaCl2, were dis-pensed into the wells. The gel plate was incubated at37#{176}Cfor 20 hand then observed with an ultraviolet lighttransifluminator. We plotted the logarithm of the units ofDNase I dispensedagainst the diffusion radius (re - r0).

To determine DNase I activity in biological samples,we dissolved lyophilized tissue extracts and saliva sam-ples at 10-20 g/L in 10 mmol/L Tris-maleate, pH 6.5,containing 5 mmol/L CaCl2. Urine samples were dis-solved in the same buffer at 0.10-0.15 g/L, and semenand erythrocyte lysates were diluted 10-fold in the samebuffer. Sera, leukocyte lysates, and breast milk were notdiluted. We placed 2 pL of sample solution in the wells,and incubated the gel plates at 37#{176}Cfor 20 h. Theradius of each dark circle was then measured. Whensample activity was weak or not detectable, incubationswere continued for an additional 24 h.

Effects of EDTA and G-Actin on DNase Activity

We investigated the effect of EDTA on the SREDDNase I assay by adding 10 mmol/L EDTA to thereaction buffer instead of 20 mmoIJL MgCl2 and 2mmolfL CaCl2. We analyzed the effect of G-actin byadding G-actin and ATP (disodium salt, grade I, Sigma),an inhibitor of polymerization of the actin, to the moltengel to give final concentrations of 20 mg/L and 0.5mmol/L, respectively.

ConventionalDNase I Assays and Unit DefinitionforDNase I

We used two different assays. In the first, the precip-itation assay (5, 11), salmon testis DNA was digestedwith test samples of DNase I. Soluble DNA fragmentswere then measured at 260 mu after precipitation withethanol. We also used the hyperchromicity assay ofKunitz (18) as modified by Liao (19).

In this study, DNase I activity is given in Yasudaunits. One Yasuda unit of DNase I activity is theamount of enzyme that gives an increase in the absor-bance at 260 mu of 1.0 A after 15 mm in the precipita-tion assay (5). One Yasuda unit of purified human urineDNase I corresponded to -0.18 Kunitz unit as deter-mined in the hyperchromicity assay.

Assaysand Unit Definitionsfor Other Enzymes

The assay and unit definition for human DNase IIwere described previously (15). We determined phos-phodiesterases I and II according to the supplier’s in-structions.

DNase I Phenotyping

We determined the DNase I phenotype of serumsamples by polyacrylamide gel isoelectric focusing, fol-lowed by electrophoresis (3,20).

Protein Determination

We determined proteins by the Bio-Rad protein assay(Bio-Rad, Richmond, CA), using bovine serum albuminas a standard.

6.0-

3.O

E

E

‘9‘9C0U,)

a

160E

Co 3.0‘A

C I

-5 -4 -3 -2Log (DNase I activity, unit)

Fig.3.Standardplots for thedeterminationofhuman DNase I activitybytheSRED methodPurified human urine DNase I was used as the standard. The DNase I assayIs described in Materials and Methods. Values are means of six measure-ments. The plots were obtained after the gel plate was incubated at 37#{176}Cfor(#{149})3 h, (0) 8 h, (A) 20 h, and (U) 50 h

Fig.1.SRED gelplate,demonstratingthedarkcirclesproduced bydiffused ONase ISee Materials and Methods for assay details. 5.9, 12, 24, 47, and 94 x 10unit of purified urine DNase I in 2 ul. of 10 mmoVL Tris-maleate, pH 6.5,contaIning 5 mmol/L CaCI2, were applied to the wells from left to right;respectively. The gel plate was incubated at 37#{176}Cfor 20 hand then observedwith an ultraviolet light transilluminator (312 nm)

