/. Biochem.

, 79, 1223-1234 (1976)

Purification and Characterization of a Proteinase from

Pineapple Fruit, Fruit Bromelain FA21

Fumiko YAMADA, Noriko TAKAHASHI,

and Takashi MURACHP
Department of Biochemistry, School of Medicine, Nagoya City
University, Mizuho-ku, Nagoya, Aichi 467

Received for publication, November 25, 1975

Fruit bromelain FA2, the main proteinase component of the juice of pineapple fruit,
has been purified and characterized.
1. Efficient extraction of this enzyme from the crude material was possible using
" Cellulosin AP," a microbial polysaccharidase preparation containing cellulase, hemi-
cellulase, and pectinase. The enzyme was purified mainly by successive applications
of anion-exchange chromatography, yielding an apparently homogeneous protein as
judged by several physical, chemical, and immunochemical criteria. Properties of
FA2 include: molecular weight, 31,000; isoelectric point, pH 4.6; absorbance at
280 nm of a 1% solution at pH 7.0 per cm, 19.2.
2. FA2 gave only alanine phenylthiohydantoin upon amino-terminal group analysis
by the Edman procedure. Stepwise degradation yielded the amino-terminal sequence
Ala-Val-Pro-Gln-Ser-Ile-Asp-Trp-Arg-Asp-Tyr-Gly-Ala. The amino acid composition
of FA2 was not markedly different from that of stem bromelain, except for a much
smaller lysine content and a smaller alanine content relative to glycine in FA2.
FA2 contained neither amino sugars nor neutral carbohydrates as determined by
several methods, so FA2 is not a glycoprotein.
3. By labeling the reactive cysteine residue (CYS) with [uC]iodoacetate, the following
partial amino acid sequence has been determined.

Asn-Glx-Asn-Pro-Cys-Gly-Ala-CYS

1
This work was supported in part by a grant from the Ministry of Education, Science and Culture, of
Japan. A part of this work is taken from a dissertation submitted by F. Yamada to the Graduate School
of Nagoya City University in partial fulfillment of the requirements for the degree of Doctor of Medical
Science, Feb. 1974.
1
Present address: T. Murachi, Department of Clinical Science, Kyoto University Hospital, Kyoto Uni-
versity Faculty of Medicine, Sakyo-ku, Kyoto, Kyoto 606.
Abbreviations: BAEE, a-N-benzoyl-L-arginine ethyl ester; BAA, ar-N-benzoyl-L-arginine amide; Tris,
tris(hydroxymethyl)aminoniethane.
Enzymes: bromelain (fruit), EC 3.4.22.5; bromelain (stem), EC 3.4.22.4; papain, EC 3.4.22.2; ficin,
EC 3.4.22.3; trypsin, EC 3.4.21.4.

