THE JOURNAL OF BIOLOGICAL CHEMISTRY

Vol. 243, No. 23, Issue of December 10, pp. 6123-6129, 1968
Printed in U.S.A.

Molecular Characteristics of Rat Liver Arginase*

(Received for publication, April 12, 1968)

HELGA HIRSCH-K LB AND DAVID &I. GREENBERG

From the Cancer Research Institute, University of California School of Medicine, San Francisco, California
9/,122

SUMMARY EXPERIMENTAL PROCEDURE

The molecular weight of approximately 1,500-fold purified Materials and Methods

arginase from rat liver (specific activity = 19,500 pmoles of
urea per min at 25 per mg of protein nitrogen) was deter- ar-Isonitrosopropiophenone was obtained from Eastman Kodak

Downloaded from http://www.jbc.org/ by guest on October 29, 2016

mined by the method of sedimentation equilibrium to be Company, Rochester, New York; n-arginine (free base) from
118,000. The sedimentation velocity constant is 6.1 S, the Sigma; tris(hydroxymethyl)aminomethane (A grade) from
diffusion coefficient 5.2 X 10U7cm2 per sec. Alteration in Calbiochem. Deuterium oxide, 99.9%, was purchased from
Bio-Rad Laboratories, Richmond, California, and carboxy-
pH, removal of Mn2+, and replacement of Mn2+ by Co2+did
not yield evidence of a changein molecular size. The partial methyl cellulose from Schleicher and Schuell, Keene, New Hamp-
specific volume (7) was determined by a differential sedi- shire.
mentation equilibrium technique in heavy water. Disso- Rat liver arginase was purified by a procedure reported by
ciation of the enzyme in 8 M urea yielded a protein which Schimke (3) which consists of certain steps used by Greenberg
sedimented in a single boundary and gave one protein band (5) for horse liver arginase and Grassmann, Hormann, and
Janowsky (6) for the isolation of calf liver arginase. The purifi-
after acrylamide gel electrophoresis. Sedimentation equi-
librium analysis of the dissociation compound gave a molecu- cation procedure was usually carried out with the livers of 5 to
lar weight of 30,800, which suggeststhat rat liver arginase is 10 Sprague-Dawley rats (approximately 400 g in weight).
composed of four subunits. Amino acid analysis has been The fresh livers were either immediately worked up or frozen
carried out and the values compared to those of arginases at -20.
from other species. In crude extracts the protein was measured by the biuret
method (7) with crystalline bovine serum albumin (Nutritional
Biochemicals Corporation) as standard. For more purified
arginase preparations, the absorbance was read in a Gilford
spectrophotometer at 280 and 320 rnp and the protein nitrogen
concentration was calculated according to the method of Bach
Differences in substrate inhibition, molecular weight, K,, and Killip (8). The nitrogen content of rat liver arginase was
and antigenicity of arginases from ureotelic and uricotelic determined by micro-Kjeldahl procedure to be 14.3%.
species have been found (1). The recently reported preparation Enzyme activity was assayed either by incubation of arginase
of highly purified rat liver arginase (n-arginine amidinohydro- in arginine solutions for 10 min at 25 and determination of the
lase, EC 3.5.3.1) (2, 3) is of interest because of its different urea with isonitrosopropiophenone (9-12) or by following the
electrochemical characteristics from that of purified horse (4, 5) enzymatic reaction in a Cary model 14 double-beam spectro-
and calf liver arginase (6). Whereas the latter are electronega- photometer at 2057 A (13). Because of the high absorbance of
tive in the physiological pH range, rat liver arginase is electro- protein in the far ultraviolet, this last method can only be used
positive. for purified preparations of arginase. By using quartz cells
Lack of information on the molecular characteristics of this with lo-mm, l-mm, and 0.5-mm path lengths the enzyme reac-
species of arginase, as well as the lack of information on the tion can be followed at substrate concentrations of lo- to
number of polypeptide chains per molecule of any species of 2 x 10m2 M. Fig. 1 shows a substrate dependence curve, apply-
arginase, prompted us to undertake a study of the molecular ing both methods. It should be mentioned that the spectro-
properties of rat liver arginase. photometric procedure is a very simple and rapid method, but
can only be used with difficulty at concentrations of substrate
* Aided by Research Grants CA-02915 and CA-03175 from the saturation.
National Cancer Institute, National Institutes of Health. Re- One enzyme unit is defined as the amount of enzyme that
port was presented at the meeting of the Federation of American
Societies for Experimental Biology, Atlantic City, New Jersey, produces 1 pmole of urea per min at 25 (9). Specific activity
April 15 to 20, 1968. is expressed in enzyme units per mg of protein nitrogen.