450 CLINICALCHEMISTRY, Vol.39,No.3,1993

ResultsDNase I Assay by the SRED Method

The SRED method is based on the fact that ethidiumbromide fluoresces only with unhydrolyzed DNA andnot with DNA digested by DNase (17, 21). A darkcircular zone, visible under ultraviolet light, is formedas DNase diffuses from the well into the agarose gelcontaining DNA and ethidium bromide (Figure 1). Theradius of this dark circle is proportional to the amount ofenzyme added to the well (Figure 2). Figure 2 indicatesthat the radius of the dark circle dependson the amountof active DNase I applied to the well, rather than on thesample volume or on the well size. The diffusion radius(r8 - r0), was linearly proportional to the logarithm ofDNase I activity within the range of (0.02-10) x i0unit after an 8-h incubation. Longer incubations of 20and 50 h increased assay sensitivity 4- and 10-fold,respectively (Figure 3). Coefficients of variation (CV, n= 6) at each time point in Figure 3 were within therange of 1.7-4.1% (mean, 2.5%). The lowest limit ofdetection was 2 x 106 unit with a 50-h incubation. Theupper limitof linearity was 1 x 10_2 unit, 2 x iOunit, and 5 x i0 unit with incubations of 3.-8, 20, and50 h, respectively. The within-run and between-runprecision studies were performed with 10 mniol/L Tris-maleate, pH 6.5, containing 5 mmoIJL CaCl2, 100 g/Lbovine serum albumin, and purified human urineDNase I at two selected concentrations (at the low endand high end of the assay). The within-run precisionsatDNase I concentrations of 3.4 and 3.4 x iO unitWL were4.3% (n = 12) and 2.6% (n = 12), respectively. Thebetween-run precisions at the same two concentrationswere 8.0% (n = 12) and 5.4% (n = 12), respectively. Wemeasured DNase I activity in sera from four differentindividuals by the SRED method before and after add-ing a known amount of purified urine DNase I (1.8 X10 and 1.8 x iO unit), and recovered 86-103%(mean, 94%) of the activity. We performed the samemeasurement using other body fluids and tissues (seeBiological Samples) and achieved similar analyticalrecovery. Bovine serum albumin at concentrations of�100 gIL did not influence the assay results.

Because two distinct types of DNase, DNase I andDNase II, occur in tissues and body fluids, we tested theeffect ofDNase 11 activity on results obtained with theSRED method. Under the present assay conditions,purified human urine DNase II caused no formation ofdark circles on the gel at concentrations �0.01 unitj

-5 -4Log (ONase I activity, unit)

Fig. 2. Relationshipbetweentheamount of DNase I appliedto thegelplateandthesizeofdark circleproducedbythediffusedenzymein the SRED methodSee MaterialsandMethodsfor assay details. Values are meansofquadrupli-cate measurements.The CVof each value was within the range of 1.3-4.1%.Logarithmof units of ONase I dispensedis plotted against the diffusion radius(r5 - i,), and a line of best fit is drawn (, = 0.990). (#{149}), 1.0mm, v5= 2.0

= 1.5 mm, v5 = 2.0 L;(A)#{231},= 1.5 mm, v, = 4,0 pL;(U)r0 1.5mm, v = 6.0 pi.

well. Calculated from the activity of DNase 11(4.6 X 10units/g protein), this corresponds to 2 ng/well. Theoptimum pH of DNase II is -5.0 with no activity evidentat pH 6.5 in the presenceof Ca2 and Mg2 (15,22,23).Hence, it is likely that the dark zone formation reflectedDNase I, not DNase II, activity in crude samples appliedto the gel. Phosphodiesterase I is present in humanurine (24), but neither authentic phosphodiesterase Ifrom snake venom nor phosphodiesteraseII from bovinespleen (0.4 and 0.9 jig/well, respectively) produced adark zone when applied to the wells. It has been re-ported that EDTA abolishes DNase I activity and thatG-actin is a potent and specific inhibitor of DNase 1(2,4,

2-3Total

3344222

101

Table 1. DNase I ActivIty In Human Serum Table 2. DNase I Activity In Human Tissues and BodyFluids

CLINICALCHEMISTRY, Vol. 39, No.3, 1993 451

DNa.. I phenotype Numberof samples Actlvltya

4.1 ± 2.04.5 ± 1.64.8 ± 1.7(3.9± 1.8)

44 ±

a Valuesare given as mean ± SDofall analyses in units/L. DNaseI activity

was assayed by the SRED method (see Materials and Methods).

b The value corresponded to (65 ± 27) x iO units/g protein.

5,25). We found that EDTA completely inhibited puri-fied human urine DNase I in the gel, and G-actininhibited the DNase I activity by 95%. These inhibitiontests confirm that the enzyme detected on the reactiongel plate was indeed DNase I.

