PHYSICAL CHEMICAL STUDIES OF (I) A PHOSPHATASE FROM PIG, (II)
CYTOCHROME P450 FROM Pseudomonas putida, AND (III) ELECTRONIC
QUENCHING OF Al AND Ga ATOMS IN RARE GAS MATRICES
DAVID CARL SCHLOSNAGLE
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS OF THE
DEGREE OF DOCTOR OF PHILOSOPHY .
UNIVERSITY OF FLORIDA
To my parents and my wife, Sara
I would like to thank the following:
Dr. J. H. Ammeter for his assistance on the A1/Ga project.
Dr. F. W. Bazer and Dr. R. M. Roberts for crude allantoic
fluid and the invitation to work on the phosphatase.
Dr. J. C. M. Tsibris and Dr. W. Weltner, Jr. for their support
and encouragement in this entire undertaking.
The many others who have helped in various ways during the
last four years.
TABLE OF CONTENTS
ACKNOWLEDGEMENTS. ... .. .. .. .. .. .. .. . iii
LIST OF TABLES. .. .. .. .. . .. ... .. v
LIST OF FIGURES . .. .. ,. ,. .. ,, vi
SYMBOLS AND ABBREVIATIONS .. .. .. ,,. . ... viii
ABSTRACT. .. .. . . .. .. . .. .. x
I. PROPERTIES OF A PROGESTERONE-INDUCED, IRON-CONTAINING
PHOSPHATASE FROM THE ALLANTOIC FLUID OF PIGS, ,,. .. . 1
Introduction. . . .. .. . ... . 1
Materials and Methods .. ... .. .. .. 5
Results and Discussion. .. .. . ... .. . 13
General Discussion. . . ... .. . 66
References. .. .. .. .. .. . . 75
II. INTERACTION OF NITRIC OXIDE WITH CYTOCHROME P450 AND
P420 FROM Pseudomonas putida. .. .. .. .. .. .. .. 79
Introduction. ... .. .. .. ... .. .. 79
Ilaterials and Methods .. .. .. .. ... .. 84
Results and Discussion. .. .. .. .. .. . 86
Conclusions .. .. ... .. .. .. .. 102
References. ... ... .... . .. .. .. 109
III. ELECTRONIC QUENCHING OF Al AND Ga ATOMS ISOLATED IN
RARE GAS MATRICES .. .. .. ... .. .. ... . 112
Introduction. ... .. .. . .. .. .. 12
Materials and Methods . ... .. .. .. .. ... 115
Results . .. .... . ... .. . . 117
Discussion: The Crystal Field Model. .. ... 152
Conclusions .. . .. .. .. .. . .. 167
References. ... .. .. .. .. .. .. .. .. 171
BIOGRAPHICAL SKETCH . .. . . .. .. .. . 173
LIST OF TABLES
I-1. Amino acid analysis of purified Fraction IV
protein from pig uterus. .. . ... ... .. 3
I-2. Hydrolytic activity of PP towards various esters 44
I-3. Rate effectors of pNPPase activity .. ... 52
I-4. Effect of reductants and oxidants on the pNPPase
activity of the purple form of PP. . ... .. 55
1-5. Reconstitution of pNPPase activity by addition
of metals to apoPP . .. .. .. .. ... .. 61
II-1. Optical data of cytochrome P450 in the presence
of nitric oxide. .. .. .. . . . .. 89
III-1, Optical absorption spectra of Al atoms in rare
gas matrices . . .. . .. . . . . 122
III-2. Optical absorption spectra of Ga atoms in rare
gas matrices .. .. .... .. .. . .. 123
III-3. Average radii of the valence orbitals of some
atoms with an outermost p shell. .. ... .. 126
III-4. Matrix-shifts for (n+1)s=-np transitions .. .. 128
III-5, ESR data for matrix-isolated Al atoms. . .. . 130
III-6. ESR data for matrix-isolated Ga atoms. .. .. 135
III-7. Temperature shifts of the magnetic parameters. . 148
III-8. Non-zero matrix elements of 9 so + ax. applied
to the six-dimensional complex basis .. . ... 155
III-9. Crystal field analysis of the ESR data of
natrix-isolated 27Al atoms .. .. . .... 164
III-10. Crystal field analysis of the ESR data of the
g values of matrix isolated gallium atoms. . 165
LIST OF FIGURES
I-1. Elution pattern of CM-cellulose positive
proteins, . . .. ,,,, ,, , 7
I-2. Optical absorption spectra of PP. . .. ... 15
I-3. Effect of 8202 on the optical spectrum of
the pink form of PP . .. .. .. .. .. 18
I-4. Circular dichraism spectrum of PP . .. .. .. 22
I-5. Liquid nitrogen ESR spectrum of PP in
acetate, buffered, pH 4.9 ... .. . .,. 23
I-6. ESR of PP at 77oK (iron signal) ... . ... 26
I-7. ESR of PP at 770K (copper signal) .. . . 27
I-8. Sulfhydryl titration. . .. .. .. .. .. .. 29
I-9. Linearity of reaction with enzyme concentration 31
I-10. Effect of ionic strength. . . .. .. .. 33
I-11. pH profile. .. .. .. .. .. .. .. . 34
I-12. Linearity of reaction with time .. .. .. .. 37
I-13. Reaction velocity versus substrate
concentration ... .. .. .. ... .. 42
I-14. Lineweaver-Burk plot of the data of
Figure I-13 ... ... .. . . . .. 43
I-15. Denaturation of diluted PP. .. ... . 45
I-16. Heat stability. .. .. .. .. . . 47
I-17. Energy of activation. .. . ... .. ... 51
I-18. Effect of 2-ME on the reaction velocity .. .. 56
I-19. Restoration of enzymatic activity by addition
of copper and iron to apoPP .. .. .. . 64
II-1. The heme group. .. .. .. . .. .. .. 82
II-2. Optical absorption spectra of complexes of
cytochrome P450 and nitric oxide. ... .. 88
II-3, Effects of nitrogen isotopic substitution on
the ESR spectra of nitric oxide complexes of
reduced P450. .. .. .. .. .. .. .. 93
II-4. ESR of natural abundance (0.10 mM) and 57-Fe
enriched (0.14 mM) reduced cytochrome P450
plus 15NO at 770K ... .. .. .. . ... 95
II-5. ESR spectra of reduced P420 plus 15N0 at 770K 98
II-6. The ESR spectrum of reduced horseradish
peroxidase plus 15NO at 770K, ....... 107
III-1, Electronic absorption spectra of matrix-isolated
aluminum atoms at 4.20K .. .. .. . 119
1II-2. Electronic absorption spectra of matrix-isolated
gallium atoms at 4.2oK. .. .. .. .. ... 121
1II-3. Simplified energy level diagram and resonance
transitions of Group III metal atoms. .. .. 125
III-4. ESR spectra of aluminum atoms in an argon
matrix at 4.2oK ,. . ... .. .. .. 132
III-5(a). ESR spectra of matrix-isolated gallium atoms
in an argon matrix at 4.20K prior to annealing. 139
III-5(b). ESR spectrum of matrix-isolated gallium atoms
in a krypton matrix at 4.20K prior to annealing 141
III-6(a). ESR spectrum of matrix-isolated gallium atoms
in argon at 4.20K after annealing ... .. 143
III-6(b). ESR spectrum of low field parallel lines of
gallium in argon at 4.20K after annealing . .. 145
III-6(c). ESR spectrum of low field parallel lines of
gallium in argon at 4.2oK after annealing .. 147
III-7. ESR spectrum of gallium atoms in xenon at
4.2oK .. .... .. .. .. .. .. .. 151
III-8. Energy levels of an (np)1 atom in an axial
crystal field ... .. .. .. .. .. .. 158
III-9. g tensor of 2P atoms in an axial crystal field. 161
ABBREVIATIONS AND SYMBOLS
A250 absorbance at 250 nm
AMP adenosine monophosphate
apoPP PP without metal cofactor
ATP adenosine triphosphate
CD circular dichroism
ESR electron spin resonance
x g times gravity
811 g parallel
.1 g perpendicular
Km Michaelis constant
Pi inorganic phosphate
pl isoelectric point
PP progesterone-induced, iron-containing phosphate
SDS sodium dodecylsulfate
SH sulfhydry1 group
A crystal field splitting parameter
5 spin-orbit coupling constant
Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
PHYSICAL CHEMICAL STUDIES OF (I) A PHOSPRATASE FROM PIG, (II)
CYTOCHROME P450 FROM Pseudomonas putida, AND (III) ELECTRONIC
QUENCHING OF Al AND Ga ATOMS IN RARE GAS MATRICES
David Carl Schlosnagle
Chairman: William Weltner, Jr.
Major Department: Chemistry
A progesterone-induced, iron-containing protein from the allantoic
fluid of pigs has been purified to homogeneity. There was found to be
1 mole of iron per mole of protein. The absorption maximum of the
protein (PP) varies from 500 to 550 nm but the molar extinction
coefficient (2200) is independent of wavelength. Reducing and oxidizing
agents were found to shift the spectrum of PP to higher and lower
energy respectively. ESR studies at liquid nitrogen temperatures have
indicated that the iron is present as high-spin ferric. Testing for
physiological function demonstrated that PP is an acid phosphatase with
maximal activity at pH 4.9. Of number of substrates tested, PP was most
active toward p-nitrophenylphosphate The average value of Km was
found to be 2.2 mM at 300 and 2.8 mM at 00. The apparent energy of
activation was found to be +11.1 kcal/mole. Under optimum conditions
the greatest specific activity obtained was 170 micromoles/min/mg. This
corresponds to a turnover number of 5400 molecules of substrate/min/mole-
cule of enzyme. The rate of hydrolysis was found to be greatly inhibited
by mercuric ion, fluoride and molybdate. Generally activity was increased
by reducing agents such as 2-mercaptoethanal. The amount of enhancement
was variable but could be as much as 2- to 4-fold with no change in Km.
The iron could be removed from the protein under reducing conditions
with concomitant loss of enzymatic activity; activity could be restored
by addition of ferric or cupric ions.
The complexes of nitric oxide with cytochrome P450 and P420 from
P. putida were studied by optical and liquid nitrogen electron spin
resonance (ESR) spectroscopy. Oxidized P450-NO has an absorption
spectrum of a low-spin ferrihemoprotein and no ESR signal. The ESR
spectrum of reduced P450-NO indicates a paramagnetic center of rhombic
symmetry with g values of 2.08, 2.004 and 1.97. The well-resolved
triplet at g = 2.004 with hfs of 20 G. gave rise to a doublet with
hfs of 28 G. upon isotopic substitution with 1NO. A difference in
the spectra was noted depending upon the presence or absence of sub-
strate camphor. 57e-enriched P450 was also studied, The heme group
of P420 was found to be weakly bound to the protein. Reduced P420-N0
exhibits an axially symrmetric ESR spectrum with g values of 2.08 and
Aluminum and gallium atoms have been trapped in Ne, Ar, Kr and
Xe matrices and studied by optical and ESR spectroscopy at liquid
helium temperatures. The results indicate that both metal atoms occupy
axially distorted sites in all rare gas lattices, The 2 ,2P
electronic transitions are shifted by about +1000 cm-1 (in Xe) to
about +6000 em-1 (in Ne) relative to the free metal atom values.