Vol. 79, No. 6, 1976 1223

1224
4. FA2 showed pH optima of 8.0 and 8.3 toward hemoglobin and casein, respectively.
The Michaelis constant for a-N-benzoyl-L-arginine ethyl ester was 43 mM at pH 6.0
and 25°. FA2 cleaved bradykinin between Gly*-Phes and between Phe5-Ser* at
comparable rates; it hydrolyzed angiotensin II preferentially between TyrMle 6 .
5. Similarities and differences found thus far between fruit bromelain FA2, stem
bromelain SB1, and papain are tabulated.
Bromelain is the generic name of the proteo-
lytic enzymes contained in pineapple plant
tissues; those contained in its fruit are called
fruit bromelain, and those contained in its
stem tissue, stem bromelain. More detailed
descriptions of purification and characterization
are available for stem bromelain (1,2) than
for fruit bromelain (3, 4), although the latter
enzyme was discovered much earlier (5). Ota
et al. (3, 4) reported purified fruit bromelain
was an acidic protein with alanine as the major
amino-terminal residue, while valine is the
amino-terminal group of stem bromelain (3).
Preliminary experiments in this laboratory con-
firmed the results of Ota et al., but indicated,
at the same time, that our preparation of fruit
bromelain did not contain any carbohydrates,
unlike the preparations obtained by Ota et al.
(3, 4), which always contained firmly bound
carbohydrates. This apparent discrepancy
seemed to require further investigation, partic-
ularly in view of the fact that the major
components of stem bromelains, called SB1 and
SB2, have been established to be basic glyco-
proteins, containing 2 per cent carbohydrates
(2). A discrepancy also existed regarding the
molecular weight: Ota et al. reported a value
of 18,000 for their last preparation, while our
preparation was found to migrate more slowly
upon gel electrophoresis in the presence of so-
dium dodecyl sulfate than stem bromelain,
which has a molecular weight of 28,000 (2).
It was thus thought necessary to obtain a
highly purified preparation of the major com-
ponent of fruit bromelains, in order to resolve
these discrepancies and permit a more detailed
comparison between fruit and stem bromelains
and other thiol proteinases of plant origin, in-
cluding papain and ficin. Such a comparison
is of great interest in relation to the phylogeny
and ontogeny of these thiol proteinases.
F. YAMADA, N. TAKAHASHI, and T. MURACHI
MATERIALS AND METHODS
Materials—Crude fruit bromelain from the
Dole Company, Honolulu, Hawaii, was used.'
It is a light-brown dry powder precipitated
with acetone from the juice of pineapple fruits.
"Cellulosin AP" was a product of Ueda Chem-
ical Industries Co., Osaka. The preparation
contains cellulase, hemicellulase, and pectinase.
The commercial preparation was found to be
contaminated with a very slight proteolytic ac-
tivity which was significantly suppressed in the
presence of 0.5 mM each of ethylenediamine-
tetraacetate and sodium tetrathionate (Na2S4O8)
(6). Trypsin (twice recrystallized) was a prod-
uct of Worthington Biochemical Corporation.
Hammarsten casein was a product of E.
Merck, Darmstadt. a-N-Benzoyl-L-arginine
ethyl ester (BAEE), bradykinin and angiotensin
II were obtained from the Protein Research
Foundation, Osaka. The molecular weight ref-
erence proteins, bovine serum albumin, oval-
bumin, bovine chymotrypsinogen A, papain,
horse heart cytochrome c, and porcine pepsin
[EC 3.4.23.1], were obtained from Mann Re-
search Laboratories. Sodium tetrathionate was
from K & K Laboratories. [^Jlodoacetate,
sodium salt, was purchased from New England
Nuclear Corp., and used for labeling the SH
group of the enzyme. It had a specific radio-
activity of 13.9 mCi/mmole.
DEAE - cellulose (10 meq/g) was from
Whatman. ECTEOLA-cellulose (0.45 meq/g)
was obtained from Seikagaku Kogyo Co.
Dowex-lx2 (Cl- form, 200—400 mesh) was pur-
chased from Dow Chemical Co., Amberlite IR-
120 (H+ form) from Rohm and Haas Co., Se-
1
We are indebted to Dr. D.R.V. Golding for a
generous supply of this material.
/ . Biochem.
FRUIT BROMELAIN FA2
phadex G-25 and G-75 from Pharmacia, Sweden,
and Bio-Gel P-4 from Bio-Rad Laboratories.
The plates for thin-layer chromatography,
Kieselgel Fi54, were products of E. Merck,
Darmstadt.
Measurement of Enzymatic Activity—Pro-
teinase activity toward casein was determined
as described by Murachi (1). One activity
unit is defined as the amount of enzyme which
gives an increase in absorbance at 275 nm of
1.00 after incubation for 10 min at pH 7.2 and
30°.
Proteinase activity in the acidic pH range
was determined by the method of Anson ( 7 )
using denatured hemoglobin as a substrate.
Esterase activity toward BAEE was deter-
mined as described by Murachi (1). Hydrol-
ysis of the ester substrate was followed in a
Radiometer model SIU2/ABU11/TTT1 pH-stat
at pH 6.0 and 25°.
Biological Assay—Bradykinin and angio-
tensin II were determined by measuring con-
traction of the uterus of a rat previously in-
jected intraperitoneally with 0.2 mg of estradiol
(8). The uterus specimen was incubated in
10 ml of oxygenated de Jalon solution at 30°.
Physical Characterization — Sedimentation
analysis was performed with a Hitachi UCA-1
ultracentrifuge equipped with a schlieren cy-
lindrical lens system. A 0.43% solution in 0.02
M sodium phosphate buffer, pH 6.8, containing
0.1 M NaCl was used. Measurement of molec-
ular weight by gel filtration was performed
according to the method of Andrews (9). The
sample protein and reference proteins were
filtered through a column (1.0x110 cm), elut-
ing with 0.02 M potassium phosphate buffer,
pH 7.2, at room temperature. Because the gel
nitration was carried out without inactivating
the proteases, there was a possibility of auto-
digestion. In order, therefore, to assure the
integrity of the eluted materials, the elution
positions of the proteases were located by
measuring their specific caseinolytic activities.
The other marker proteins were located by
determining the absorbance at 280 nm.
Electrophoresis on polyacrylamide gels was
performed by the methods of Ornstein (10)
and Davis (11). The run was carried out at
pH 9.4 for 1 hr. Proteins were stained with
Vol. 79, No. 6, 1976
1225
Amido Black 4B or Coomassie Brilliant Blue.
Glycoproteins were stained according to
Zacharius and Zell (12).
Gel electrophoresis in the presence of so-
dium dodecyl sulfate was carried out in order
to determine the molecular weights of proteins,
according to the procedure originally described
by Weber and Osborn (13) and Dunker and
Rueckert (14) and modified by Takahashi et
al. (2).
The isoelectric point was determined with
an isoelectric focusing apparatus. The column
volume was 110 ml, 1% carrier ampholine was
used in the pH range of 3—10, and the run
was carried out at 4° and 570 volts for 60 hr.
Analysis of Protein—Protein concentration
was determined by the method of Lowry (15),
and also by measuring the absorbance at 280
nm. For amino acid analysis, approximately
0.3 /imole of the sample was hydrolyzed with
5.7 N HC1 in an evacuated, sealed tube at 110°
for 20 or 70 hr. The hydrolysate was analyzed
by the method of Spackman et al. (16) using
a Hitachi KLA-3B amino acid analyzer.
Determination of tryptophan and tyrosine
residues was done by measuring the absorb-
ance values at 280 nm and 288 nm in 6 M
guanidine hydrochloride, according to the
method of Edelhoch (17). The content of
tyrosine was also calculated from the incre-
ment in absorbance values at 295 nm and 300
nm at pH 12 over those at pH 7 (17), taking
^2% nm=2,480 and ^300 nm = 2,270. The meth-
od of Goodwin and Morton (18) was also em-
ployed ; measurements were made at 280 nm
and 295 nm in 0.1 N NaOH, taking ^£280nm =
1,580 and ^£295nm = 2,360 for tyrosine, and
^£2so nm = 5,250 and 4i295nm = 2,280 for trypto-