6123
 Rat Liver Arginuse Vol. 243, No. 23

at a rotor speedof 5784 rpm. Enzyme concentrationsranged

from 1.5 to 3.0 mg per ml. Initial protein concentrationswere
determined from the number of fringes obtained in a synthetic
boundary run as describedby Richards and Schachman(17).
The molecularweight wascalculated from the slopeof the loga-
rithm c versusrz.
The partial specific volume was determined by a differential
sedimentationequilibrium techniqueaccordingto the method of
Edelstein and Schachman(18). The method is basedupon the
change produced in the equilibrium concentration distribution
a 4-I when the density of the solution is increasedby the useof heavy
water. For the experimentsarginasepreparationsof 0.4, 0.8,
and 1.2 mg per ml in either water or 99% deuteriumoxide (each
0.01 M Tris-HCl buffered, pH 7.5) were dialyzed for several
days against solvent.
The amount of hexosesin the enzyme preparation wasassayed
with anthrone accordingto the method of Trevelyan and Harri-
(S) MOLES / LITER son (19) by usinggalactoseasstandard.
Amino acid hydrolyses were carried out with 6 s HCl as
FIG. 1. Concentration activity curve using two different described by Moore and Stein (20). Hydrolysis temperature

Downloaded from http://www.jbc.org/ by guest on October 29, 2016

methods for activity determination. Substrate solutions were
brought to pH 9.5 with dilute HCl. Temperature was 25. was 105. The amino acids were determinedwith a Beckman
C- - -0, spectrophotometric activity test (shaded area); V-- -v, amino acid analyzer, model 120.
color assay with isonitrosopropiophenone. For the study of the dissociationof rat liver arginase,enzyme
preparations at concentrations of 0.3 to 1.0 mg per ml were
Polyacrylamide gel electrophoresiswas carried out as de- dissolvedin and dialyzed against 8 M deionized urea (pH 7.3)
scribedby Davis (14) in a gel of 6.5% acrylamide concentration. at 4. The deionization was achieved by shaking the urea
Then, 10 to 25 pg of protein were applied to the gel. solution with a mixed bed resin of AG 501-X8, 20 to 50 mesh,
Analytical ultracentrifugation was performed with a Beck- analytical grade (Bio-Rad Laboratories). In order to prevent
man-Spinco model E ultracentrifuge, equipped with a cali- the oxidation of sulfhydryl groups, mercaptoethanolwas added
brated temperature control unit and Rayleigh interference and to the urea solution to give a final concentration of 10e4M.
schlieren optics. Prior to analytical centrifugation, enzyme
sampleswere dialyzed extensively against buffer solutions of RESULTS
appropriate ionic strength and pH. For the sedimentation
EnzymePurijication
velocity experiments,pictures were taken with schlierenoptics
and evaluated as describedby Schachman(15). Following the isolation of the enzyme according to the pro-
The diffusion coefficient was determined in the analytical cedure of Schimke (3), a further purification of the arginase
ultracentrifuge, using a capillary type double-sectorcell and could be obtained by a secondpassagethrough a CM-cellulose
a rotation speedof 12,590rpm. The rotor temperaturewaskept column (10 x 1 cm). Fig. 2 showsa typical elution pattern.
at 20. The enzyme samples(4 to 12 mg per ml) were dialyzed After the first passagethrough CM-cellulose(Fig. 2.-l), Fractions
carefully against buffer solution (0.01 M Tris-HCl, pH 7.5) to 12 to 15 (1 ml each) were pooled,lyophilized to a smallvolume,
prevent effects due to different ionic strength between the and dialyzed against0.01 M Tris-HCl buffer, pH 7.5. A second
protein solution and the overlayering buffer solution. Schlieren CM-cellulose column was equilibrated with this buffer, and the
pictures, taken at intervals of 8 min, were magnified with a enzyme from the first passagewasintroduced onto the column.
Nikon microcomparator. The gradient curve and base-line The enzyme then was eluted from the column with a 0.2 M