Increasing the matrix temperature slightly causes a reversible red
shift in these transitions. The ESR spectra exhibit axial symrmetry
and show effects of preferential orientation. While the g values of
the free Group III atoms in their Kramers degenerate 2P4 ground level
are approximately equal to 2/3, the observed g values of matrix-
isolated Al and Ga are near free spin, i.e., almost complete quenching
of the free atom angular momentum. The dependence of the g values on
the matrix and on temperature can be described by a simple crystal
field splitting parameter. The splitting of the aluminum and gallium
p shells are very similar for both atoms and range from about 1000 to
3000 em- and increase from light to heavy rare gas atom.
I. PROPERTIES OF A PROGESTERONE-INDUCED, IRON-CONTAINING
PHOSPHATASE FROM THE ALLANTOIC FLUID OF PIGS
It has been found that protein levels in the uterine secretions
of pigs change both quantitatively and qualitatively during the nor-
mal estrous cycle. There is a slow increase in protein concentra-
tion during the first 9 days after estrous followed by a rapid
change after day 10.1 After reaching a maximum of 50 mg per gilt
on day 15, the values decrease markedly. This decline occurs at a
time when the corpora lutea regress and progesterone levels in
plasma fall. Two new protein fractions isolated via Sephadex G-200
chromatography were not present before day 9 or after day 16 of the
normal cycle. These were called Fractions IV and V, the latter
having at least six components with an estimated average molecular
weight of 20,000 as shown by polyacrylamide gel electrophoresis.
Fraction IV has a characteristic purple color and a somewhat
higher molecular weight than the proteins of Fraction V.2 It could
be resolved into a single band which moved toward the cathode at
pH 8.0 on gel electrophoresis, indicating that the protein is basic.
Treatment with periodic acid Schiff stain for carbohydrates gave
a positive reaction, showing that Fraction V is a glycoprotein.
Gilts were ovariectomized on day 4 after onset of estrous and
administered progesterone (12.2 ag/kg) and estrogen (1.1 yg/kg)
daily until day 15 at which stage the uterine lumen was flushed with
0.33 M sodium chloride.2 The yield of total protein was twice that
obtained from gilts given only progesterone and nearly four times that
of untreated animals. In each set of experiments, the basic purple
protein was found. By contrast, ovariectomized control gilts produced
only small amounts of protein and no protein containing a net basic
charge at pH 7.0. The purple protein was not found in the serum of
pigs in any of the above experimental groups. Thus the purple protein
appears to be progesterone-induced.
PP was purified from uterine flushings via ion-exchange and gel
filtration chromatography.2 Isoelectric focusing experiments gave a
pl of around 9.7. SDS gel electrophoresis and equilibrium ultracen-
trifugation gave a molecular weight of 32,000. Amino acid analysis,
Table I-1, showed a high content of basic amino acids as expected
from the very basic nature of the protein. Ten half-cystine residues
were found via treatment with performic acid, but the oxidation states
of these were not determined. Four neutral sugars were identified by
gas-liquid chromatography: fucase, manncse, glucose and galactose in a
ratio of 2:5:4:4. Sialic acid could not be detected. The protein was
found to contain eight males of glucosamine and one mole of galactosamine
per mole protein. Thus, each molecule of PP contains fifteen residues
of neutral sugar and nine residues of amino sugar, On this basis, the
total carbohydrate content would be about 12% by weight.
Partially purified protein was subjected to are emission analysis
which indicated the presence of iron, about a tenth as much copper and
Table I-1. Amino acid analysis of purified Fraction IV protein
from pig uterus.a
males/mg protein Residues/32,000 mol wtb
Lysine 0.58 19
Histidine 0.27 9
Arginine 0.39 13
Aspartic acid 0.70 23
Threonine 0.46 15
Serine 0.47 15
Glutamic acid 0.56 19
Proline 0.37 12
Glycine 0.58 19
Alanine 0.58 19
Cysteinee 0.30 10
Valine 0.43 14
Methioninee 0.12 4
Isoleucine 0.28 9
Leucine 0.64 21
Tyrosine 0.17 6
Phenylalanine 0.39 13
Tryptophan 0.23 8
Glucosamine 0.24 8
Galactosamine 0.03 1
a From Ref. 2
bCalculated to the nearest whole number.
c Results of analyses on performic acid-oxidized protein,
a trace of zinc. No manganese was found. The uterine protein had an
absorption maximum at 545 am and the color was stable to prolonged
dialysis against EDTA. The allantoic fluid from pregnant animals also
contained the purple basic glycoprotein having the same molecular
weight as the uterine protein as determined by electrophoresis.
A female lamb was immunized by injection with the purple uterine
protein and the presence of antibody was verified in the serum.2 The
basic protein from allantoic fluid was found to be immunologically
identical to the uterine protein. The antiserum (anti-IV) did not
cross react with any of the acidic proteins obtained from the uterine
flushings nor with extracts from tissues of the following organs:
heart, lung, stomach, intestine, liver, spleen, kidney and oviduct
tissue. Further, no reaction occurred with the serum of a pig obtained
on day 15 of estrous. Sheep anti-IV was administered to gilts several
times early in pregnancy and the embryos and placentae later were
examined." There was a significant reduction in placental and fetal
crown-rump length. Between days 8 and 16 of gestation, when the
purple protein is present in large amounts, rapid elongation of the
blastocyst occurs. It is during this period that the major portion
of embryonic deaths occur.5
Only two other progesterone-induced proteins have been purified
to homogeneity: avidin from chicks and blastokinin from rabbit.7 The
purple glycoprotein from pig (PP) has been found in both the uterine
flushings and the allantoic fluid, but in no other tissue. Further,
this protein appears to be necessary to the growth and development of
the fetus. This work was undertaken to determine the biological func-
tion of this purple protein and study the role of the iron.
Materials and Methods
Purification from Allantoic Fluid
Brown allantoic fluid from day 45 was received after having been
passed through a Millipore filter to remove bacteria. The following
purification was carried out in the cold. The crude solution was
dialyzed against 10 mM sodium acetate for 6 hours with two changes.
The dialyzed fluid was centrifuged for 30 minutes at 27,000 x g and
the pellets discarded. A column (1 x 12 cm) of CM-cellulose was
prepared using Whatman CM 52 treated according to manufacturer's
instructions. The column was washed with several volumes of 10 mM
sodium acetate and the allantoic fluid was added. Proteins having a
net positive charge at pH 7.5 will be retarded by this column. The
effluent was light yellow-brown and was later lyophylized and stored
at -200C. The OR column soon took on color, as a brown protein present
began to accumulate on the resin.
After the last of the allantoic solution was added, the column
was washed with 2 volumes of 10 mM sodium acetate followed by 0.1 M
and 0.2 M acetate which eluted the brown protein. The visible absorp-
tion spectrum was taken of an aliquot of this fraction and was found to
be typical of hemoproteins with alpha, beta and Soret bands. No attempt
was made at identification. At one point a narrow single band of pink-
orange coloration was seen moving down the column nearly free of any
other visible bands. The optical spectrum of this fraction resembled
the reduced form of cytochrome c. 1 M sodium acetate was used to elute
the violet-colored band still remaining on the column. An aliquot of
this fraction was scanned in the visible region and found to have a
broad absorption at 505 nm, a weak peak from contaminating hemoprotein
and a shoulder around 320 nm. Part of the more dilute violet fractions
were lyophilized and stored at _200 for later use. The remaining frac-
tions were pooled and concentrated via vacuum dialysis.
Half of this protein solution was placed on a room-temperature
column (2 x 100 em) of Sephadex G-100, previously prepared and equili-
brated with 0.1 M sodium acetate. The Sephadex beads serve to separate
molecules according to size. The transmission at 280 nm was continuously
and automatically monitored. The second aliquot of the purple protein
was passed through the same Sephadex G-100 column, this time equili-
brated with 0.1 M (Na $) acetate buffer, pH 4.8. The two elution
patterns are shown in Figure I-1. In the first case, there are three
bands, two of which are colored. The first fraction is pink and the
second purple with absorption maxima of 505 nm and 535 am respectively.
In the second fractionation, only one colored band was obtained. The
leading shoulder on this band was from an impurity. SDS polyacrylamide
gel electrophoresis was used to determine purity.
Protein concentration was determined by a number of ways. Pure PP
could be quantitated by optical absorption where an optical density of
1.0 at 280 nm corresponds to 1.0 mg protein/m1. A molar extinction
coefficient of 2200 at 550 nm was also used to calculate PP concentration.
Two common analytical procedures in use are the LowryB and the
biuret methods. The Lowry is at least twenty times more sensitive than
the biuret, but suffers from non-linearity at higher protein concentra-
tions. A good relationship was found between PP dry weight and the Lowry
method. The biuret method is quicker but requires more protein.
ELUTION VOLUME ---
Figure I-1. Elution pattern of CM-cellulose positive
proteins. The proteins were fractionated via Sephadex G-100
chromatography (see text) A: The column was equilibrated
with 0.1 M NaAc at pH 7.5. The first band is pink, the second
purple. B: The column was equilibrated with 0.1 M acetate
pH 4.8, The fiikst main band is of violet coloration. The
two patterns are not on the same abscissa.
The Lowry reagents are prepared as follows:
A. 2% Na2CO3 in 0.10 N Na0H
B. 0.5% CuSO *5H20 in 1% tartrate
C. 50 ml A plus 1 ml B (good one day only)
D. 1 N Folin Ciocalteu reagent (from Sigma)
Procedure: To 0.2 ml of each protein sample (0.05 to 0.5 mg/ml) add 1
m1 of reagent C and let stand 10 minutes. Then add 0.10 ml reagent D
with rapid mixing and allow to develop for 60 minutes. Read absorbance
at 750 nm. Bovine serum albumin was used to prepare the standard curve.
The biurat solutions are prepared as follows. Reagent A contains
1.5 g CuSO '5H20 and 6.0 g solution potassium tartrate in a final volume
of 0.50 1. Reagent B is 10% carbonate free sodium hydroxide. The binret
reagent was prepared prior to use by mixing equal volumes of A and B. The
unknown protein solution (1 to 10 mg/ml) is added to the biuret solution
in a volume ratio of 1:5. The absorbance at 540 nm is recorded after 15
minutes. Bovine serum albumin was used as the standard.
The method below is that described for nonheme iron in serum using
tripyridyl-S-triazine (TPTZ) as the chromogen.
To 0.5 ml of protein add 0.5 ml 1 N HC1 and allow 10 minutes for the
iron to be split off. Add 0.5 ml of 10% TCA to precipitate the protein.
Centrifuge at 2500 x g for 15 minutes, then remove 1.0 ml of protein-free
supernatant. Add 0.5 ml of a freshly prepared mixture of 50% ammonium
acetate, 10% hydroxylammonium chloride and 0.004 M TPTZ in a volume
ratio of 2:1:1. After 10 minutes record the absorbance at 593 nm.
Ferrous armmonium sulfate was used as standard. All glassware was
soaked in 1 N HC1 overnite and rinsed in deionized or glass-distilled
water and dried prior to use.
All UV visible absorption spectra and 0.D. measurements were taken
on a Beckman Acta C III double beam instrument with digital readout. The
recorder scale was variable in increments of 0 to 0.1, O to 1, O to 2
and 0 to 3 0.D. units. Thunberg-type cuvettes were used for anaerobic
Electron spin resonance spectra were obtained using Varian E3
and E9 spectrometers which operate at X-band. 100 kd~z modulation was
used in both cases. Spectra at 770 K were obtained by means of a
Scanlon dewar insert or with a Heli-Tran variable temperature accessory.