DNase I Activity in Tissues and Body Fluids

The DNase I activity of 17 different tissue extractsand body fluids was analyzed by the SRED method. Thesamples that formed a dark circle on the gelplatewerethen subjected to inhibition tests with EDTA and G-ac-tin. The activity in all these samples was inhibited byEDTA and G-actin. To use an enzyme as a diagnosticmarker or to perform a metabolic study on an enzyme,one must know the normal activity of that enzyme inhuman serum. Table 1 shows the mean valuesofDNaseI activity in 101 serum samples obtained from differentdonors. Because serum DNase I is genetically polymor-phic (3), these samples were grouped into four mainphenotypes (types 1, 1-2, 2, and 2-3), and their activitywas then determined. The intra-individual coefficientsof variation in serum DNase I activity were -10% of themean value. Activities in serum samples from femaleand male subjects were 3.9 ± 1.3 (n = 20) and 4.6 ± 1.9unitsfL (n = 81), respectively. Significant phenotype- orsex-related differences in specific activity were notfound when the results were analyzed statistically bythe two-sample t-test. The level of activity of DNase I inother samples from different tissues and body fluids issummarized in Table 2. A wide range of DNase Iactivity was observed. Urine had the highest specificactivity of all samples tested.The DNase I activity insemen and saliva was also relatively high.

DIscussIon

Three types of nonradioactive DNase I assays havebeen reported: the hyperchromicity and the precipita-tion assays (see Materials and Method.s) and the fluoro-metric assay (10). Preliminary attempts to determineDNase I activity in tissue extracts and body fluids by thefirst two assays were unsuccessful. The sensitivityofboth assays was low and most test samples had to bedissolved at high concentrations. This procedure gaverise, occasionally, to chyle-like turbidity that interferedwith these spectrophotometric assays. Although thefluorometric assay is reported to be more sensitive thanthe others, it requires large amounts of pure covalentlyclosed circular DNA of bacteriophage PM2, which is not

ActMty, 10 x unlts/g proteinNo. of

lissue or body fluId samples RangV Msan ± SD

Pancreas 7 <0.05 [1],0.28-26.2 8.1± 8.6Kidney 7 <0.05(11,0.80-1.92 1.15 ± 0.62Submandibulargland 5 0.20-0.45 0.33 ± 0.09Liver 6 <0.05(1], 0.17-0.30 0.24 ± 0.15Spleen 7 <0.05 [2], 0.09-0.37 0.15 ± 0.13Placenta 7 <0.05 [41,0.22-5.1 0.80 ± 1.8Heart 7 <0.05(5], 0.48-3.7 0.60 ± 1.3Thymus 8 <0.05(6], 0.95-1.50 0.31 ± 0.55Brain 7 <0.05 [7] <0.05Lung 7 <0.05 [7] <0.05Leukocytest’ 5 <1.0 [5J <1.0Erythrocytes 11 <0.05(111 <0.05Urine 13 2900-10 000 6000 ± 2200Semen 9 16.6-34 25 ± 4.8Saliva 10 1.1-3.8 2.4± 0.94Breastmilk 3 0.10-0.23 0.16± 0.05

a The values In brackets are the number of samples whose activity wasbelowthe minimum limit of detection.The activitywas arbitrarily set atzero forthe calculationof mean ± SD.

bEach of the test Iysates (2 tL) contained >20 000 leukocytee. Becausethe number of leukocytes obtained from single donors was limited, moreconcentratedcell lysatescould not be tested.

commercially available. The SRED method is a simple,nonphotometric assay. The results described above in-dicate that the SRED method is -1000-fold more sensi-tive than the hyperchromicity and precipitation assays.It can be used to detect accurately a very small amountof enzyme present in concentrated crude samples. More-over, reagents needed for the SRED method are com-mercially available and inexpensive, the sample sizerequired is small, and the technical procedures involvedare simple.