phan.
Analysis of cysteine and cystine contents
was performed by the method of Inglis and
Liu (19). The relative color yield of S-sulfo-
cysteine upon reaction with ninhydrin was
0.75, the color value for aspartic acid being
taken as unity.
The content of neutral sugar was deter-
mined by the orcinol-HiS04 method (20), us-
ing mannose as a standard.
Glucosamine was determined either by a
modification of the Elson-Morgan reaction
1226
{21), or by amino acid analysis.
Amino acid sequence analysis was mostly
performed by the Edman degradation proce-
dure (22). The final identification of the
phenylthiohydantoins was done by thin-layer
chromatography according to the method of
Edman and Begg (23), using the solvent sys-
tems described by Jeppsson and Sjoquist (24).
High-voltage paper electrophoresis and
paper chromatography were employed for the
separation and purification of peptides. Toyo
No. 51 paper was used. High-voltage paper
electrophoresis was carried out at pH 3.5 or
6.5 at 2,000 volts for 1 hr. Descending paper
chromatography was run with 1-butanol: pyri-
dine : acetic acid: water (15 : 10 : 3 : 12, v/v) at
room temperature for 2 days. The peptides
and amino acids were located by staining with
ninhydrin, Pauly, and Sakaguchi reagents.
Other Methods—Radioactivity was meas-
ured on a Packard Tri-Carb model 3380 liquid
scintillation spectrometer. The sample solu-
tion (10 or 50 n\) containing uC-labeled com-
pounds was mixed with 10 ml of Bray's scintil-
lator solution, and counted. For radioactive
samples on paper after paper chromatography
and paper electrophoresis, each segment of the
paper strip was placed in 10 ml of scintillator
solution, and counted directly.
Rabbit antisera to crude fruit bromelain
were prepared according to the method of
Sasaki etal.(25). Immunoelectrophoresis was
carried out as described by Iida et at. (26).
RESULTS
Purification of Fruit Bromelain—All pro-
cedures were carried out at 4°. To prevent
autodigestion during purification, all buffer
solutions used contained 0.1 mM sodium tetra-
thionate.
Step 1. Extraction: Acetone powder of
the juice of pineapple fruit (10 g) was sus-
pended in 100 ml of cold 0.02 M potassium
phosphate buffer (pH 7.2) containing 100 mg
of sodium tetrathionate, and the mixture was
allowed to stand in a cold chamber for 1 hr.
This crude extract was so viscous that it could
not be clarified by conventional centrifugation
or nitration.
F. YAMADA, N. TAKAHASHI, and T. MURACHI
Step 2. Treatment with Cellulosin AP:
Three ml of 1% Cellulosin AP solution con-*
taining 0.5 mM ethylenediaminetetraacetate
and 0.5 mM sodium tetrathionate was added
to the fruit powder suspension obtained in step
1. This mixture, when stirred at 0° for 30
min, transformed itself into a solution of very
low viscosity. This treatment greatly facili-
tated extraction of the enzyme from the thick,
10% suspension. The solution was centrifuged
at 6,000 rpm for 10 min.
Step 3. Ion-exchange chromatography with
DEAE-cellulose: The centrifugal supernatant
from step 2 was applied to a 7.6x7.0 cm col-
umn of DEAE-cellulose previously equilibrated
with 0.02 M potassium phosphate buffer (pH
7.2). The column was washed with the same
buffer and then with 0.4 M potassium phos-
phate buffer (pH 7.2) (Fig. la). The eluted
fractions which showed caseinolytic activity
I 2 0 I
EFFLUENT VOLUME ( l i t e r )
Fig. 1. Purification of fruit bromelain. (a) DEAE-
cellulose chromatography of fruit bromelain (step 3
in Table I). The material, extracts (100 ml) from
acetone powder (10 g), was applied to a 7.6x7.0 cm
column of DEAE-cellulose equilibrated with 0.02 M
potassium phosphate buffer, pH 7.2. After the
column had been washed with the same buffer,
elution was carried out with 0.4 M potassium phos-
phate buffer, pH 7.2 as indicated by the vertical
arrow, (b) ECTEOLA-cellulose chromatography of
fruit bromelain (step 4 in Table I). The material,
0.78 g in 40 ml, was applied to a 2.7 x 17.5 cm column
of ECTEOLA-cellulose. After the column had been
washed with 0.02 M potassium phosphate buffer, pH
7.2, stepwise elution was carried out with 0.14 M
potassium phosphate buffer. • , protein concentra-
tion (absorbance at 280 nm) on the left-hand
ordinates; O, enzyme activity (units per 50/*1 of
each fraction) on the right-hand ordinates. The
fractions pooled are indicated by horizontal bars.
J. Biochem.
FRUIT BROMELAIN FA2
were combined and ammonium sulfate was
added to make 0.4 saturation. The precipitate
thus produced was collected, dissolved in 35
ml of water, and dialyzed overnight against
0.02 M potassium phosphate buffer.
Step 4. Ion-exchange chromatography with
ECTEOLA-cellulose: The dialyzed solution
was applied to a 2.7x17.5 cm column of
ECTEOLA - cellulose previously equilibrated
with 0.02 M potassium phosphate buffer (pH
7.2). Elution was carried out with 0.14 M potas-
sium phosphate buffer (pH 7.2). The elution
pattern is shown in Fig. lb. The fractions
with caseinolytic activity were collected, and
the enzyme protein was precipitated by 0.4
saturation with ammonium sulfate.
Step 5. Rechromatography with
ECTEOLA-cellulose: The enzyme was further
purified by rechromatography with ECTEOLA-
cellulose in the same way.
The specimen thus obtained was named
fruit bromelain FA2.4 The results of purifica-
tion are summarized in Table I. From 10 g
of acetone powder of the juice of pineapple
fruit, 260 to 300 mg of FA2 was obtained,
with a roughly 2.5-fold increase in specific ac-
tivity and 23 to 27% recovery of the total ac-
tivity.
Purity of FA2—The ultracentrifugation
of FA2 solution produced a single, symmetri-
4
FA stands for "fruit, acidic," conforming to the
nomenclature proposed previously (2). FA1 is a
minor proteinase component present in some of the
preparations of fruit juice acetone powder. It was
found to emerge from the DEAE-cellulose column
ahead of FA2 under the conditions employed (6).
TABLE I. Summary of the purification procedure for fruit bromelain FA2.
Step Volume Total protein
(ml)
1. Extraction 1 100
2. Treatment with Cellulosin AP 100
3. DEAE-Cellulose 40"
4. ECTEOLA-Cellulose 35"
5. ECTEOLA-Cellulose 28°
» Acetone powder (10 g) of pineapple fruit juice was suspended in 100 ml of 0.02 M potassium phosphate
buffer, pH 7.2. ° After (NH4)tSO4 precipitation and re-dissolution.
Vol. 79, No. 6, 1976
1227
cal schlieren pattern. Electrophoresis in poly-
acrylamide gel gave a single band. End group
analysis also gave a single spot of alanine
phenylthiohydantoin. In immunoelectrophore-
sis, only one precipitation arc against anti-
crude fruit bromelain antibody was obtained
(Fig. 2). These results indicate that the puri-
fied FA2 was homogeneous.
Physical Properties — The physical con-
stants determined are shown in Table V (see
below). Figure 3 shows the results of gel
electrophoresis in the presence of sodium
dodecyl sulfate. A value of molecular weight
31,000 was obtained.
Chemical Properties—(1) Amino acid com-
position : The amino acid composition of FA2
is shown in Table II. The values are the
means of four to five hydrolysis runs for 20
hr and for 70 hr.
The contents of tyrosine and tryptophan
were measured by several methods, and the
results are presented in Table III. There are
some differences depending upon the method
(b)
Fig. 2. Immunoelectrophoresis of crude fruit bro-
melain (a) and purified fruit bromelain FA2 (b).
Electrophoresis was carried out at 6-8 volts per cm
for 2 hr in 1.2% pure agar, 0.03 M barbital buffer,
pH 8.6. The plate was incubated for 16 hr at 37°
after application of the antisera to the central
trough.
Specific activity Total activity
(g) (units/mg protein) (units X10"')
2.8 4.5 128
2.8 4.5 138
0.78 10.3 W
0.47 10.6 50
0.26 11.6 30
1228 F. YAMADA, N. TAKAHASHI, and T. MURACHI