were traced by using half transparent paper. The area under arginine-0.005 M Tris-HCl solution, pH 7.5 (Fig. 2B). The
the gradient curve was measuredwith a compensatingpolar fractions which contained the enzyme activity (12 to 14) were
planimeter (Keufel and EsserCompany, New York). pooled, lyophilized, dialyzed against 0.01 M Tris-HCl buffer,
High speedsedimentationequilibrium analyseswere carried pH 7.5, and stored by freezing at -20. There was only slight
out according to the method of Yphantis (16) at a calculated lossof activity over a period of severalweeksof storage.
optimal rotation rate of 20,410 rpm by using a multichannel This arginase preparation gave a single protein band after
cell. Arginase samples were prepared at concentrations of acrylamide gel electrophoresisat pH 6.0 and 8.3 (Fig. 3A) and
0.4 to 1.5 mg per ml and dialyzed against solvent prior to each sedimentedin the analytical ultracentrifuge in a singleboundary
analysis. Pictures were taken with interference optics immedi- (Fig. 4A).
ately after reaching full speed, then after 36 and 48 hours. Rat liver arginasecould be chromatographedover Sephadex
Attainment of equilibrium was judged by following the blank- G-100 without lossof activity (0.01 M Tris-HCI buffer, pH 7.5).
corrected fringe displacementuntil it approached a constant The enzyme was eluted in a symmetrical protein peak with a
value. During the run the rotor temperature was kept at 20. constant specificactivity (19,500).
The interference patterns were magnified 50-fold with a Nikon The preparation of highest purity obtained by the above
microcomparator and the fringe displacementwas determined procedure had a specific activity of 19,500 after activation in
at intervals of 10 to 20 p. 0.05 M MnC& (pH 7.5) at 55 (21). The preparation of Schimke
Low speedsedimentationequilibrium runs were carried out (3) was estimated to have a specific activity of 14,800 under the
Issue of December 10, 1968 H. Hirsch-Kolb and D. M. Greenberg 6125

same conditions. This estimation was made by reducing the

data of Schimke (37) to 25 by means of a temperature coeffi-
cient of 2.5, calculated from the activation energy of the argi-
nase reaction. Under similar conditions the preparation of
Kossmann et al. (2) yielded a specific activity of 4,700.

Sedimentation Velocity Coeficient

In sedimentation experiments, carried out in a 12-mm single-
sector cell and a rotor speed of 50,740, the enzyme moved in
a single boundary as shown in Fig. 4. In order to determine
the concentration dependence of the sedimentation velocity
coefficient, experiments were performed at protein concentra-
tions of 2 to 12.5 mg per ml in 0.01 M Tris-HCI buffer, pH 7.5.
Corrections of the observed sedimentation coefficients, sobs,
to standard state (20, water) were made (15), and no change
could be observed in the sedimentation velocity coefficient
of the enzyme, when the experimentswere carried out at tem-
peraturesbetween10 and 25.
The s20,Wvalues, calculatedfor various protein concentrations,
are plotted in Fig. 5. For zero enzyme concentration, the sedi-

Downloaded from http://www.jbc.org/ by guest on October 29, 2016

mentation coefficient sio,, wasestimatedby extrapolation to be
6.1 S.
It wasof interest to seeif arginase,whoseactivity is reported
II
to be rather unstable below 7.0 and above 8.0 (5), would show
any dissociation or aggregation dependence on pH. By dialysis
FIN. 3. Acrylamide gel electrophoresis of rat liver arginase.
againstTris-HCl buffer solutions,enzyme sampleswere brought A showsthe migration band for native enzymeat. pH 8.3 after 1
to pH 5.8, 8.3, and 9.6. Table I gives the sedimentationcoeffi- hour of electrophoresis (left) and pH 6.0 after 12 hours of elec-
cientsobtained in this seriesof experiments. trophoresis (right) in Tris-alvcine buffer (14) at 6 ma ner nel.
No changein the sedimentationbehavior could be detectedat B gives the migration pattern 12 hours) of urka-dissociatedknzyome
in 4 M urea at pH 6.0.