To determine the effect of endogenousDNase I inhib-itors on the measurement of DNase I activity in biolog-ical samples, we prepared active DNase I by acid extrac-tion (0.125 mol/L H2S04) from tissues and body fluids.This procedure has been commonly used to avoid endog-enous inhibition (18). However, the DNase I activity ofthe samples thus obtained was no higher (in some casesit was even lower) than that of the tissue extracts andbody fluids prepared by the method given in the Mate-rials and Methods section. As described previously, wewere able to recover known amounts of human DNase Ifrom serum and other body fluids and tissues, quantita-tively. Therefore, the influence of endogenous DNase Iinhibitors was negligible under the present experimen-tal conditions.

Our data indicate that neither DNase I phenotype norsex influences serum DNase I activity. However, thenumber of serum samples in this study was small com-pared with other studies (13, 14), and other factors thatmay influence DNase I activity should be considered.Measurement of DNase I activity in many serum andother body fluid samples by the SEED method is underway in our laboratory. As shown in Table 2, DNase I

452 CLINICALCHEMISTRY, Vol. 39, No. 3, 1993

activity is high in urine. DNase I is thought to beproduced in the pancreas, kidney, liver, and other tissuesand subsequently released into the bloodstream. It maythen be concentrated in the kidney, giving rise to mark-edly higher concentrations of DNase I in urine than inother body fluids. This mechanism has been reported forother enzymes (26). We have proposedthat the humanDNase I present in various tissues and body fluids mightbe encodedby the same single gene, becausethe biochem-ical and genetic properties of this enzyme in urine,serum, pancreas, kidney, and liver are quite similar asdescribed in our previous papers (3-5, 27). Leukocytesamples prepared in this study showed no DNase Iactivity. We do not believe that this is caused by thepresence of heparin, which has been described as inhib-iting DNase activity (28), because the leukocytes ob-tained from heparinized blood were thoroughly washedwith isotonic Tris . HC1 buffer before the lysate prepara-tion.

This work was supported in part by the Uehara MemorialFoundation, the Research Foundation for Traffic and PreventiveMedicine, the Research Foundation for Cancer and CardiovascularDiseases, theShimabara Science Promotion Foundation, and aGrant-ia-Aid for Scientific Research from theMinistry of Educa-tion, Science, and Culture of Japan (04152052 to K.K., 04836007 toT.Y., and 04770356 to D.N.).

References1. Laskowski M. Deoxyribonuclease I. In: Boyer PD, ed. Theenzymes, 3rd ed., Vol.4.New York:Academic Press, 1971:289-311.2. Moore S. Pancreatic DNase. In: Boyer PD, ed.The enzymes, 3rded., Vol. 14. New York:Academic Press, 1981:281-96.3. Kishi K, Yasuda T, Ikehara Y, Sawazaki K, Sate W, lida R.Human serum deoxyribonuclease I (DNase I) polymorphism: pat-tern similarities among isozymes from serum, urine, kidney, liver,andpancreas. Am J Hum Genet 1990;47:121-6.4. Nadano D, Yasuda T, Kiahi K. Purification and characteriza-tion of genetically polymorphicdeoxyribonuclease I from humankidney. J Biochem(Tokyo) 1991;110:321-3.5. Yasuda T, Awazu S, Sate W, lida R, Tanaka Y, Kishi K. Humangenetically polymorphic deoxyribonuclease: purification, charac-terization, and multiplicity of urine deoxyribonuclease I. J Bio-diem (Tokyo) 1990;108:393-8.6. Rosenstreich DL, Tu JH, Kinkade PR, et al. A human urine-derivedinterleukin 1 inhibitor homologywith deoxyribonucleaseI. J Exp Med 1988;168:1767-79.7.Drew HR. Prayers AA. Structural junctions in DNA: theinfluence of flanking sequence on nuclease digestion specificities.Nucleic Acids Res 1985;13:4445-67.8. Zollner EJ, MUller WEG, Zahn RK, Brehm G, Korting GW.MOglichkeiten zur gezielten Diagnose durch Bestimmung der sau-ran Desoxyribonucleasen im Urin gesunder Kontrolipersonen undXeroderma pigmentosum-Patienten. Kiln Wochenschr 1971;49:1290-4.9. Shak S, Capon DJ, Helimiss R, Marsters SA, Baker CL.Recombinant human DNase I reduces the viscosity of cysticfibrosis sputum. Proc Natl Acad Sci USA 1990;87:9188.-92.