TABLE II. Amino acid composition of fruit bromelain

10'
FA2 and stem bromelain SB1.»
3terum albumin

ovalbumii
Number of residues per molecule
pepsin1
FA2
tu Amino acid
chymotrypsinoQfn SB1°
pa pain Nearest
Average integer
ID
myoglobin
Aspartic acid 32.3 32 23.3
o cytochroms Threonine 13.7 14 10.1
Serine 30.0 30 21.3
Glutamic acid 25.4 25 18.6
0.5 1.0 Proline 10.0 10 11.3
RELATIVE MOBILITY
Glycine 35.3 35 27.3
Fig. 3. Estimation of the molecular weight of
purified fruit bromelain FA2 using 10% polyacryl- Alanine 25.3 25 29.9
amide gel containing 0.1% sodium dodecyl sulfate. Valine 19.7 20 15.9
The seven marker proteins were: bovine serum Half-cystine 8.9 9 9.0
albumin, mol. wt. 68,000; ovalbumin, mol. wt. Methionine 5.0 5 3.3
43,000; porcine pepsin, mol. wt. 35,000; bovine Isoleucine 17.5 18 16.8
chymotrypsinogen A, mol. wt. 25,700; papain, mol. Leucine 10.0 10 6.9
wt. 23,400 ; sperm whale myoglobin, mol. wt. 17,600;
Tyrosine 16. 0c 16 14.3
and horse heart cytochrome c, mol. wt. 12,400. The
arrow indicates the mobility of FA2. Phenylalanine 8.4 8 8.2
Lysine 8.3 8 14.2
Histidine 0.9 1 1.2
of determination; the average values are Arginine 9.2 9 8.1
shown in Table II, which also contains data Tryptophan 6.8= 7 6.8
for stem bromelain SB1. Taking into con- Glucosamine 0.0 0 1.3
sideration the difference in molecular size be-
tween FA2 and SB1, it is apparent that the a
Methods of analysis and calculation are described
two enzyme proteins do not differ markedly in the text. " Taken from Takahashi et al. (2).
c
in amino acid composition, except for the fol- Average value (see Table III).
lowing two points. (1) FA2 has markedly less
lysine than SB1, a basic protein, while the TABLE III. Contents of tryptophan and tyrosine in
contents of arginine and histidine are com- fruit bromelain FA2.
parable. (2) FA2 contains less alanine than
glycine, while SB1 contains more alanine than Method Tyrosine Tryptophan
glycine. Amino acid analysis 19.3
( 2 ) Contents of carbohydrates: Qualita-
tive and quantitative analyses of carbohydrates Edelhoch (19) 280 nm« 15.4
7.4
were carried out by four different methods. 288 nm» 17.6
(a) Periodate-Schiff staining was per- 295 nm" 14.4
7.0
formed after gel electrophoresis. Stem brome- 300 nm" 14.4
lain, which is known to contain 2% neutral
sugars and 2 residues of N-acetylglucosamine Goodwin and Morton (20) 16.0 6.0
per molecule (2), was used as a reference.
Average 16.0 6.8
The stem enzyme gave a distinct reddish-purple
band, while FA2 gave a turbid band of pro- * In the presence of 6 M guanidine hydrochloride,
tein when treated with 12.5% trichloroacetic calculated from the absorbance at 280 and 288 nm.
acid. b
Calculated from the increment in absorbance values
(b) Quantitative analysis for neutral sug- at 295 and 300 nm at pH 12 over those at pH 7.