FIG+. 4. Sedimentation velocity patterns for native enzyme

3t 001 M TRIS 02 M arcintne (A) at 50,740rpm in 0.01M Tris-HCI buffer, pH 7.5, at a protein
--
-I- concentration of 8.2 mg per ml, and for dissociated enzyme (B)
in 8 M urea at 59,780 rpm at a concentration of 6 mp ner ml. Pit-
B tures were taken after 1 hour and 2) hours after reaching fall
speed, respectively.

various pH values. The small differencesin the sedimentation

FIG. 2. Elution pattern of rat liver arginase from carboxy-
methyl cellulose columns (A and B, 10 X 1 cm). Arginine solu-
tions were prepared with 0.005 M Tris-HCl buffer. The second
peak contains thearginaseactivity, while in the first peak only
traces of activity were found. In A pooled fractions from 12 to
15 ml had a specific activity of 13,800, fract,ions from 12 to 14 ml in
B yielded a specific activity of 19,500.
coefficientsgiven in Table I are due to differencesin the protein
concentration.
In another experiment arginase was dialyzed extensively
against a 5 x 10e3M EDTA solution (pH 7.5) to remove the
manganesefrom the enzyme complex. A samplethus treated
was subsequentlydialyzed against a 5 x 10m3M Co(NO&-
solution (pH 7.5) for metal ion activation. As shown in Table
I removal of Mn* and subsequentreplacement of Mn* by
Co* did not changethe sedimentationbehavior of arginase.
6126 Rat Liver Arginase Vol. 243, No. 23

Extrapolation for zero concentration yielded D~o,~ = 5.2 x

lo- cm2 per sec.

Partial SpeciJic Volume (PI

For the determination of the partial specific volume, parallel
sedimentation equilibrium experiments were carried out. Pa-
rameters such as temperature, angular velocity, and protein
concentration were kept constant, but the density of the solvent
was varied by using either DzO or H20.
Fig. 6 shows a plot of log j uersus r* for arginase in Hz0 or
DzO. From the slopes, P was calculated by Equation 2

I >
0 5 JO 15
[cjin mg/ml where c is the concentration, r the distance from the center
FIQ. 5. Concentration dependence of the sedimentation of rotation, p the density of the solvent in milligramsper ml, and
coefficientof rat liver arginasein 0.01MTriz-HCl buffer, pH 7.5. k the measureof the exchangeability of hydrogen atomsof the
protein. Experiments to determine the weight increase by
TABLE I deuterium exchangehave been carried out with a number of
E#ect of pH, removal of Mn2+, and substitution of Co* on enzyme proteinsand gave a value of k = 1.0155(23, 24).

Downloaded from http://www.jbc.org/ by guest on October 29, 2016

activity and sedimentation behavior of rat liver arginase This method yielded a partial specificvolume of 0.75 ml per g,
-
Protein Act%@ which is in good agreementwith P calculated from the amino
Sample Metal ion PH comx!n- - acid compositionasdescribedby Cohn and Edsall (25), by which
tration
1>H 9.5 pH 6.5 a value of 0.74 ml per g wasobtained.
-- -- .-
m&d % % s
Determinationof Molecular Weight
I. Native enzyme Mn* 5.8 2.7 78 12 5.9
Mn+ 7.0 5.0 100 15 5.8 From Sedimentationand Diffusion Coegicients-Themolecular
Mn* 8.3 4.7 97 15 5.8 weight (AfJ was calculatedaccordingto the Svedbergequation
MnP+ 9.6 4.3 92 14 5.8 (26) and a value of 116,066wasobtained.