In the latter case, the temperature was monitored using a chromel vs.
gold (with 0.07 atom % iron) thermocouple. The microwave frequency was
calibrated by insertion of a quartz capillary containing solid DPPH
(g = 2.0037).10
Circular dichroism measurements were made with a Jasco CD/0RD 20.
The instrument was calibrated with 10-camphorsulfonic acid prior to us6.11
Measurement of Phosphatase Activity
Phasphatase activity was determined by two methods: either by
measurement of inorganic phosphate or, when p-nitrophenyl phosphate
(pNPP) was used as substrate, measurement of p-nitrophenol (pNP) as the
nitrophenylate ion. Controls containing no enzyme were employed with
every series of trials. The specific activity of an enzyme is expressed
as micromoles of product formed per minute per mg protein added under
the conditions of the assay. Eppendorf pipettes, used in the addition
of microliter volumes, were calibrated with distilled water at 220.
Inorganic phosphate determination
This methodl2 is a modification of the Fiske-SubbaRow assay.13
Reagent A consists of 1%b Elon (Kodak) and 3% sodium bisulfite. Reagent
B is prepared by diluting 272 ml concentrated sulfuric acid to 700 ml
with deionized water and the solution allowed to cool. 50 g of armmo-
nium molybdate are added to 100 ml deionized water. These are then
mixed and diluted to a final volume of 2000 mL.
After suitable incubation of substrate with enzyme, the reaction
is stopped by addition of reagent B. Reagent A is then added and the
absorbance at 660 nm recorded after 15 minutes. The volume ratio of
reaction mixture:reagent A:reagent B is 8:1:1. It has been reported
that adenosine phosphates and pyrophosphates interfere with the assay
by the formation of a colorless complex with the molybdate.14 A 4-fold
excess of molybdate over substrate insures normal color development.
K2HPO4 was used as the standard.
p-Nitrophenylate ion determination.
The enzyme will cleave pNPP into inorganic phosphate (Pi) and pc
nitrophenol. The anion formed upon ionization of the phenolic proton
absorbs around 410 nm, whereas neither pNPP nor pNP absorbs appreciably
above 400 nm.15 Thus the amount of product can be easily determined
Continuous assay. The initial assays were carried out at room
temperature in 0.1 M ammonium acetate, pH 6.0, 12 mM in pNPP, Enzyme
is added to an optical cuvette containing the buffered substrate solu-
tion, the contents mixed and the absorbance at 405 nm is followed vs.
time using the chart drive of the Acta. pNP in buffer was used as
standard. The linear portion of the trace was used in calculating the
initial reaction velocity.
Discontinuous assay. The reaction is carried out under the desired
conditions and stopped by addition of 0.25 N KOH. pNP liberated is
determined using a molar extinction coefficient of 16,200 at 410 na.15
In the early assays, the data were accumulated as the average of
six trials since there was a considerable fluctuation in successive
determinations. Better consistency was obtained with 0.1 M sodium
chloride present. The best results were achieved using silanized test
tubes, with which replication of assays was within 5% of the average
value. The test tubes were first washed thoroughly and dried. They
are then immersed for about one minute in a solution of 1% dimethyl
dicl0rosilane in benzene at 550. The hydrophobic layer formed on the
surface of the test tube apparently prevents the basic protein from
absorbing onto the glass.
Sulfhydryl groups will react with a number of metals to form mer-
captides. Divalent mercury has two sites open to coordination, thus
only monomercaptides are formed when the mercury is covalently bound
to an organic residue. One such organometallic is p-mercuribenzoate
(pMB) which gives rise to an ultraviolet spectral shift upon mercaptide
formation. The change is greatest around 250 nm where there is an
increase of absorption. The nature of this reaction has been discussed
The method used was that of Riordan and Vallee.17 About 0.08 mM pMB
is prepared by dissolving in a slight excess of alkali, diluting to
approximate concentration with desired buffer and centrifuging to remove
precipitate. One ml of this solution is added to one of a pair of
matched optical cuvettes; the other cuvette contains 1.00 ml buffer
only. Ten microliters (2-3 nanomoles) of PP are added to both reference
and sample cuvette. Therefore, any contribution of the protein to the
UV absorbance is blanked out with the double beam spectrometer. The
contents are mixed and the increase in absorbance due to mercaptide
formation is followed at 250 nm. When the O.D. no longer changes, the
absorbance is recorded and another aliquot of protein added. All the
pMB has reacted when the absorbance no longer increases. A plot of
A250 against nanomoles of PP added will show two straight lines, the
intersection of which gives the stoichiometry of the reaction. The
concentration of pHE can be standardized either by using a molar
extinction coefficient of 16,20016 at 233 nm, pH 7.0, or by titration
with a solution of reduced glutathione, a small molecule containing a
single thiol group. These methods agreed to within about 5%.
Electrophoresis was carried out at pH 4.5 using 6-alanine buffer as
described by Reisfeld et al.18 A current of 6 to 8 mA per tube was run
for 30 minutes with the anade at the top. Gels were stained for pro-
tein using 0.125% (w/v) Coomassie blue in 10% (v/v) acetic acid and 40%
Results and Discussion
Chemical and Physical Properties
Based on the iron determination of PP from different pigs at differ-
ent stages, there is 1.0 (f 0.1) mole of iron per 32,000 molecular weight.
PP can be obtained in various forms which absorb between 500 nm (the
pink form) and 550 nm (the purple form). The optical spectrum of the
purple species, shown in Figure I-2, also has a shoulder at 320 nm and
peaks at 280 nm and 215 nm (not shown). The absorption at 280 nm is due
to the aromatic amino acids tryptophan, tyrosine and phenylalanine which
are present in a ratio of 8:6:13. This band at 280 nm strongly resembles
that of tryptophan as this amino acid absorbs much more strongly than
the other two.l9 The absorption in the far UV is due to transitions of
the peptide bond.20 The spectrum of the pink form is similar to that
of the purple except for the shift of the visible absorption maxima to
higher energy. Also the 320 nm shoulder is either absent or hidden under
the 280 nm peak as it is no longer distinguishable. One can find a cor-
relation between the color of the protein and intensity of the 320 nm
shoulder; the more purple the color, the stronger the 320 nm absorption
Figure I-2. Optical absorption spectra of PP.
The intensity of trace A has been increased 10-fold
and that of trace B 2-fold compared to that of trace C.
The concentration of PP is 45 fM in 0.1 M acetate at
N C 1 -
and vice versa. A mixture of the pink and purple species will have a
visible absorption maximum between the two noted extremes.
The molar extinction coefficient at 550 nm for the purple form was
calculated to be 2200 based on iron content. The absorbance of 1.0
0.D. at 280 nm is indicative of 1.0 mg protein per ml as determined both
by the biuret method and by the Lowry method. This gives a theoretical
ratio of A280/A550 of 14. Approximate purity of protein preparations
could be readily determined from this ratio.
As the pink form was eluted from the Sephadex column with an
apparent molecular weight greater than that of the purple, it was
thought that perhaps the former was a dimer. A change in protein con-
formation upon going from monomer to dimer could influence the environ-
ment about the iron atom which would explain the shift in the visible
region of the spectrum. Since this protein is rich in sulfhdryl groups,
dimerization could be due to the formation of intermolecular disulfide
bonds. These would be broken by sulfhydryls. 0.1 M 2-ME however, had
no effect on the visible spectrum of the pink form. On the other hand,
2-ME caused the color of the purple form to change to pink within about
ten minutes. Cysteine and ascorbic acid had the same effect, whether by
direct addition or by dialysis and subsequent removal. The spectrum of the
nascent pink was the same as the naturally occurring species. The addition
of low concentrations of dithionite, a powerful reducing agent, also
caused the same spectral shift but with gradual loss of visible absorption.
At higher concentrations, dithionite caused immediate loss of color.
Prolonged exposure to 2-ME or ascorbate also has this bleaching effect.
Fractions of the pink form could be made purple by oxidizing agents,
either within a few minutes by addition of 1 mM hydrogen peroxide
(Figure I-3) or more slowly by dialysis against ferricyanide.
Figure I-3. Effect of H202 on the optical
spectrum of the pink form of PP. Solid line:
1.0 ml of PP as isolated after dialysis against 0.1 M
acetate at pH 4.8; dotted line: same sample scanned
2 minutes after the addition of 0.1 ml of 8.8 mM
oxidant. As mentioned in the text, reducing agents
shifted the spectrum of the purple form of PP to
4/1 m N s**
The rotation of the plane of polarized light and the unequal absorp-
tion of right- and left-circularly polarized components, that is, opti-
cal activity, is a property of a molecule that is not superimposable on
its mirror image.21 Such molecules are referred to as dissymmetric and
can have no axis of improper symmetry. Commonly, dissymmetric molecules
have no symmetry at all or belong to point groups having only proper rota-
tion such as Cn or Dn These molecules have the property of possessing
different molar absorbancies and refractive indices for circularly
right- and left-polarized light.
A beam of plane polarized light can be regarded as two beams of
right- and left-circularly polarized light rotating in phase with equal
amplitudes. A circularly polarized beam is one in which the electric
vector rotates through 2n in a direction perpendicular to the direction
of propagation. The resultant electric vector is the sum of the right
and left rotating electric vectors and appears to be confined to a
plane when viewed along the direction of propagation and traces out a
sine wave when viewed perpendicular to the direction of propagation.
When plane-polarized light passes through an optically active sample,
the difference in the refractive index causes a retardation of one of
the electric vectors and the difference in molar absorbancy causes a
difference in vector amplitude. The vector sum will trace out an
ellipse. 6 may be defined as the angle whose tangent is the ratio of
the minor to major axis of the ellipse. a is directly proportional to
the difference in molar extinction coefficients between right and left
polarized light, i.e., (El Er) = AE. The observed ellipticity can be
converted to the molar ellipticity by
O = (100-90bs)/(1*C)
where 1 is pathlength of light through the sample and C is the molar
concentration of sample. The units of OM are (deg~cm2)/dmol. The unequal
absorption and unequal velocity of transmission of right- and left-
circularly polarized light is called the Cotton effect.
The circular dichroism spectrum of PP is shown in Figure I-4, where
protein concentrations of 0.367 mM and 0.045 mM in 1 cm pathlength cells
were used in the visible and ultraviolet regions respectively. The sam-
ple of enzyme used had an absorption maximum at 545 nm whereas the maxi-
mum molar ellipticity occurred at 520 nm and was calculated to be 5700
(deg~cm2)/dmol. The positive peak around 275 nm is due to the aromatic
amino acids. The sharp negative peak at 294 nm and positive peak at 289
nm is probably due to tryptophan; studies with model tryptophanyl compounds
have shown this to be the region where C.D. maxima occur.22 In addition,
there are bands of both positive and negative ellipticity in the near
ultraviolet region. The ellipticity in this region is not due to the
aromatic amino acids as tyrosine and phenylalanine do not contribute
above 300 nm and tryptophan can generate ellipticity only up to 320 nm?2
The ellipticity of the peaks are of the same magnitude as found for a
number of disulfide bonds.23 The position and intensity of the UV absorp-
tion depends upon the dihedral angle of the disulfide group. However,
molecules containing disulfide bonds which absorb above 300 nm have
extinction coefficient about ten times less than that which can be
approximated for the 320 nm shoulder of PP.24,25
The ESR spectrum of PP at pH 5, taken near 770 K, is shown in
Figure I-5. Seen are two prominent absorptions at about g = 4.3 and
g = 2.0. The point of zero derivative in the low field peak corresponds
Figure I-4. Circular dichroism spectrum of PP.