/ . Biochem.
FRUIT BROMELAIN FA2
ars by the orcinol-HiSO« method (20) was re-
peated six times, but the data showed that
FA2 contained no more than 0.05 mannose-
equivalent per molecule.
(c) Six repetitions of quantitative anal-
ysis for amino sugars by the Elson-Morgan
method (21) with glucosamine as a standard
showed that FA2 contains no appreciable
amount of these sugars.
(d) The amino acid analysis chart did
not show any detectable peak corresponding
to hexosamines.
It was therefore concluded that FA2 is not
a glycoprotein.
( 3 ) Amino-terminal analysis: Following
Edman's procedure, FA2 gave only the phenyl-
thiohydantoin derivative of alanine on thin-
layer chromatograms using two different sol-
vent systems. When Edman's degradation
procedure was applied to freshly prepared FA2
(approximately 10 mg), the following sequence
was obtained: Ala-Val-Pro-Gln-Ser-Ile-Asp-
Trp-Arg-Asp-Tyr-Gly-Ala.
Active Center— [uC]Carboxymethyl labeling
of the reactive SH group of FA2 was carried
out according to the method described by
Takahashi et al. (2) for stem bromelain.
After performic acid oxidation, the product
(390 mg) was digested in 40 ml of incubation
mixture with 3.9 mg of L-l-tosylamido-2-
phenylethyl chloromethyl ketone-treated tryp-
sin at pH 8.0 and 37° for 24 hr, in the pres-
ence of 0.01 M calcium chloride. The digest,
after centrifugation, contained 68.7% of the
initial radioactivity. Radioactive peptides in
the digest were fractionated successively on
Sephadex G-25, Dowex-lx2 resin, and Bio-Gel
P-4, as shown in Fig. 4. From peak A-2
(Fig. 4c), a ninhydrin-positive, radioactive pep-
tide was recovered after high-voltage paper
electrophoresis and paper chromatography with
a yield of 5% of the parent protein in terms
of radioactivity. The material from peak A-l
resisted further purification by these methods.
The purified [14C]carboxymethyl - peptide
gave asparagine phenylthiohydantoin alone
upon amino-terminal analysis. The amino acid
composition (molar ratio) of this peptide was:
cysteic acid (including oxidized product of
carboxymethylcysteine), 1.54; aspartic acid,
Vol. 79, No. 6, 1976
1229
10 20 30
FRACTION NUMBER
Fig. 4. Preparation of radioactive peptides from
"C-labeled fruit bromelain FA2. (a) Gel filtration
of a trypsin digest of [14C]carboxymethylated FA2 on
Sephadex G-25. [14C]carboxymethylated and per-
formic acid-oxidized FA2 (12 /imoles) with a total
radioactivity count of 1.18xl07cpm was digested
with trypsin, and the digest was applied to a 2.5 X
70 cm column of Sephadex G-25. Elution was car-
ried out with 0.05 M sodium acetate, pH 5.2. (b)
Ion-exchange chromatography of the radioactive
peptide fraction obtained from the trypsin digest of
[14C]carboxymethylated FA2. The pooled fraction
(see (a)) was applied to a 1.4x70 cm column of
Dowex-lx2 (Cr form). Gradient elution was started
at the position indicated by the arrow (1) using an
autograde containing 300 ml each of water, 0.005 N
HC1, 0.05 N HC1, and 0.1 N HC1. From the position
of the arrow (2), elution with 0.1 N hydrochloric
acid was carried out. (c) Gel nitration of peak A
material from (b) above on Bio-Gel P-4 (1.0x110 cm).
Elution was carried out with 0.2 N acetic acid. Size
of each fraction: 12.5 ml in (a), 50 ml in (b), and
10 ml in (c). The horizontal bar in each figure
indicates effluent fractions that were pooled for
further fractionation or analysis.
1.40; glutamic acid, 1.00; proline, 0.66; gly-
cine, 1.07; and alanine, 0.85. Edman degra-
dation gave the sequence:
Asn-Glx-Asn-Pro-Cys-Gly-Ala-CYS
where CYS denotes the reactive cysteine res-
idue.
Enzymatic Activity—(1) Optimum pH:
pH profiles of the proteinase activities of FA2
1230
Fig. 5. The pH dependence of the proteinase ac-
tivity of purified fruit bromelain FA2 ( • ) and the
crude enzyme (O). Denatured hemoglobin (a) and
casein (b) were used as substrates. In (a), the pH
was adjusted with 1 N NaOH or 2 N HC1. In (b),
glycine-NaOH buffer was used for pH 7-10. Each
incubation tube contained 50 eg of FA2 or 45 MS of
the crude enzyme. Other experimental details are
described in the text.
are shown in Fig. 5. The optimum pH for
FA2 is 8.0 toward hemoglobin and 8.3 toward
casein. The starting material, acetone powder
of pineapple fruit juice, showed a broad pH
profile toward hemoglobin with an optimum
around pH 5. This implies that the crude
preparation contained, in addition to FA2,
another protease or proteases with optimum
pH on the acidic side.
(2) Specific activity and Km value: These
values for FA2 toward casein and BAEE as
substrates were determined and compared with
those of stem bromelain (SB1) and papain.
The results are shown in Table V. The spe-
cific activity of FA2 toward casein is slightly
higher than that of SB1. The Km value of
FA2 for BAEE is 43 mM, which is much
smaller than the Km value of SB1 (0.17 M).
(3) Specificity: Digestion of bradykinin
and angiotensin II by FA2 was studied to in-
vestigate the specificity of this enzyme. The
mode of cleavage of bradykinin by stem brome-
lain was reported earlier (27).
Figure 6 shows the time course of the loss
of biological activity of bradykinin or angio-
tensin II as determined from the contraction
F. YAMADA, N. TAKAHASHI, and T. MURACHI
20 0
INCUBATION ( m i n )
Fig. 6. Action of fruit and stem bromelain on
bradykinin (a) and angiotensin II (b). The ordinate
shows the ratio in percent contractile response of
isolated rat uterus as recorded on the kymogram.
Contraction of the rat uterine muscles in response
to 0.03 fig of either of the peptides was defined as
100%. • , with fruit bromelain FA2, 10/ig in (a)
and 32 eg in (b); O, with stem bromelain 10 fig in
(a) and 25 fig in (b). The conditions of incubation
are described in the text.
of rat uterine muscles. Contraction of the rat
uterine muscles in response to 0.03 fig of either
of the peptides was defined as 100%, and the
ratios of the heights of contractile responses
to the peptides after a given period of incuba-
tion were expressed as percentages of this
value. Because of the existence of a uterine
muscle contraction threshold, the contractile
ratio does not necessarily represent the ratio
of decrease in bradykinin or angiotensin II.
However, the results shown in Fig. 6 suggest
that the action of FA2 on these peptides is
different from that of stem bromelain, probably
both qualitatively and quantitatively.
In order to determine the site or sites of
cleavage, the following experiments were car-
ried out. One mg of the substrate peptide
was incubated with 1 mg of FA2 at 37° in the
presence of 2 mM 2-mercaptoethanol for 60
min in the case of bradykinin and 120 min in
the case of angiotensin II. Parallel runs with
stem bromelain in place of FA2 were conducted.
The reaction mixture was then subjected to
high-voltage paper electrophoresis at pH 3.5.
The paper was stained with ninhydrin, Pauly
and Sakaguchi reagents. The results are shown
in Fig. 7. The peptide fragments correspond-
ing to these spots were recovered from the
/ . Biochem.
FRUIT BROMELAIN FA2 1231