II. Enzymedislyzec 7.5 4.5 22 5 5.8

against5 X
lo- MEDTA

III. Sample II dia cot+ 7.5 3.2 28 32 5.9

lysed agains ; 3.0
5 x lo- II 0
Co(NOdn .-t
- - - - - E
0Activity testswere carried out spectrophotometricallyat pH
f 2.5
9.5 (optimum) and pH 6.5. The activity increaseof SampleIII
at pH 6.5 reflects the shift of the pH optimum curve of arginase s
whenCo* is usedasthe activating metal ion (5). E
z!
2
Diffusirm Coejici4mt 0 2.c
The diffusion coefficient of rat liver arginasewas determined 5
al
in the analytical ultracentrifuge for four different enzyme con-
F
centrations,ranging from 4 to 12 mg per ml. 'C
The apparent diffusion coefficients D,,,, were calculated ; 1.5
from the time dependenceof the diffusion procedureby Equation s
1 accordingto the methodof Gosting (22),
(AC)
D (1)
*PP= 4s. t(ac/ar)b.. 43.5 44.0 44.5
where ACis the concentration shift, t is the corrected time, and r2 [cm21
(~Yc/ar),,~ is the maximal concentration gradient. The concen- FIQ. 6. Sedimentation equilibrium analyzes of rat liver ar-
tration gradient (acm),., is proportional to the maximal height ginase. Two separateexperimentswere carried out in a multi-
of the gradient curve, (H,,,&, while the concentration shift AC channel cell. Operational speed waz 26,410rpm. The plots
is proportional to the area under the gradient curve (A). The represent the fringe displacementin the cell at equilibrium
correctedfor the blank for arginaeein water (A- - -A) and DzO
experimental time t was corrected by plotting (A/H,,,J2 uersus (C- - -0). Protein concentration wae 169cg in 0.1 ml. Each
time. solution contained1pmoleof Tris-HCl buffer, pH 7.5.
Issue of December 10, 1968 H. Hirsch-Kolb and D. M. Greenberg 6127

From Sedimentation Equilibrium--For low protein concentra-

tions, sedimentation equilibrium experiments were carried out at
a high rotor speed according to the method of Yphantis (16).
When the logarithm of the fringe displacement was blotted
against r*, the points fell on a straight line.
Experiments at higher enzyme concentrations (1.5 to 3.0 mg
per ml) and low rotor speed were carried out according to the
method of Richards and Schachman (17).
The molecular weight (M,) was calculated for each concen-
tration by Equation 3
2RT .~a In c (3)
Mw = (1 - V&a ar* I
P
1.0 2.0 3.0
where R is the gas constant, T the absolute temperature, c the
concentration, 7 the partial specific volume, o the angular
[cl in mg/ml
velocity in radians per set, and T the distance from the center
of rotation. The molecular weight obtained from these two Fm. 7. Concentration dependence of the molecular weight of
different sedimentation equilibrium techniques were plotted in rat liver arginase determined by sedimentation equilibrium
according to the methods of Yphantis (16) and Richards and
a l/M uersus c diagram and fell on a straight line. Extrapola-
Schachman (17). H, high speed sedimentation equilibrium

Downloaded from http://www.jbc.org/ by guest on October 29, 2016

tion to zero concentration (Fig. 7) yielded a molecular weight of experiment in HzO; A-A, same experiment in DzO (both 0.01
118,500. M Tris-HCl buffered, pH 7.5); O-O, low speed sedimentation
To obtain further evidence of the homogeneity of the enzyme equilibrium.
preparation M, was calculated from these experiments according
to the method of Van Holde and Baldwin (27). The weight and TABLE II
z-average molecular weights, extrapolated to zero concentration, Molecular weight of rat liver arginase and of its unit polypeptide
are summarized in Table II. The small differences of the determined by sedimentation velocity and sedimentation
equilibrium ultracentrifugation
molecular weights obtained by three different methods as well
as the close agreement of the calculated w- and z-averages Preparation Method of determination & ME
can be taken as good evidence for the homogeneity of the argi-
nase preparation. Native enzyme so and Do 116,000

Frictional Coegicient Sedimentation 118,500 120,000

equilibrium
From the data obtained from the diffusion experiments and
the sedimentation equilibrium, the frictional coefficient was Subunit (in 8 M urea) Sedimentation 30,800 31,000
determined as described by Tanford (28) and a value of 1.26 equilibrium
was obtained.