The protein was buffered in citrate-phosphate at pH 7.
Visible region: 0.367 mM PP, right-hand ordinate;
ultraviolet region: 0.045 mM PP, left-hand ordinate.
A 1 cm pathlength cell was used in both cases; the
baseline was taken using buffer only.
(~O"P/Z"3'8ap) ~-D1 X W~
r- \o vr m N rl O rl N
O vl O O YI O
m N rJ I
5 E-O~ XW~
Figure 1-5. Liquid nitrogen ESR spectrum of PP in acetate,
buffered, pH 4.9. The low field peak is at g = 4.3; the high
field peak is at g = 2.05. The microwave power was set at 25
mWatts and the modulation amplitude at 5 G.
to g = 4.29. There is a low field shoulder and a peak to higher field
at g = 4.04. Thus there appear to be three transitions in this region.
Figure I-6 shows this in expanded scan.
The signal at g = 2.05 is shown expanded in Figure I-7. Clearly
resolved are at least 6 hyperfine lines with splitting of 15 gauss. The
sample of PP used for this ESR experiment was then dialyzed for 11 hours
at 40 against 0.01 M bipyridine at pH 5 in acetate buffer, followed by
dialysis for 26 hours and 3 changes which removed the chelating agent.
ESR spectra were taken using the same quartz sample tube. The signal at
g = 2 was found to be decreased in intensity by about 50% relative to the
g = 4.3 peak. A standard sample of 0.1 mM CuC12 was prepared in 1.0 mM
EDTA and 0.1 M acetate, pH 5.26 The ESR signal intensity around g = 2 was
at least ten times stronger than the g = 2 signal from the protein. The
ratio of the g = 2 peak to the g = 4.3 peak was found to vary in different
preparations of PP.
A third signal, having a peak-to-peak separation of about 120 gauss
was found around g = 11. The intensity of this varied with respect to the
g = 4.3 signal, depending upon the preparation of protein. This signal
was not identified. The signals at g = 4.3 and g = 2.0 which were assigned
toFe3 adC+2 respectively, will be discussed later in this chapter.
The abundance of half-cystine residues in the protein indicates a
strong possibility of PP being an iron-sulfur protein, 7'2 although the
visible spectrum is not similar to the rubredoxins, wherein the sulfur
atoms of four cysteine residues are coordinated about the iron in a dis-
torted tetrahedron. Addition of mercurial to rubredoxin leads to a loss
of visible spectrum.2
Figure I-6, ESR of PP at 77"K, The spectrum
is indicative of high-spin ferric iron, Conditions
are as in Figure I-5.
Figure I-7. ESR of PP at 770K. This signal was assigned
to the cupric ion. The 6 hyperfine lines are split by 15 G.
Conditions are as in Figure I-5.
Titration of PP with a sulfhydryl reagent could possibly give informa-
tion as to the ligands of the iron atom in PP. p-mercuribenzoate was
selected as it is known to be a specific, reversible sulfhydryl reagent.17
The reaction can be quantitated spectrally by the use of difference spec-
pMB at 0.08 mM was found to react very slowly with PP at pH 7, and the
number of reacting groups could not be determined. However, when the
protein was preincubated for 30 to 90 minutes in 1 M guanidine hydro-
chloride, a denaturing agent, the titration showed 9 (r 1) reacting sulf-
hydryl groups of the 10 found by amino acid analysis (see Fig. I-8). The
visible spectrum of guanidine treated PP was found not to greatly change;
the spectrum remained unchanged when a stoichiometric amount of p:MB was
added. From this evidence it seems certain that the iron is not completely
coordinated by sulfurs. From the difference in reactivity of pMB toward
the untreated and guanidine treated PP, it appears as though most of the
sulfhydryls are buried within the three-dimensional structure of the
protein. Away from the surface of the molecule, these sulfhydryl groups
would be somewhat protected from oxidation and the possible formation of
inter-molecular disulfide bonds.
Because it has been reported that phosphatase activity increases in the
early pregnant uterus,293 crude uterine and allantoic fluids were tested
for the ability to hydrolyze pNPP. No activity was found in tris-buffer
at pH 8.2, but there was indeed phosphatase activity at pH 6.0 in ammonium
acetate. Testing the basic proteins showed activity in the purple frac-
tions but not in the eluent corresponding to Fraction V. Samples of PP
10 20 30 40 50 60
Figure I-8. Sulfhydryl titration. Aliquots of guanidine-
treated PP (at 216,uM) were added to 1.00 n1 of 78.0pM pMB
as described in the methods section. The absorbance at 250) am
at pH 7 was recorded against PP added. The curve indicates a
stoichiometry of 9 (f 1) reacting sulfhydryl groups on the protein.
having the highest purity as determined by gel electrophoresis were found
to have the greatest activity, The effect of the concentration of PP
on the rate of hydrolysis at pH 6.0 is shown in Figure I-9, and indicates
a linear relationship. This experiment was later repeated at 300, pH
4.9. The initial rate of reaction increased linearly over a 10-fold PP
Effect of ionic strength
The effect of ionic strength on the rate of reaction at pH 6.0 was
studied using sodium chloride and potassium chloride. The concentrations
of these salts in the reaction mixture were varied from 0 M to 2 M. The
results are shown in Figure I-10. Increasing salt concentration decreases
enzyme activity whereas some phosphatases have been found to be unaffected
or stimulated by salts. Also, certain phosphatases have been found to be
specifically affected by either Na+ or K ; this is not the case with PP.
PP functions as an acid rather than an alkaline phosphatase. However,
it was desirable to know more accurately the pH at which maximum activity
occurs. 300 was selected as the temperature at' which to run the reaction
with 12 mY pNPP. The following were used to prepare buffers of 0.1 M concen-
tration: citrate, acetate, maleate and imidazole. The pH range examined
was from 3 to 8. 5 minutes after enzyme was added to the reaction mixture
at 300, KOH was added and the pNP liberated was determined, Figure I-11
shows the pH optimum to be at 4.9 in 0.1 M acetate buffer.
5 10 15 20)
Figure I-9. Linearity of reaction with enzyme concentration.
Protein was added to 12 mM pNPP in 0.1 M NH4Ac, pH 6, at 22".
The absorbance at 405 nm was recorded versus time with a Beckman
Acta cIll. The linear portion of the trace was used to determine
initial reaction velocity.
Figure I-10. Effect of ionic strength. The
reaction with pNPP was run at room temperature in
0.1 M NH4Ae at pH 6.0. Activity was determined as
pNP liberated. O : NaCl; AS: KC1.
0.2 0.4 0,6 0.8 1.0
MOLARITY OF SALT
1,2 1.4 1.6 1.8
Figure I-11. pH profile. The reaction was run in 0.1 M
buffer using pNPP as substrate. The reaction was stopped with
KOH and the absorbance at 410 nm was recorded. Buffers were:
citrate; acetate; o maleate; b imidazole-HCL.
Product formation versus time
The linearity of product formation versus time was tested with enzyme
solutions of varying specific activities. Generally, 5 ml of 12 mM pNPP
in acetate at pH 4.9 was incubated in the presence of enzyme at 30o. 0.5
m1 aliquots were withdrawn periodically and added to a known volume of
KOH to stop the reaction and develop the color. The results of one trial
is shown in Figure I-12 for both 12 mM and 1.2 mM pNPP. It was found
that for solutions of 12 mM pNPP, the reaction would be linear up to about
10 minutes if the amount of enzyme added produced an optical density of
less than 1.5 0.D. at 410 nm in a final volume of 6 ml (1 ml reaction
mixture plus 5 ml of KOH). This corresponds to about 0.5 mM pNP produced
in the original reaction mixture, i.e., hydrolysis of 4% of the substrate.
A quantitative theory on enzyme kinetics was first developed by Henri,
This theory now bears the names of Michaelis and Menten who extended Henri's
work and devised graphical methods for evaluating the constants Km and
Vmax. The theory assumed rapid formation of an intermediate complex which
would break down in the rate-limiting step to form product.31
E+ S ES
ES ----E + P
Here E, S, P and ES stand for enzyme, substrate, product and the enzyme-
substrate complex, The equilibrium position of the first reaction was
believed not to be significantly disturbed by breakdown of ES. However,
Figure I-12. Linearity of reaction with time.
The reaction was run at 30* in 0.1 M acetate at pH 4.9
using pNPP as substrate, Aliquots were taken at various
time intervals and analyzed for product (see text).
*: 1.2 mM pNPP, 1.22 yg PP/ml; o: 12 mM pNPP; 0.61
REACTION TIME (MIN)
Briggs and Haldane showed that this assumption was unnecessary, that
a steady state could be attained in which the change of [ES] with time
kl[E][S] = k2[ES]+kg[ES]
kl[E][S]-(k2+ k3)[ES] = 0.
In studying enzyme reactions, the molar concentration of substrate
is generally very much greater than the concentration of enzyme. As a
result the amount of S bound by E at any given time is negligible com-
pared to [S], the total concentration of substrate. The total concen-
tration of enzyme, [Eo] is equal to that of the free enzyme, [E], plus
the concentration of complex, [ES], i.e.,
[Eol = [E] + [ES],
Solving this equation for [E], substituting into equation (2) and
dropping the bracket notation for concentration gives
kl(S)(Eo ES) (k2 + k3)(ES) = 0.
Solving for ES gives
ES = kl(Eo)(S) .(3)
The rate of reaction is then
v = g(E) =klk3(Ea)(S)
Since S is the initial concentration of substrate, v is the initial
SKm + (S)
Km = k2+k3 (5)
Km is referred to as the Michaelis constant. When the substrate
concentration is much smaller than Km, S may be neglected in the denomi-
nator of (3),
v = (Eo) (S).
The kinetics are first order in substrate concentration. When S is
much greater than Km, equation (4) becomes
v = k3 Eo = Vmax (6)
and the kinetics are independent of substrate concentration. A plot
of rate versus substrate concentration at constant Eo should be linear
for low substrate concentrations and asymptotically approach Vmax at
high substrate concentrations. Substituting the right side of equation
(6) into equation (4) gives
Km + S
The condition for half-maximal velocity, i.e., v = 1/2*Vmax is readily
seen by solving equation (7) and is
S = m.
Km is thus the concentration of substrate necessary to half-saturate
the enzyme and is a useful property in the characterization of enzymes.
Km has additional significance in two special Cases. If k2>>k3 in
equation (4), then Km = k2/k1, the enzyme-substrate dissociation con-
stant. If k2<
Km and Vmax are not fixed values. Both vary with substrate, pH
and temperature. The type of plot mentioned above is not convenient
to use in the determination of Km as it requires an extrapolation to
Vmax. If one takes the reciprocal of equation (7), one obtains the
1/v = (Km/Vmax*1/S) + 1/Vmax. (8)
Plotting 1/v versus 1/S results in a straight line with slope of Km/Vmax,
x-intercept of -1/Km and y-intercept of 1/Vmax. Putting the experimental
data in this form makes Km and Vmax readily accessible providing Michaelis-
Menten kinetics are obeyed. In this work, Km was determined from a least
squares analysis of the data.
The basic assumptions in the formulation of equation (4) are as
follows.32 (i) There is an intermediary complex of enzyme and substrate.