TABLE IV. Amino acid composition and amino-terminal residues of the fragments obtained by FA2 diges-
tion of bradykinin.*

Total
Amino acid BK1 BK2 BK3
Found Theoretical
Serine 0.56 0.43 0 0.99 1
Proline 1.57 0.49 1.36 3.43 3
Glycine 0.48 0 0.74 1.22 1
Phenylalanine 1.57 0.43 0 2.00 2
Arginine 0.73 0.40 0.46 1.59 2

Amino-terminal residues Arginine, Serine Arginine

phenylalanine

• Fragments BK1, BK2, and BK3 were eluted from the paper (see Fig. 7). Figures in the table are molar
ratios. Since the data for BK1, BK2, and BK3 were obtained from a single sample of the parent peptide,
they may approximate to the relative yields of the fragment peptides. Methods of analysis and calculation
are described in the text.

(a)BRADYKININ
(a)BRADYKININ
BKI 1 2 3 4 5 6 7 8 9
I Arg-Pro-Pro-Gly— Phe-rSer-Pro-Phe-Arg
2 BKI |
I- -I BKI
00 BK2
BK3 I-
h -i
(b) ANGIOTENSIN
ATI
(WANGI0TEN3NK
1 2 3 4 5 6 7 8
5 •
Asn-Arg-Val-Tyr-lle —Hls-Pro-Phe
6 00 ATJ !
10 20 30 j AJJ2
DISTANCE FROM ORIGIN (an)