Determination of Hexose Content Amino Acid Composition

Greenberg, Bagot, and Roholt (4) found from the amino acid After hydrolysis of arginase samples as described by Moore
analysis of horse liver arginase that only 75% of the dry weight and Stein (20) no brown coloration of the samples could be
was represented by amino acids, whereas 107% of the nitrogen detected, which may be taken as a further indication that no
appeared in the form of amino nitrogen. It was concluded that considerable amount of sugar was present in the preparation.
this low recovery of amino acid residues per 100 g of protein Table III represents the amino acid composition of rat liver
could be at least partially attributable to amino acid degradation arginase. The values are averages of three different experiments.
during hydrolysis. The amino acid analysis by Grassmann et al. Tryptophan was calculated from the ultraviolet absorption
(6) of calf liver arginase yielded 102.2% of the nitrogen, but only spectrum according to the method of Edelhoch (29). The value
61.5% of the dry weight represented amino acid residues. An of serine was extrapolated from three different hydrolysis times
assay with anthrone gave a hexose value of 43.7% in the en- (12, 20, and 30 hours).
zyme preparation. From these results Grassmann et al. con-
cluded that arginase might be a glycoprotein. Dissociation into Subunits
In our experiments, samples of rat liver arginase (specific In 8 M urea, dissociated arginase showed no enzyme activity
activity = 19,500) were assayed with anthrone according to the with the use of the spectrophotometric test. In the analytical
method of Trevelyan and Harrison (19) and the absorption values ultracentrifuge the enzyme sedimented in a single boundary with
compared with a standard curve obtained with galactose. The a sedimentation coefficient of .s&,~ = 1.5 S (Fig. 4B).
assays yielded a hexose content of 1 to 3% in the arginase prep- Sedimentation equilibrium analysis of the dissociated protein
aration. This low percentage is also in good agreement with at 44,770 rpm yielded a straight line in the log fringe displace-
the results of the amino acid analysis. From the above data, ment versus r* diagram. In calculating the molecular weight,
however, it cannot be concluded whether the observed hexose we assumed the partial specific volume of the subunit to be the
content is a slight contaminant or whether a small amount of same as for the native enzyme; no information to the contrary
hexose is bound by the protein molecule. is available so far. The M, and M, molecular weights obtained
6128 Rat Liver Arginase Vol. 243, No. 23

TABLE III and replacement of Mn* by Co2+ did not yield evidence of a
Amino acid composition of rat liver arginase change in molecular size (Table I).
The report of Grassmann et al. (6) on the carbohydrate con-
No. of residues per
Amino acid mole of arginase tent of their arginase preparation, and the low values for amino
- acid residue recovery per unit weight of protein indicated that
Lysine. ...... 87.3 arginase contained a high content of bound carbohydrate.
Histidine ...... 23.1 The comparatively low nitrogen content of rat liver arginase
Arginine 32.5 (14.3%) does suggest the possibility of a non-nitrogen containing
Aspartic acid. 87.0 contamination. The nitrogen value obtained from the amino
Threonine ...... 68.1 acid composition is lower than the above; it does not include
,s erine.
. ...... 60.0 amide nitrogen. This tends to support the conclusion that the
Glutamic acid. ...... 87.0
low nitrogen content of the arginase is not a result of a non-
Proline. 71.1
nitrogen containing contamination. The analytical results with
Glycine. 90.0
Alanine. 64.2 anthrone show that rat liver arginase has very low or no bound
Half-cysteine. 7.2 carbohydrate. Also the value obtained for the partial specific
Valine. 85.8 volume (p = 0.75 ml per g) is characteristic of unconjugated
Methionine 12.8 proteins. It can be expected that a substantial amount of
Isoleucine ...... 51.0 hexoses (v = 0.5 to 0.6 ml per g) attached to the protein would
Leucine. ...... 71.1 substantially lower the value of v derived from sedimentation
Tyrosine. ...... 24.0 equilibrium. The fact that the partial specific volume v cal-