Though this seems quite reasonable, not all workers studying enzyme cataly-
sis in the early days accepted this postulate. (ii) One substrate is
bound at one catalytic site. If this is not true the rate expression in
equation (4) will involve powers of S and the reciprocal plot of 1/v versus
1/S will not be linear. Of course, one protein may have more than one
active site. (iii) The free substrate concentration is equal to the
total amount of substrate added since Eo<
allows a further observation. The derivative of equation (4) can be taken
with respect to Eo. This gives (dv/dEo)S = kg*S/(Km + S), a constant,
which means v is proportional to enzyme concentration. This is important
as it allows one to vary the initial enzyme concentration during measure-
ment of initial rates. (iv) The overall reaction is irreversible.
Enzymes are known to catalyze reactions in both the forward and reverse
directions; there is no change in the value of the equilibrium constant.
However, if the measurements of the rate are made early in the reaction
while the concentrations of products are low, the assumption of irrever-
sibility is justified.
Km was determined in 0.1 M acetate, pH 4.9, using concentrations
between 1.2 mM and 12 mM pNPP. From five determinations Km was calculated
to be 2.2 mM (f 0.3 mM) at 300 and was little changed at 00, namely,
2.8 mM for three determinations. Compared to other phosphatases, these
values are somewhat high. Under optimum conditions and using the acti-
vating agents to be described later, the greatest specific activity
obtained was 170 nanomoles/min/,ag. This corresponds to a turnover num-
ber of 5400 molecules of substrate per minute per molecule of enzyme. The
graphs of one experiment are shown in Figures I-13 and I-14.
The location of PP in vivo implies a possible function of the enzyme
to be that involving general hydrolysis. PP could be supposed to be
active in some phase of the catabolism of substances found in the allan-
toic fluid. With this in mind, a number of molecules were tested as
substrates at pH 4.9. As can be seen from Table I-2, no esterase, sulfa-
tase or phasphodiesterase activity was found. The phosphate groups of
the protein phosvitin were hydrolyzed only poorly. PP was then assayed
for the ability to hydralyze a number of phosphate-containing molecules
of biological importance. Of the phosphomonoesters and anhydrides
tested, the only appreciable activity was found towards ATP and pyro-
phosphate. Thus PP is most active towards molecules containing high-
energy phosphate bonds.
The specific activity changed very little when the enzyme was stored
at 40 at concentrations of about 1 ag/m1. However, when the enzyme was
diluted about 100 times for use in assays, the liability increased. The
denaturation of diluted enzyme when stored at 40 in 0.1 M acetate at pH
4.9 is shown in Figure I-15. After 20 days there is still about 20%
activity remaining. 2-ME, described later as an activating agent,
0 2 4 6 8 10 12
Figure I-13. Reaction velocity versus substrate concentration.
The initial concentration of pNPP was varied and the concentration
of enzyme was held constant (2.07yg). The reaction was carried out
at 30" in 0.1 M acetate, pH 4.9.
0 2 4 6 8 10
Figure I-14. Lineweaver-Burk plot of the data of Figure I-13.
A least squares analysis gives a Km of 1.6 mM.
Table I-2, Hydroytic activity of PP towards various esters.a
Substrate (nmoles Pi/5min)
Sodium pyrophosphate 58
pyridoxal phosphate 9
p-nitrophenyl sulfate 0
aReactions carried out at 5 mM substrate concentration using
2 118 enzyme. Buffer was 0.1 M acetate, pH 4.9 containing 0.1
M NaC1. Phosvitin was approximately 5 mM with respect to
2 4 6 8 10 12 14 16 18 20 22 24
DAYS AT 4o
Figure I-15, Denaturation of diluted PP. The protein was diluted
to 6.1;ug/ml and stored at 4o. Aliquots were withdrawn periodically
and the specific activity towards pNPP was determined at pH 4.9, 300
could not increase the activity to that found at the start of the
experiment. The rate at which the enzyme lost activity could be signif-
icantly increased by storing the enzyme in acetate buffer containing
0.5 M NaC1. To test the stability of PP at increased temperatures, the
enzymle was preincubated in sealed test tubes in water baths at 300, 500
and 700. At the higher temperatures, the enzyme solutions were cooled
in an ice-water bath and then a few minutes were allowed for the tempera-
ture to stabilize at 300, at which temperature the reaction was run.
Increasing the preincubation temperature and the length of exposure
resulted in decreased activity, shown in Figure I-16. The diluted
enzyme is relatively stable to heat; there is still 12% residual activ-
ity after 30 minutes at 700.
It is known that proteins in dilute solution become denatured at
surfaces. Various surface active agents have been used to retard
denaturation; these compete with the protein for the available surface
area. One such surfactant used successfully in the protection of acid
phosphatases is Triton X100.3 '84 A room temperature enzyme solution
containing 0.005% Triton in acetate buffer was assayed at pH 4.9 and
compared to an identical solution which lacked the Triton. The former
was found to lose activity at twice the rate o'f the control. Possibly
the detergent caused a loosening of the protein structure. No other
surface active agents were examined.
Energy of activation of the catalytic reaction
The temperature variation of a chemical reaction follows the equa-
tion of Arrhenius
a In k/3 (1/T) = -Ea/R
5 10 15 20 25 30
PREINCUBATION TIME (MIN)
Figure 1-16. Heat stability, The enzyme was preincubated
at elevated temperatures prior to assay with pNPP at pH 4.9, 300
* : 300; a : 50"; 0: 70D.
where Ea is a constant, the energy of activation. Ea will depend upon
the activation energies of the elementary reactions which are given in
equation (1). In the case of equation (1) a plot of In kg against 1/T
would give a straight line, the slope of which is -Ea/R. If ES is kept
constant as the temperature is varied, then v is proportional to k3 and
Ea could be determined from a plot of In v versus 1/T. If the concentra-
tion of substrate is kept constant but very large, then ES will approxi-
mate Eo, the total concentration of enzyme, and thus ES can be taken to be
constant. If S is not large, then ES will equal (Eo)(S)/(Km + S) as seen
in equation (3). Experimentally it is easier to work at constant substrate
rather than constant ES. The error involved has been determined by Gibson35
and found to be
Ea Ea = Km/(S+Km) R a In Km/3 (1/T) (9)
Ea is the apparent activation energy and the term containing the partial
derivative is equal to (1/R) AHm, the change in heat content accompanying
the formation of ES, provided Michaelis-Menten kinetics apply. In the
theory of Briggs and Haldane, the formation of a pre-equilibrium is not
required. Therefore Km is not an equilibrium constant unkess k2>>ks.
In any case one can find the value of the partial derivative by using
the experimental values of Km at 00 and 300 given before. The right-hand
side of equation (9) becomes +1.3 kcal'Km/(S + Km). The concentration of
substrate used in the activation energy determination was 40 mM. Thus the
value of Km/(S + Km) will be about 0.05. Equation (9) reduces to
Ea Ea = (0.05)*(1.3kcal) = 0.065 kcal .
The difference between the apparent and true activation energies is found
to be less than experimental error under the conditions of the assay.
The apparent activation energy was determined at pH 4.9 in 0.1 M
acetate buffer, 0.1 M NaC1. pNP liberated was measured after the reac-
tion had proceeded for 5 minutes at the desired temperature. The natural
log of the absorbance at 410 nm in the undiluted reaction volume was
plotted against 103/T, where the temperature is on the Kelvin scale.
The plot was linear as seen in Figure I-17 and the apparent activation
energy was found from a least squares analysis to be 11.1 kcal/mole. This
value can be compared to the energies obtained for alkaline phosphatase
from bone,36 9.2 kcal, and from P. laevis,37 4.9 kcal and for acid phospha-
tase from T. confusum,3 13.0 kcal, and from P. laevis,39 8.2 kcal. How-
ever, it seems likely that some of these experimental values may contain
the errors mentioned above.
A number of metal ions and polyanions were tested as inhibitors of
the hydrolysis of pNPP at pH 4.9. The results are shown in Table I-3.
Molybdate was found to be extremely inhibitory and, to a lesser degree,
arsenate and phosphate. The type of kinetic inhibition was not studied.
pNP was also tested and found to show no effect when at 1 mM. Of the
metal ions tested, mercury(II) showed the greatest effect. pME and zinc
were also inhibitory.
Several experiments were carried out using cysteine or 2-ME to reacti-
vate the Hg+2 inhibited PP. The concentration of mercury used was 1uM.
Enzyme with and without mercury was preincubated at room temperature for
10 minutes in acetate at pH 4.9. To one series of test tubes, 2-ME in
buffer was added to give a concentration of 10 pM; the control series
was buffer without the sulfhydry1. After an additional 10 minutes of
Figure I-17. Energy of activation.
The apparent energy of activation (Ea*)
was determined as described in the text
with 40 mM pNPP in 0.1 M acetate 0.1 M
NaC1 at pH 4.9. The reaction was stopped
by the addition of 5 ml of 0.25 N KOH and
the absorbance at 410 nm recorded. The
slope of the curve gives Ea* = +11.1 kcal/mole.
1. 2 O
1 I I
3.2 3.3 3.4 3.5 3.6 3.7
103/T ( K)-1
a Reactions were carried out using 12 mM p-nitrophenylphosphate
in 0.1 M acetate buffer, pH 4.9, for 5 minutes. Each compound
was inlcubated with the enzyme for 10 minutes at room temperature
before addition of substrate Phosphatase activity was measured
as p-nitrophenot released.
Table I-3. Rate effectors of pNPPase activity
Added Concentration Activity
Compound (mM) (% Control)
preincubation, pNPP was added. Repeating this experiment several times
gave the same results: that the low concentration of 2-NE used did not
affect the specific activity compared to that of the control containing
no 2-ME; that while Hg+2 was a powerful inhibitor, a 10-fold molar excess
of 2-ME over metal completely restored activity with respect to the control.
The same results were obtained with 10 VM cysteine and 10 I1M ascorbic
Enzymes which require the presence of an SH group in or near the
active site, the so called sulfhydryl enzymes, are often found to be
inhibited by the mercuric ion through mercaptide formation. Also, reacti-
vation can occur by adding excess sulfhydryl which competes for the bound
mercury. Presumably ascorbate can also interact in the same manner.
Since the effect of Hg+2 is not specific for sulfhydry1 groups, it can
not be assumed from the data that the protein requires a thio1 group for
catalysis. Webb4 has cautioned that different sulfhydryl reagents will
have varying degrees of inhibition on sulfhydry1 enzymes. One particular
reagent may completely inhibit one enzyme at a given concentration and
yet have little effect on another. The slightly stimulating effects,
seen in the table, of EDTA and bipyridine on PP activity may be due to
chelation of potentially inhibitory metals present in the reaction solution.
A third point which is illustrated by Table I-3 is the lack of
significant inhibition by ATP, PPi and pyridoxal phosphate on the hydrolysis
of pNPP as measured by the determination of pNP. The concentration of
each was 1 mM compared to 12 mM pNPP. Apparently these compete only
poorly with pNPP for the active site of the enzyme. Km determinations
were not carried out for these substrates.
The final information in Table I-3 concerns the effect of the
halide ions. It was found that neither sodium nor potassium ions had a
specific effect on phosphatase activity; therefore, the sodium salts
were prepared in buffered solutions. Chloride, bromide and iodide are
seen to have little effect, yet fluoride is a strong inhibitor. A concen-
tration of 0.1 mM fluoride results in 67% activity remaining.