Fig. 7. Paper electrophoresis of the digests of
bradykinin (a) and angiotensin II (b) by fruit and
stem bromelain. Electrophoresis at pH 3.5 and 4,000
volts, for 1 hr. 1, FA2 digest; 2, bradykinin as a
reference ; 3, stem bromelain digest; 4, FA2 digest;
5, angiotensin II as a reference; 6, stem bromelain
digest. The tracings are of ninhydrin stainings; Y
is yellowish, S is Sakaguchi-positive, and P is Pauly-
positive.
paper, and amino acid analysis and amino-ter-
minal group determination of each fragment
were carried out. The results for the brady-
kinin fragments are shown in Table IV. It is
apparent that peptide BKI is a mixture of at
least two peptides. From the known amino
acid sequence of bradykinin the modes of
cleavage of the peptide by FA2 were deduced
Fig. 8. Mode of cleavage of bradykinin and angio-
tensin II by fruit bromelain FA2.
to be as shown in Fig. 8a. From the relative
yields of peptides BK2 and BK3 (Table IV), it
is apparent that cleavage of bradykinin by
FA2 between Gly*-Phe5 and between Phe5-Ser«
took place at comparable rates. The angio-
tensin II fragments gave the following results:
for ATI, aspartic acid (1.52 in molar ratio),
valine (1.22), tyrosine (1.00), and arginine (1.02)
with asparagine as the amino-terminal group;
for AT2, proline (1.26), isoleucine (0.66), phenyl-
alanine (1.30), and histidine (0.77) with iso-
leucine as the amino-terminal group. Pre-
ferential cleavage of angiotensin II by FA2
between TyrMle 5 is evident (Fig. 8b).

Vol. 79, No. 6, 1976

1232 F. YAMADA, N. TAKAHASHI, and T. MURACHI

TABLE V. Comparison of the properties of fruit bromelain FA2, stem bromelain SB1, and papain.

FA2 from SB1 from Papain from

Properties
pineapple fruit1 pineapple stem b papaya latex c

Molecular weight 31,000 28,000 23,400

d
Isoelectric point 4.6 9.55 8.75=
Absorbance at 280 nm of a
1% solution
19.2 20. Id 25.0

Molar extinction coefficient 59,500 56,300 58,500

Sedimentation coefficient, s 2.75 S 2.77S 2.42 S
Amino acid composition :
Lys+Arg
17/57 22/42 22/39
Asp+Glu
Gly/AJa (molar ratio) 35/25 27/30 28/14
Tyr+Trp (mole %) 8.30 8.54 11.32
Carbohydrate content (%) None 2.1i None6
Amino-terminal sequence Ala-Val-Pro-GIn Val-Pro-Gln Ile-Pro-Gln
Carboxyl-terminal residue Glyf Gly Asn
Sequence around the reactive
Pro-Cys-Gly-Ala-CYS Pro-Cys-Gly-Ala-CYS Ser-Cys-Gly-Ser-CYS
SH group (CYS)
pH optima for: casein 8.3
hemoglobin 8.0 5-6e
BAEE 5-8i
BAA 5-7.5"
1
Specific activity
(units/mg protein) toward :
casein 11.6 (1.00) 6.86 (0. 59) J 6.69 (0. 58) J
0.01 M BAEE 12.4 (1.00) 1.10 (0.089) J 16.9 (1.36) J
Am (M) for: BAEE 0.043 0.17<i 0.023
BAA 0.0040' 0.00121 0.040
Hydrolysis of peptide k :
i i I 1
bradykinin Gly-Phe, Phe-Ser Phe-Ser' Gly-Phe, Phe-Seri
1
angiotensin II Tyr-Ile Several points™
Double immunodiffusion Cross-reaction No visible
Cross-reaction" cross-reaction"
(vs. anti-stem bromelain) with spur"
Percentage cross-reaction
with anti-3tem bromelain 20° 100° 3"

1
Present paper, except where otherwise indicated. b Takahashi et al. (2), except where otherwise indicated.
c
Glazer and Smith (28), except where otherwise indicated. d Murachi ( / ) . • Smith and Kimmel (29).
' Ota et al. (4). e Unpublished observation. h Arnon (30). ' The ratio of specific activities is shown
in parentheses, taking the value for FA2 as unity. J Murachi and Neurath (33). k Arrows indicate peptide
bonds susceptible to hydrolysis. ' Murachi and Miyake (27). m Present paper. ° Sasaki et al. (32).
The major component of fruit juice acetone powder was used.