Downloaded from http://www.jbc.org/ by guest on October 29, 2016

Phenylalanine. ...... 31.0 culated from the amino acid composition (0.74 ml per g) is
Tryptophan. 12.2
in excellent agreement with the experimentally determined value
is further evidence for the absence of large amounts of hexose.
The results obtained with urea dissociated arginase lead to
from these experiments and extrapolated to zero concentration
the reasonable conclusion that the rat liver enzyme is com-
are given in Table II.
posed of four subunit polypeptide chains with a molecular
An acrylamide gel electrophoresis run of the dissociated
weight of about 30,000. The linearity of the slope of the loga-
protein at pH 6.0 in 8 M urea solution gave a single protein
rithm of fringe displacement (log f) uersus the square of the
band (Fig. 3B).
distance from the center of rotation (r2) indicated a compara-
DISCUSSION tively homogeneous dissociation compound. Acrylamide gel
electrophoresis gave a single protein band, suggesting that the
A number of criteria support the conclusion that rat liver
polypeptide subunits are probably identical. It should be
arginase prepared by us is a single molecular species of protein. emphasized, however, that small differences in the sequence of
These are the nature of the ultracentrifuge sedimentation peak, the individual subunits would not be detected by the methods
the appearance of only a single protein band on acrylamide gel
employed in this study. Studies on the identity of the subunits
electrophoresis, and the good agreement of the calculated W- by peptide mapping and determination of the terminal amino
and z-average molecular weights. It should be noted from the
acids are in progress.
recalculated values of the specific activities of the enzyme Further studies on the comparative molecular properties
preparations of Schimke (3, 21) and Kossmann et al. (2) as well of arginases from different genera of ureotelic vertebrates should
as from our findings that arginase activity is highly dependent
yield interesting information of evolutionary processes and the
on manganese activation. relation of structure and enzyme activity to changing physio-
The molecular weight M, of the enzyme, estimated by three logical requirements.
different methods, is close to 118,000. This value is lower than
the 138,000 to 142,000 reported by Schimke (21) based on sucrose REFERENCES
gradient experiments with crystalline alcohol dehydrogenase as 1. MORA, J., TARRAB, R., AND BOJALIL, L. F., Biochim. Biophys.
a standard, where no corrections for the partial specific volume Acta, 118, 206 (1966).
2. KOSSMANN, K. T., HINTZE, R., LANGE, K., AND MENNE, F.,
or the frictional ratio have been made.
Hoppe-Seylers Z. Physiol. Chem., 346, 163 (1966).
Mora, Tarrab, and Bojalil (1) reported that there is a remark- 3. SCHIMKE, R. T., J. Biol. Chem., 239, 3808 (1964).
able difference between arginases from ureotelic and uricotelic 4. GREENBERG, D. M., BAGOT, A. E., AND ROHOLT, 0. A., JR.,
species. For chicken liver arginase (30), lizard liver arginase Arch. Biochem. Biophys., 62. 446 (1956).
(30), and arginase from Neurospora crassa (1) molecular weights 5. GREENBERG, D. M., in P. D. BOYER, H. LARDY, AND K.
MYRB;~CK (Editors), The enzymes, Vol. 4, Ed. 2, Academic
of 276,000 and 278,000 were obtained. Greenberg found a
Press, New York, 1960, p. 257.
molecular weight of 138,000 (5) for horse liver arginase. Our 6. GRASSMANN, W., HORMANN, H., AND JANOWSKY, O., Hoppe-
calculated molecular weight for rat liver arginase therefore fits Seylers Z. Physiol. Chem., 312, 273 (1958).
very well into the group of the much lower molecular weights 7. LAYNE, E., in S. P. COLOWICK AND N. 0. KAPLAN (Editors),
Methods in enzumoloau, Vol. III, Academic Press, New
for arginases from ureotelic species.
York, 1957, p. 447.
Rat liver
arginase is a comparatively stable enzyme and 8. BACH, S. J., AND KILLIP, J. D., Biochim. Biophys. Acta, 29,
does not dissociate readily. Alteration in pH, removal of Mn* 273 (1958).
9. VAN SLYK~, D. D., AND ARCHIBBLD, R. M., J. Biol. Chem.,
1 R. Reddy and J. W. Campbell observed the occurrence of an 166. 293 (1946).
arginase of molecular weight about 27,606 in earthworm gut 10. HUNTER, A., AND DOWNS, C. E., J. Biol. Chem., 166, 173
(31). (1944).
Issue of December 10, 1968 H. Hirsch-Kolb and D . M. Greenberg 6129