Effect of oxidants and reductants
Because oxidizing and reducing agents were found to have a decided
effect on the absorption spectrum of PP, these were tested as to their
influence on the rate of hydrolysis of pNPP. Various concentrations of
oxidant or reductant were preincubated with the enzyme for 10 minutes;
the reaction was then run at 300, pH 4.9 as usual. The results of one
such series using enzyme which had maximum visible absorption at 545 nm
are shown in Table I-4. Reducing agents such as 2-ME and ascorbate were
found to give a pronounced increase in enzymatic activity. Dithionite,
a powerful reducing agent, also increased activity at low concentrations
but when present at, or greater than 10 mM, completely inhibited hydrolysis.
This would correspond to the concentrations of dithionite needed to bleach
the purple color of the protein. 1 mM hydrogen peroxide had an inhibitory
effect on the reaction. Figure I-18 is a plot of activity versus concen-
tration of 2-ME following a 10 minute preincubation. Increasing the
concentration of 2-ME increases activity, but the slope of the curve is
found to be greatest at low concentrations. Two Km determinations with
pNP gave an average value of 2.2 mM in the presence of 0.1 M 2-ME. Thus,
2-ME affects the rate but not the binding of substrate. It was found
later that an excess of 2-ME, ascorbate and dithionite causes loss of
Table I-4. Effect of reductants and oxidants on the pNPPase
activity of the purple form of PP.a
Added Concentration Activity
Species (mM) (% Control)
Na dithionite 0.1 117
Na dithionite 1 127
Na dithionite 10 0
Na dithionite 100 0
2-ME 1 132
2-ME 20 171
2-ME 120 222
Ascorbate 100 245
H202 1 28
a Reaction conditions specified in Table I-3.
.02 0.4 .06 .08 .10 .12 .14 .16 .18 .20
Figure I-18. Effect of 2-ME on the reaction velocity.
The enzyme was preincubated with 2-MIE for 10 minutes atoroom
temperature. The assay was carried out with pNPP at 30 pH 4.9.
iron from the protein with concomitant loss of activity. This will be
discussed in more detail later. Mild reducing agents had either no
effect or a slightly deactivating influence on the pink form (512 nm)
of PP. As with the purple form, pink PP was completely inhibited by
10 mM dithionite. 1 mM hydrogen peroxide left only 2% residual activity.
The effect of 2-ME and ascorbate was found to be variable and depen-
dent not only on the color of the enzyme but on the age as well. As
mentioned before, preparations of diluted enzyme which were allowed to
denature could not be completely reactivated by 2-ME.
Preparation of ApoPP
In the spectral studies of PP, dithionite in sufficient concentra-
tion causes rapid bleaching of the purple color. Also found was a
definite effect of dithionite upon enzymatic activity; as the concentra-
tion of dithionite was increased, the activity first increased, as with
all reducing agents tried, and then decreased to zero. Based on these
observations, the following experiment was attempted in order to deter-
mine if these conditions cause loss of iron, or at least decreased
affinity of the protein for the iron. Dithionite was added to a solution
of PP which was 1 mM in bipyridine and the solution quickly became pink,
the color of the ferrous-bipyridine complex. The loss of iron, as deter-
mined by the pink color, took on the order of 10 seconds, about the same
time required for dithionite to bleach the color of PP in solution alone.
Removal of iron
A 3.5 ml solution of 55 MM PP with A280/A545 = 13.4 was transferred
into a dialysis bag. The bag was sealed and then inverted several times
to mix protein solution with any residual water left in the bag. The
bag was cut in half, with about equal volumes of solution in each half,
and resealed. This careful procedure was used to insure that each bag
would contain solutions of equal protein concentration. One bag was
placed in 100 ml of 0.1 M acetate, pH 5.0, the other in 100 ml of buffer
and 50 mM dithionite. Both contained 1 mM bipyridine. Solutions were
made from glass-distilled water and all dialysis was done in the cold.
Within 5 minutes the contents of the dithionite treated bag began to
turn pink. After 3.5 hours with one change of the respective solutions,
the dialysate was changed to buffer plus bipyridine in both cases. After
2 hours the dialysate was changed to buffer only. After 5.5 hours and
one additional change of buffer, the optical spectrum of each solution
was taken. The control had A280/AS45 = 13.4 as before; there was no
iron lost. The treated protein had no distinct absorption in the visible
region and no 320 nm peak. The specific activities of the two solutions
were measured as usual at pH 5 using pNPP as substrate, and found to be
0.3 and 36 for treated and control. The assay was repeated after 10
minutes preincubation in the presence of 0.1 M 2-ME for 10 minutes. The
specific activities were found to be 0.5 and 108. The iron content of
each was determined by the TPTZ method. The control was found to contain
1.08 male Fe per male protein while the dithiodlite treated sample had
0.05 mole Fe per mole protein.
The remainder of each solution was dialyzed against 1 mM ferric
chloride in acetate for 4 hours and then against buffer only to remove
excess iron. Assays gave the specific activities to be 21 and 7 for
treated and control respectively, and 20 and 66 for samples praincubated
for 10 minutes in 0.1 M 2-ME. The decrease in activity of the control
is most probably due to oxidation by ferric iron. The 9-fold increase
in activity upon treatment with 2-ME (the greatest increase yet ob-
served) shows that oxidation is reversible if the protein is not
allowed to age. That is, the oxidation of a critical group or groups
may be followed by a slow irreversible denaturation. However, 2-ME
had little effect on the treated sample. There was not enough of
either sample to obtain visible spectra or to do an iron analysis.
An experiment was undertaken as above but 50 mM ascorbic acid in
acetate, pH 5, was used in place of dithionite. Again the contents of
the ascorbate treated bag turned pink. After 4 hours the dialysate
was changed to acetate buffer and dialysis continued to remove all ascor-
bate. The optical spectrum of each sample was recorded. The treated
was found to have only half of the visible absorption as the control
and decreased 320 nm shoulder. The specific activities towards pNPP
at pH 5 were found to be each 33. However, when both were preincubated
in 0.1 M 2-ME for 10 minutes the activity of the control had increased
about three times while the ascorbate treated sample was unchanged.
Thus it appears in this case that the ascorbate treated protein is
already fully "reduced", insofar as the required groups are concerned,
as it would appear to contain only half the iron of the control, seen
from the visible absorbance, but still has the same specific activity.
An iron determination was not carried out however.
Reconstitution of apoPP
Enzymatic reconstitution. Three experiments were carried out to
remove the iron from PP and determine reconstitutability enzymatically.
Two involved using ge1 filtration and one used dialysis as means of
separating the liberated iron and excess dithionite from the protein. In
the first method, either 50 mM dithionite, in one case, or 50 mM dithionite
plus 1 mM bipyridine in the other, was added to solutions of PP. After
allowing 15 minutes at Oo, the treated solutions of PP were passed through
Sephadex G-10 column (1.5 x 15 cm) equilibrated in acetate buffer at room
temperature. Fractions were collected and scanned in the UV region.
There was good separation of the smaller molecules from the protein.
Fractions of apoPP were pooled and aliquots were added to pH 5 buffered
solutions which were 8 IIM in one of the following metals: Mg+2 Mln Z,
Cr+3,Fe+3, Co+ or Cu+2 The protein concentration at this point was
1.2 yIM. The solutions were allowed to stand at 40 for 20 minutes after the
addition of protein;then the mixtures were assayed with pNPP at pH 5.
The results are shown in Table I-5. The control used was 1.2 yM apoPP
in metal-free buffer. The table shows that the ferric and cupric ions
give significantly greater activity over the control, while the other
ions show little effect. The order of the entries in the table is the
chronological order in which the assays were done. As additional time
is allowed for the ferric or cupric ions to bind to the apoprotein, more
activity is found. The second column experiment investigated the change
in specific activity versus time for apoPP plus either iron or copper.
The results were similar to the third experiment which is described
A solution of PP was dialyzed in the cold against 50 mM dithionite
and 1 mM bipyridine in acetate, pH 5. Changes of buffer during 23 hours
insured removal of the iron-bipyridine complex and excess dithionite.
Aliquots of apoPP were added to solutions of either 9-fold molar excess
Table I-5. Reconstitution of pNPPase activity by addition of
metals to apoPP,
Metal Presenta Specific Activity
none 3, 6b, 10c
Fe+ (chloride) 1,3b 3
Cr+ (K-sulfate) 3
Co+ (acetate) 4
Mn+ (chloride) 4
Mg+2 (sulfate) 5
a apoPP was added to 40 solutions of 8.0,aM in metal (a 7-fold molar
excess) buffered in 0.1 M acetate at pH 4.9. Twenty minutes after mixing
these solutions were assayed for activity towards pNPP. The order of the
table is the order in which the assays were done,
bsaoebtteasywsrpetd9 iue fe iig
SAs above but the assay was repeated 90 inutes after mixing,
Cu+2, 8-fold excess Fe+3 or to metal-free buffer which was used as the
control. The mixtures were at room temperature and aliquots were with-
drawn periodically and assayed with pNPP at pH 5. A plot of specific
activity (based on total protein present) versus time after addition of
enzyme is shown in Figure I-19. As stated above, these results are
essentially the same as for the second G10 column experiment. The activity
of the copper enzyme reaches a maximum after about one hour past mixing.
The activity of reconstituted Fe+3-PP reaches a maximum after about two
hours and in two of the three experiments had a specific activity of about
twice that of Cu+2-PP. The control shows slow restoration of activity;
this is no doubt due to low levels of contaminating iron found in the
reagents used to prepare the buffer. The protein is present at approxi-
mately 1 pM. Thus 1 nanomole Fe/ml of solution would provide the stoichio-
metric amount of metal required for reconstitution. If the copper- or
iron-restored protein is stored at 40, the loss of activity is only 15%
and 10% respectively over 16 hours. This is about the same rate of decay
as found for native protein which had been diluted as discussed previously.
The effect of a 10 minute preincubation with 0.1 M 2-ME was studied on
the reconstituted iron and copper enzymes. No increase in activity was
Spectroscopic reconstitution. ApoPP was prepared by treatment of about
1.4 mg protein with dithionite and bipyridine, followed by dialysis as
before. The apo-enzyme was scanned and found to have no definite visible
absorption bands. A 10-fold molar excess of ferric chloride was added to
both the sample cuvette containing apoPP and the reference cuvette contain-
ing the dialysis buffer; this would compensate for iron absorption in the
Figure I-19. Restoration of enzymatic activity
by addition of copper and iron to apoPP. apoPP was
prepared as described in the text and then added to
room temperature solutions of 0.1 M acetate at pH 5
containing no metal (0), 9-fold molarexcess Fe 3 (1)
or 8-fold molar excess Cu+ (). Aliquots of each
solution were withdrawn at the indicated time intervals
and assayed with pNPP in acetate at 300. The specific
activity calculated was based on total protein present.
20 40 60 80 100
blue and UV portion of the spectrum. The mixture was scanned periodi-
cally. After 8 minutes, the absorption at 540 nm was about 2/3 that of the
maximum obtained after 40 minutes. The spectrum was identical to that
of naturally occurring PP. The mixture was dialyzed overnight to remove
excess iron and the spectrum then recorded. The spectrum was found to
be virtually unchanged.
ApoPP was prepared as described in the above paragraph and the spec-
trum recorded. Aliquots from a stock solution of copper sulfate were
added to reference and sample cuvette to give a 4-fold molar excess of
Cu+2 After 12 minutes the spectrum was taken. No increase had occurred
in the visible region but there was an increase in the 280 nm peak of 4%.
Then Fe+ was added to a 6-fold excess and slow increase in absorbance at
550 nm was followed versus time. After 90 minutes the spectrum was recorded.
The visible region contains the same features as native PP.