10. Love JD, Hewitt RR. The relationship between human serumand human pancreatic DNase I. J Biol Chem 1979;254:12588-94.11. Ito K, Minamiura N, Yamamoto T. Human urine DNase I:immunological identity with human pancreatic DNase I, andenzymic andproteochemical properties of the enzyme. J Biochem(Tokyo) 1984;95:1399-406.12. Taper JiB, Deckers CO, Deckers-Passau LO. Increase in nu-clease activity as a possible means for detecting tumor cell sensi-tivity to anticancer agents. Cancer 1981;47:523-9.13. Avall-Lundqvist E, Economidou-Karaoglou A, SjOvall K, et al.Serum alkaline DNase activity in normal or nonhospitalisedindividuals. Clin Chim Acta 1989;185:35-44.14. Avall-Lundqvist EH, Sj#{246}vallK, Eneroth PilE. Influences ofdiet and surgical trauma on serum alkaline DNase activity levels.Clin Chim Acta 1992;205:43-9.15. Yasuda T, Nadano D, Awazu S, Kishi K. Human urinedeoxyribonuclease II (DNase II) isoenzymes: a novelimmunoaffin-ity purification, biochemical multiplicity, genetic heterogeneityand broad distribution among tissues and body fluids. BiochimBiophya Acta 1992;1119:185-.93.16. Spudich JA, Watt S. The regulation of rabbit skeletal musclecontraction. I. Biochemical studies of the interaction of the tropo-myosin-troponin complexwith actin andtheproteolytic fragmentsof myosin. J BiolChem1971;246:4866.-.71.17. Yasuda T, Mizuta K, Ikehara Y, Kishi K. Genetic analysis ofhuman deoxyribonuclease I by immunoblotting andthezymogrammethod following isoelectric focusing. Anal Biochem 1989;183:84-8.18. Kunitz M. Crystalline desoxyribonucleaseI. Isolation andgeneral properties. Spectrophotometricmethod for the measure-ment of deaoxyribonuclease activity. J GenPhysiol 1950;33:349-.62.19. Liao T-H. Reversible inactivation of pancreatic deoxyribonu-clease A by sodium dodecyl sulfate. Removal of COOH-terminalresiduesfrom the denatured proteinby carboxypeptidase A. J BiolChem1975250:3831-6.20. Yasuda T, Nadano D, Tenjo E, Takeshita H, Kishi K. Thezymogram method for detection of ribonucleases after isoelectricfocusing analysis of multiple forms of human, bovine, andmicro-bial enzymes. Anal Biochem 1992206:172-7.

21. Kim HS, Liao T-H. Isoelectric focusing of multiple forms ofDNase in thin layers of polyacrylamide gel and detection ofenzymatic activity with a zymogram method following separation.AnalBiochem1982;119:96-101.22. Bernardi G. Spleen aciddeoxyribonuclease.In: Boyer PD, ed.The enzymes, 3rd ed., Vol. 4. New York: Academic Press, 1971:271-87.23. YsnnnAka M, Tsubota Y, Ansi M, et al. Purification andproperties ofaciddeoxyribonucleasesof human gastric mucosa andcervix uteri. J BiolChem1974249:3884-9.24. Ito K, Yamamoto T, Minamiura N. PhoephodiesteraseI inhuman urine: purification and characterization of the enzyme. JBiochem(Tokyo)1987;102:359-67.25. Lazarides E, Lindberg U. Actin is the naturally occurringinhibitor of deoxyribonuclease I. Proc Nati Acad Sci USA 1974;71:4742-6.26. Kishi K, Yasuda T, Mizuta K. New genetic markers detectedin human urine: DNase I, RNase 1, and 43-kDa glycoprotein. In:Ogita Z, Markert CL, eds.Isozymes: structure, function, andusein biologyand medicine. New York: Wiley-Lies, 1990:891-913.27. Kishi K, Yasuda T, Awazu S, Mizuta K. Genetic polymor-phism of human urine deoxyribonuclease I. Hum Genet 1989;81:295-7.28. Connolly JH, Herriott RM, Gupta S. Deoxyribonuclease inhuman blood and platelets. Br J Exp Pathol1962;43:392-401.