11. ROBBINS, K. C., AND SHIELDS, J., Arch. Biochem., 62, 55 21. SCHIMKE, R. T., J. Biol. Chem., 237,459 (1962).
(1956). 22. GOSTING, L. J., Advance. Protein Chem., 11,429 (1956).
12. GREENBERG, D. M., Hoppe-Seyler/Thierfelder, Handbuch der 23. MARTIN, W. G., WINKLER, C. A., AND COOK, W. H., Can. J.
physiologisch- und pathologisch-chemischen Analyse, Vol. Chem., 37, 1662 (1959).
VI/C, Ed. 10, 1966, p. 354. 24. HVIDT, A., AND NIELSEN, S. O., Advance. Protein Chem., 21,
13. WARD, R. L., AND SRERE, P. A., Anal. Biochem., 18,102 (1967). 287 (19661.
14. DAVIS, G., Ann. N. Y. &ad S&i., 121,404 (I&). 25. COHN,E. j., AND EDSALL, J. T., Proteins, amino acids and
15. SCHACHMAN. H. K.. in S. P. COLOWICK AND N. 0. KAPLAN peptides, Rheinhold, New York. 1943. D. 370.
I *

(Editors): Metho& in enzymology, Vol. IV, Academic Press,

26. S;E~RG; T., AND PEDERSEN, K. O., The ultracentrifuge,
New York, 1956, p. 32. - --
Clarendon Press, 1940.
16. YPHANTIS. D. A.. Biochemistru. 3. 297 (1964).
27. VANHOLDE, K. E., AND BALDWIN, R. L., J. Phys. Chem., 62,
17. RICHARDS; E. G.; AND SCHACIX&A~, H. k., j. Phys. Chem., 63,
734 (1958).
1578 (1959).
18. EDELSTEIN, S. J., AND SCHACHMAN, H. K., J. Biol. Chem., 242, 28. TANFORD, C., Physical chemistry of macromolecules, John
Wilev and Sons. Inc.. New York. 1961. D. 356.
306, 1967.
19. TREVELYAN, W. E., AND HARRISON, J. S., Biochem. J., 60, 29. EDEL&H, H., Biochenkstry, 6,1948 (19f$).
298 (1952). 30. MORA, J., TARRAB, R., MARTUSCELLI, J., AND SOBER~N, G.,
20. MOOR& S.; AND STEIN, W. H., in S. P. COLOWICK AND N. 0. Biochem. J., 96, 588 (1965).
KAPLAN (Editors), Methods in enzymology, Vol. VI, 31. REDDY, R., AND CAMPBELL, J. W., Biochim. Biophys. Acta, 169,
Academic Press, New York, 1963, p. 819. 557 (1968).

Downloaded from http://www.jbc.org/ by guest on October 29, 2016

 Molecular Characteristics of Rat Liver Arginase
Helga Hirsch-Kolb and David M. Greenberg
J. Biol. Chem. 1968, 243:6123-6129.

Access the most updated version of this article at http://www.jbc.org/content/243/23/6123

Alerts:
When this article is cited
When a correction for this article is posted

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

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

Downloaded from http://www.jbc.org/ by guest on October 29, 2016

http://www.jbc.org/content/243/23/6123.full.html#ref-list-1