This experiment was repeated using a 4-fold excess of Cu+ or Fe+
ApoPP plus Fe+3 again gave the results as obtained in the previously de-
scribed reconstitution experiments. The increase in absorbance at 550
nm was 0.024 0.D., 19 minutes after addition of' Fe+3 The addition of
Cu+ to apoPP again gave only an increase in the UV region but the change
occurred within about 2 minutes. 4-fold excess Fe was then added to
the Cu+ plus apoPP solution. After 22 minutes the spectrum was taken.
The increase in A550 was found to be 0.017 0.D. From these results it
would seem that the presence of copper inhibits the binding of iron.
Whether this is because the copper binds in or near the iron binding site
or acts to distort this site cannot be determined from this preliminary
study. The increase in the 280 nm absorption which occurs after addi-
tion of copper could be caused either by amino acid copper interaction,
or by a change in the environment in one or more of the aromatic amino
acid residues, resulting from an alternation in peptide conformation.
PP has been found to be an acid phosphatase possessing a rather
narrow range of hydrolytic ability. It is produced in the uterus and later
found in the allantoic fluid of the young embryo. It is inducible by
progesterone and can be obtained in relatively large amounts. Further,
the protein is at its highest level during a critical phase in pregnancy
when embryonic mortality is high. Due to these facts, and because PP
is most active towards the biologically unimportant molecule p-nitrophenyl-
phosphate, it seems likely that the true physiological function has not
yet been discovered. Perhaps PP is important in maintaining a suitable
environment for embryonic development. It is possible that the chela-
tion of free iron is its function; however, all PP found contained, within
experimental error, 1 mole of iron per 32,000 molecular weight. It would
seem that if binding iron is the role of PP, then an appreciable amount
of apoPP should be found.
Reducing agents were found to have an activating influence on the
enzyme when present in proper concentrations. 0.1 M 2-ME was found to
increase the rate of hydrolysis of pNPP by 2 to 4 times but the value of
Km at 300, pH 4.9, was unchanged. Treatment of PP with reducing agents
of sufficient concentration or prolonged exposure resulted in loss of
activity and loss of iron. It was also found that very low concen-
trations of Hg+2 had a potent inhibitory effect upon the reaction.
Coupling these data, it seems quite possible that a free thio1 group
is required for catalysis. Oxidation either by air or hydrogen perox-
ide would generate the formation of a disulfide bond if a second sulf-
hydryl group were in close proximity. This would impose a slight con-
straint on the conformation of the protein and perhaps affect the
environment of the iron enough to cause a spectral shift to around
550 nm and the appearance of the 320 nm peak (certainly due to the iron)
seen in the spectrum as a shoulder. Addition of such as 2-ME would
reduce the disulfide to two free thiols, thus the shift of visible
absorption maximum to the blue as the ligands of the iron rearrange to
the native conformation. It should be recalled that reconstituted PP
prepared from lengthy exposure to reductants did not show increased
activity toward pNPP after preincubation with 2-ME. From this, one
might have expected the enzyme to be restored, by addition of iron, to
the pink form. At the low concentration of apoPP used for spectral
restoration it was difficult to identify the peak maximum precisely,
However, the maximum was near 540 nm, which is close to the purple
extreme found around 550 nm. Perhaps the liability of apoPP is greater
than PP and although iron can be added to give color, a high percentage
of protein is no longer capable of enzymatic activity. 'Two types of
iron-containing protein may be formed: the pink species which is a
minority component and another species which absorbs around 540 nm, but
has become irreversibly denatured. Clearly more work needs to be done
on this intriguing problem.
From experiments involving the reactions of acid and alkaline
phosphatases in H2180, it has been found that the hydrolysis is via
cleavage of the 0-P bond in C-0-P. 1 One could postulate a mechanism
of hydrolysis involving the formation of an intermediate sulfur-
phosphorous bond as molecules of the type ROP(S)(SH)2 are known.4
The observed stimulatory effect of reducing agents under this theory
would be due to the liberation of the essential sulfhydry1 group via
reduction of the disulfide bond. Thus 2-ME is seen to increase the
number of catalytically active proteins in solution rather than increas-
ing the efficiency of the reaction. The variable effect found, even
with a constant concentration of 2-ME and constant preincubation time,
is due to different protein solutions having different concentrations
of "oxidized" enzyme.
Prolonged exposure to 0.1 M 2-ME or ascorbate or 10 mM dithionite
results in reduction of the iron. One cannot be certain which event
occurs first: conversion to the ferrous ion, or release of iron from
the protein, due to some other effect, then subsequent reduction. The
metal, required for catalysis, could simply be a site of attraction for
one or two of the oxygen atoms of the phosphate group. This could serve
to orient the substrate into the proper configuration for catalysis to
occur. Assuredly, more information needs to be gathered on the effects
of sulfhydry1 reagents on PP.
Nature of Glycoproteins
Though this work has not dealt with the carbohydrate groups of the
glycoprotein, a few points can be considered here. Two recent reviews
on glycoproteins were found to be helpful.43'44 Both have demonstrated
that while the function of a molecule as a whole is understood, the
role of the carbohydrate portion is often not known at all. For
example, RNAse B, a glycoprotein, has the same enzymatic activity as
RNAse A, but the latter contains no sugar groups. There are some
general concepts pertaining to the carbohydrate moieties which will
be briefly mentioned, in that one or more of these may be important
in regard to PP. The carbohydrate group may increase the resistance
of the protein to hydrolysis by proteolytic enzymes, increasing its
life in vivo. These groups also increase protein solubility due to
their strong hydration.
Carbohydrates are important in two facets of recognition. It has
been found that if one type of terminal sugar unit has been selectively
removed from a glycoprotein which is then injected into test animals,
these altered proteins will be removed from the serum, whereas the
untreated control will not be removed.4 Secondly, carbohydrates may
be involved in recognition and the transport of the protein into the
cell. It is known that sugars are abundant on the surface of cell mem-
branes. It has been proposed that these are instrumental in inter-
cellular communication, so called contact inhibition, and in recognition
of large molecules. A final interesting finding is that a particular
glycoprotein, C'9, has been found to chelate iron; the bonds formed are
believed to be from carbohydrate groups.45
ESR studies of iron in glass gave a signal at g = 4.27 which was
analyzed by Castner et al. 6 and found to be due to the high spin ferric
ion. Since the ground state of this atom is uS, one would expect a line
at g = 2 with perhaps some broadening due to crystal fields within
the glass. These workers examined the spin Hamiltonian in regard to
D, the coefficient of the axial part of the Hamiltonian, and E, the
coefficient of the rhombic part. By considering D = 0, they were
able to predict a splitting corresponding to g = 30/7 for each of the
three principal axes for the middle Kramers doublet. The g values of
the upper and lower doublets were found to be strongly anisotropic.
In a detailed study, Blumberg 7 showed that for IE/DI = 0, i.e., axial
symmetry, then an increase in this ratio represents an increase in
rhombicity. Further, JE/DJ = 1/3 represents a completely rbombic field
and the energy separation between the three Kramers doublets is equal
and is 4r7 D/3. Also shown in this work were plots of the ratio of
microwave frequency/magnetic field in units of the Bohr magneton/
Planck's constant versus magnetic field. D was held constant at 0.75
wavenumbers and E was varied. The results showed that the effective
g values departed from 30/7 as the strength of the magnetic field
increased and as the rhombicity decreased. Dowsing and Gibson4 have
also investigated high-spin d5 systems. They have constructed six
graphs of D/hy versus H/hy and D versus H for E/D ranging from 0 to 1/3.
They conclude that for E/D near 1/3 and D greater than 0.23 wavenumbers,
absorption would be around 1500 gauss (v = 9.3 GHz). More precise
information on the values of D and E could be obtained from the experi-
mental g values if a complete set of these graphs were available. The
only conclusions that can be drawn in regard to the iron in PP is that
the symrmetry of the paramagnetic site is highly rhombic and one has a
lower limit on the value of D from the aforementioned graphs.
ESR spectra of a number of nonheme iron proteins have shown the
g = 4.3 absorption. A detailed analysis of the ESR spectrum of rubre-
doxin from P. oleovorans was undertaken by Peisach et al.4 at temper-
atures below 120 K; signals arising from the lower Kramers doublet
could be observed; as predicated from theory, they were markedly
anisotropic. The signal intensity of the observed transitions was
measured over a range of temperature from 1.4 to 400 K. The data were
fit to a Boltzmann distribution over the three Kramers doublets. From
this and knowledge of the effective g values, the values of D and E
which gave the best fit of the data on solving the spin Hamiltonian
gave D = 1.76 wavenumbers, E = 0.495 wavenumbers and E/D = 0.28; this
corresponds to 84% rhombicity. With the instruments available, this
type of evaluation could not be undertaken for PP.
In light of the arc emission data, it seems likely that the signal
at g = 2.05 is due to copper(II). In simple theory, the cupric ion
can be thought to be in the center of an Octahedron of ligands. The
d9 system can be considered as an unpaired hole in one of the eg
orbitals. The Jahn-Teller theorem states however, that the two orbi-
tals in the eg set are not degenerate, though the difference in energy
may be small. The hole is restricted to either the dz2 or dx2_ 2 orbi-
tals and the orbital contribution to the magnetism is quenched. Thus,
g values for Cu+2 are expected to be close to free spin. In the first
approximation, the deviation of the g values from free spin will be
related to the ratio of the spin-orbit coupling constant to the crystal
field splitting. Since copper complexes typically absorb light
around 600 nm, one can use 16,000 wavenumbers as a value of A. Hence
one can calculate 811 = 2.4 and gl = 2.1 from the equations given by
V'dnngard.50 More refined treatments require consideration of covalency
between metal and ligands. Characteristic of copper ESR spectra are
the hyperfine splitting due to the isotopes 6Cu and 65Cu, each with
spin of 3/2. Since the magnetic movements of the two nuclei are of
similar magnitude, the hyperfine lines from the two isotopes are
usually not separated.
Inspection of the g = 2.05 signal of PP shows this to be the per-
pendicular-type transition. The intensity of the parallel lines would
be expected to be such that they would be lost in the noise present
in the spectrum. The 6 hyperfine lines can be tentatively explained
in two ways. First, the splitting could be due to one nitrogen (I = 1)
nucleus. If the symmetry about the copper atom deviates slightly from
axial, two closely overlapping transitions would occur near g = 2.05.
Thus one would see two triplets lying fortuitously close to each other.
Such a slightly non-axial spectrum was found at 35 GHz for 63copper-
transferrin-bicarbonate, pH 7.6.3 In the low field parallel line for
65Cu-transferrin,51 three line hfs was found at X-band. Computer simu-
lation matched this experimental line assuming the presence of a single
nitrogen nucleus, splitting 9.5 gauss. The alternative explanation is
that the splitting is from 2 or more nitrogen nuclei. Six line hyper-
fine near g = 2.0, similar to that of the PP spectrum, has been observed
in Cu-transferrin, bicarbonate-free at pH 9.2,51 and Cu-transferrin,52
conditions unspecified. In the former case, 65Cu-transferrin was studied
and computer simulation of the low field parallel peak, which contained
at least seven byperfine lines, agreed well with the observed spectrum
assuming four equivalent nitrogen nuclei with splitting of 12 gauss.