/ . Biochem.
FRUIT BROMELAIN FA2
DISCUSSION
The product of purification, FA2, was found
to be satisfactorily homogeneous as judged by
several independent criteria. The finding that
FA2 had alanine as its sole amino-terminal
residue is particularly significant, because fruit
bromelain of such high purity has never been
reported before. Purification of the major
proteinase component from pineapple fruit
juice was reported by Ota et al. (3, 4), but
the product they obtained gave, upon amino-
terminal analysis, 1 mole of alanine together
with 0.3 mole each of glycine, valine, aspartic
acid, and glutamic acid (3, 4). According to
these authors, fruit bromelain has a molecular
weight of approximately 18,000 ( 4 ) and it firm-
ly binds some 3% neutral sugars (3, 4). We
have established, however, with our highly
purified preparation that fruit bromelain FA2
is not a glycoprotein but a simple protein hav-
ing a molecular weight of approximately 31,000.
Ota et al. used pineapple fruit obtained at a
town market in Japan (4), while we used the
acetone powder of fruit juice prepared in
Hawaii. However, this may not entirely ac-
count for the conflicting results.
Comparison of the properties of FA2 from
pineapple fruit tissue with those of the enzyme
from stem tissue, SB1, is of great interest.
Such a comparison can also be extended to
papain ( 2 8 - 3 1 ) from papaya (Table V). The
pineapple plant belongs to the Monocotyledonae,
while papaya belongs to Dicotyledonae. The
tabulated data may be summarized as follows.
FA2 is a simple protein with an acidic iso-
electric point, while SB1 is a basic glycoprotein.
Papain is basic, but is a simple protein. The
molecular size decreases in the order FA2,
SB1 and papain. The amino acid sequence
around the reactive cysteine residue is identical
in FA2 and SB1, whereas the amino-terminal
sequence of FA2 has one extra residue, alanine
which precedes the common sequence, Val-Pro-
Gln. FA2 and SB1 show significant differences
in catalytic properties, e.g., optimum pH to-
ward denatured hemoglobin, specific activity
toward BAEE, and kinetic parameters. It is
interesting to note that the mode of action of
Yol. 79, No. 6, 1976
1233
FA2 on bradykinin or angiotensin II differs
greatly from that of SB1 but rather resembles
that of papain.
Table V includes some of the immuno-
chemical data recently reported by Sasaki et
al. {32), which indicate an appreciable degree
of surface homology between FA2 and SB1
proteins. Nevertheless, it is apparent that
these two proteins are independent molecular
species which the pineapple plant may synthe-
size separately and store in different tissues.
REFERENCES
1. Murachi, T. (1970) Methods Enzymol. 19, 273-
284
2. Takahashi, N., Yasuda, Y., Goto, K., Miyake,
T., & Murachi, T. (1973) / . Biochem. 74, 355-
373
3. Ota, S., Moore, S., & Stein, W.H. (1964) Bio-
chemistry 3, 180-185
4. Ota, S., Horie, K., Hagino, F., Hashimoto, C ,
& Date, H. (1972) / . Biochem. 71, 817-830
5. Chittenden, R.H. (1892) Trans. Conn. Acad. Sci.
8, 281-308
6. Yamada, F. (1974) D. Med. Sc. degree thesis,
Nagoya City University
7. Anson, M.L. (1939) / . Gen. Physioi. 22, 79-89
8. Werle, E. & von Roden, P. (1939) Biochem. Z.
301, 328-337
9. Andrews, P. (1965) Biochem. J. 96, 595-606
10. Ornstein, L. (1964) Ann. New York Acad. Sci.
121, Art. 2. 321-349
11. Davis, B.J. (1964) Ann. New York Acad. Sci.
121, Art. 2. 404-427
12. Zacharius, R.M. & Zell, T.E. (1969) Anal.
Biochem. 30, 148-152
13. Weber, K. & Osborn, M. (1969) / . Bid. Chem.
244, 4406-4412
14. Dunker, A.K. & Rueckert, R.R. (1969) / . Biol.
Chem. 244, 5074-5080
15. Lowry, O.H., Rosebrough, N.J., Farr, A.L., &
Randall, R.J. (1951) / . Biol. Chem. 193, 265-275
16. Spackman, D.H., Stein, W.H., & Moore, S. (1958)
Anal. Chem. 30, 1190-1206
17. Edelhoch, H. (1967) Biochemistry 6, 1948-1954
18. Goodwin, T.W. & Morton, R.A. (1946) Biochem.
}. 40, 628-632
19. Inglis, A.S. & Liu, T.Y. (1970) / . Biol. Chem.
245, 112-116
20. Winzler, R.J. (1955) Methods Biochem. Anal. 2,
279-311
21. Rosenman, S. & Daffner, I. (1956) Anal. Chem.
28, 1743-1746
1234
22. Konigsberg, W. (1967) Methods Enzymol. 11,
461-469.
23. Edman, P. & Begg, G. (1967) Eur. J. Biochem.
1, 80-91
24. Jeppsson, J.O. & Sjoquist, J. (1967) /Ina/.
Biochem. 18, 264-269
25. Sasaki, M., Iida, S., & Murachi, T. (1973) /.
Biochem. 73, 367-375
26. Iida, S., Sasaki, M., & Ota, S. (1973) /.
Biochem. 73, 377-386
27. Murachi, T. & Miyake, T. (1970) Physiol. Chem.
Physics 2, 97-104
28. Glazer, A.N. & Smith, E.L. (1971) The Enzymes
(Boyer, P.D., ed.) 3rd ed., Vol. 3, pp. 501-546,
Academic Press, New York
F. YAMADA, N. TAKAHASHI, and T. MURACHI
29. Smith, E.L. & Kimmel, J.R. (1960) The Enzymes
(Boyer, P.D., ed.) 3rd ed., Vol. 4, pp. 133-173,
Academic Press, New York
30. Arnon, R. (1970) Methods Enzymol. 19, 226-244
31. Murachi, T. & Takahashi, N. (1970) Symposium
on Structure-Function Relationships of Proteolytic
Enzymes (Desnuelle, P., Neurath, H., & Ottesen,
M., eds.) pp. 298-309, Munksgaard, Copenhagen
32. Sasaki, M., Kato, T., & Iida, S. (1973) / .
Biochem. 74, 635-637
33. Murachi, T. & Neurath, H. (1960) / . Bio/.
Chem. 235, 99-107
/ . Biochem.