The spectrum of Cu-conalbumin was also similar to that of PP in the
To review a few facts briefly: iron is required for the catalytic
action of the protein; iron can be removed without protein precipita-
tion and added to form an active enzyme having the same spectral absorp-
tion as the native; copper is the only other metal, out of several
tried, which can be added to apoPP with at least partial restoration
of activity. This evidence gives credence to the proposal that copper
is binding at a site the same as, or similar to, the iron binding
site. It is not certain whether the copper ESR signal observed is
due to copper occupying an iron site or copper bound to another portion
of the polypeptide. It is not known if copper is able to successfully
compete with iron for the protein. If it can, then copper may be used
as a spin probe in an attempt to discover some of the ligands of the
iron, especially in regard to nitrogen hyperfine. The data currently
available are not amenable to the type of analysis performed on Cu-
transferrin. However, as increasing amounts of enzyme are purified,
ESR experiments with 65Cu-PP and Mtissbauer of 57Fe-PP will become a
PP, a number of whose properties have been-described in this work,
appears to be a member of an expanding family of iron-containing phos-
phatases. These proteins have in common a purple color and phosphatase
activity. However, they all differ in molecular weight, amino acid
composition, pl, pH maximum and substrate specificity, insofar as these
properties are known. They have been found in plants, animals and
The first was discovered in beef spleensk-58 nearly 20 years ago
and was described as a "phosphoprotein phosphatase" because of its
ability to hydrolyze phosphate groups on the milk protein, casein. The
published spectrum of the basic glycoprotein is virtually identical
to that of the purple form of PP. One papers? reported that no metals
were found to be present from either are emission analysis or ESR
spectroscopy. Almost no further work was carried out on this enzyme
during the past ten years save for a paper in 1973 by Campbell and
Zerner,59 who found the presence of 1 male of iron per mole of protein,
making this the first example of a metallo-acid-phosphatase. Previously
however, solutions containing a repressible acid phosphatase from N.
crassa had been reported to have a purple color, although no metal
analysis was attempted. This basic protein is about 9.5% carbohy-
drate and believed to exist as a dimer. An iron-containing phospha-
tase active toward ATP was d discussed at the Biochemis try/B iophysics
1974 meeting.61 Also a glycoprotein, it has broad visible absorption
at 565 nm, very similar to PP, and an ESR signal at g = 4.3 at 770 K.
The source of enzyme is the red kidney bean.
Thus PP is another example of an iron-containing purple-colored
phosphatase. The fact that the sources of these proteins are highly
diversified should not be overlooked. None of the mentioned enzymes
have a clearly understood reason for existence. Though certainly
important, their physiological significance remains to be discovered.
1. Murray, F. A., Jr., Bazer, F. W., Wallace, H. D., and Warnick, A. C.,
Reprod. 7, 314 (1972).
2. Chen, T. T., Bazer, F. W., Cetorelli, J. J., Pollard, W. E., and
Roberts, R. M., J. Bio1. Chem. 248, 8560 (1973).
3. Vallee, B. L., and Roberts, R. M., unpublished results.
4. Chen, T. T., and Bazer, F. W., J. Animal Sci. 37, 304 (1973).
5. Perry, J. S., and Rowlands, I. W., J. Reprod. Fert. 4, 175 (1972).
6. O'Malley, B. W., and McGuire, W. L., Endocrinology 84, 63 (1969).
7. Krishnan, R. W., and Daniel, J. C., Jr., Science 158, 490 (1967).
8. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.,
J. Biol. Chem. 193, 265 (1951).
9. Fischer, D. S., and Price, D. C., Clin. Chem. 10, 21 (1964),
10. Weil, J. A., and Anderson, J. K., J. Chem. Soc., 5567 (1965).
11. Cassim, J. Y., Yang, J. T., Biochem. 8, 1947 (1967).
12. Allen, C. A., private communication.
13. Fiske, C. H., and SubbaRow, Y., J. Bio1. Chem. 66, 375 (1925).
14. Seddon, B., and Fynn, G. H., Anal. Biochem. 56, 566 (1973).
15. Bessey, 0. A., Lowry, 0. H., Brock, M. J., J. Biol. Chem. 164, 321
16. Boyer, P. D., J. Amer. Chem. Soc. 76, 4331 (1954).
17. Riordan, J. F., Vallee, B. L., Methods in Enzymology (Hirs, C. H. W.,
ed.), Vol. XI, p. 541. Academic Press, New York, 1957.
18. Reisfeld, R. D., Lewis, U. J., and Williams, 0. E., Nature 195, 281
19. White, A., Handler, P., Smith, E. L., Principles of Biochemistry,
5th ed., p. 92. McGraw-Hill, New York, 1973.
20. Gratzer, W. B., Poly-ar-Amino Acids (Fasman, G. D., ed.), p. 177.
Marcel Dekker, New York, 1967.
21. Cotton, F. A., and Wilkinson, G., Advanced Inorganic Chemistry, 3rd
edition, pp. 581-585. Interscience, New York, 1972.
22. Strickland, E. H., Horwitz, J., and Billups, C., Biochem. 8, 3205
23. Beychock, S., Poly-ar-Amino Acids (Fasman, G. D., ed.), p. 293.
Marcel Dekker, New York, 1967.
24. Boyd, D. B., J. Amer. Chem. Soc. 94, 8799 (1972).
25. Boyd, D. B., J. Phys. Chem. 78, 1554 (1974).
26. Roman, L., Mllmstrom, B., Aasa, R., and V'dangard, T., J. Mol. Bio1.
5, 301 (1962).
27. Tsibris, J. C. M., and Woody, R. W., Coord. Chem. Rev. 5, 417 (1970).
28. Orme-Johnson, W. H., Ann. Rev. Biochem. (Snell, E. E., ed.) Vol. 42,
29. Bredeck, H. W., and Mayer, D. T., Mo. Agr. Exp. Sta. Res. Bul. 491,
30. Goode, L., Warnick, A. C., and Wallace, H. D., J. Animal Sci. 24, 4
31. Webb, E. C., Comprehensive Biochemistry (Flochin, M., and Stotzy,
E. H., eds.) Vol. 12, p. 1 Elsevier, Amsterdam, 1964 (and refer-
32. Webb, J. L., Enzyme and Metabolic Inhibitors, Vol. I, chapter 2.
Academic Press, New York, 1963.
33. Tsubai, K. K., and Hudson, P. B., Arch. Biochem. Biophys. 55, 191
34. Tsuboi, K. K., and Hudson, P. B., Arch. Biochem. Biophys. 55, 206
35. Gibson, K. D., Biochim. Biophys. Acta 10, 221 (1953).
36. Bodansky, 0., J. Bio1. Chem. 129, 197 (1939).
37. Saleem, M., and Alikhan, M. A., Comp. Biochem. Physiol. 49B, 129
38. Chandhary, K. O., Moorijani, S., and Lemonde, A., Can, J. Biochem.
42, 1769 (1964).
39. Saleem, M., and Alikhan, M. A., Comp. Biochem. Physio1. 49B, 129
40. Webb, J. L., Enzyme and Metabolic Inhibitors, Vol. III, chapter
7. Academic Press, New York, 1966.
41. Dixon, M., and Webb, E. C., Enzymes, 2nd edition, pp. 279-282
Academic Press, New York, 1964.
42. Cotton, F. A., and Wilkinson, G., Advanced Inorganic Chemistry,
3rd edition, p. 385. Interscience, New York, 1972.
43. Heide, K., and Schwich, H-G., Angew. Chem. Internat. Edit. 12, 721
44. Sharon, N., Sci. Amer. 230, 78 (1974).
45. Hadding, U., and Miiller-Eberhard, H. J., Irmmunology 16, 719 (1969).
46. Castner, T., Jr., Newell, G. S., Holton, W. C., and Slichter, C. P.,
J. Chem. Phys. 32, 668 (1960).
47. Blumberg, W. E., Magnetic Resonance in Biological Systems (Ehren-
berg, A., Malmstrim, B. G., and V~nng~rd, T., eds.), p. 119.
Pergamon Press, Oxford, 1967.
48. Dowsing, R. P., and Gibson, J. F., J. Chem. Phys. 50, 294 (1969).
49. Peisach, J., Blumberg, W. E., Lode, E. T., and Coon, M. J., J.
Biol. Chem. 246, 5877 (1971).
50. V~nngard, T., Biological Applications of Electron Spin Resonance
(Schwartz, H. M., Bolton, J. R., and Borg, D. C., eds.), p. 411.
Wiley-Interscience, New York, 1972.
51. Aasa, R., and Aisen, P., J. BioL. Chem. 243, 2399 (1968).
52. Windle, J. J., Wiersema, A. K., Clark, J. R., and Feeney, R. E.,
Biochem. 2, 1341 (1963).
53. Aasa, R., Malmstroim, B. G., Saltman, P., and Vlangard, T., Biochim.
Biophys. Acta 7j[, 203 (1963).
54. Sundararajan, T. A., and Sarma, P. S., Biochemical J. 56, 125 (1954).
55. Singer, M. F., and Fruton, J. S., J. Biol. Chem. 229, 111 (1957).
56. G10mset, J. A., Biochim. Biophys. Acta 32, 349 (1959).
57. G10mset, J., and Parath, J., Biochim. Biophys. Acta 39, 1 (1960).
58. Sabel, K-G., Glomset, J., and Porath, J., Biochim. Biophys. Acta
50, 135 (1961).
59. Campbell, H. D., and Zerner, B., Biochem. Biophy. Res. Comm. ff,
60. Jacobs, M, M., Nye, J. G., and Brown, D. M1., J. Biol. Chem. 246,
1419 (1971) .
61. Nochumson, S., O'Rangers, J. J., and Dimitrov, V., Fed. Proc, 33,
abstract 876, 1378 (1974).
II. INTERACTION OF NITRIC OXIDE WITH CYTOCHROME
P450 AND P420 FROM Pseudomonas putida
Cytochrome P450 is a hemoprotein whose name is derived from its
unusual visible spectrum when reduction of the iron of the heme group
is followed by addition of carbon monoxide. Whereas most ferrous
hemoprotein-CO complexes absorb maximally at 420 nm, P450-CO absorbs
around 450 nm. In 1958 Klingenbergl and Garfinkel2 discovered this
cytochrome in the microsomal fraction of rat livers. Subsequent
studies on P450 from mammals have shown this to be an enzyme function-
ing in a mono-oxygenase system. Mono-oxygenases, whose role is the
insertion of a hydroxyl group into the substrate, are important in
fatty acid oxidation, steroid metabolism and drug detoxification.
P450 present in mammals is membrane bound and can be solubilized only
with difficulty. Detergents have been used for this purpose. Early
attempts at purification resulted in an appreciable amount of an inac-
tive form of P450 which absorbed at 420 nm upon'reduction plus CO;
this was named PA20.
A soluble P450 can be induced in the bacterilrnPseudomonas putida
when grown on camphor as the sole carbon source.3,4 (From this point
P450 will be used to denote the oxidized enzyme from the bacterium and
P450cam the camphor completed species unless otherwise noted.) The
soluble nature of this cytochrome makes it very attractive for study
as a model of other mono-oxydases. The hydroxylase system consists
of a flavoprotein (Ea), and iron-sulfur protein (Eb) and cytochrome
P450 (Ec). While other flavoproteins can substitute for Ea,5 the iron-
sulfur protein, putidaredoxin, is quite specific. The first step in
the enzymatic degradation of camphor is hydroxylation at the 5-exo
The stoichiometry of the reaction, which is inhibited by carbon monoxide,
NADH + H+ + 02 + cam -M NAD+ + H20 + cam-0H
and the order of electron flow from the reducing agent, nicotinanide
adenine dinucleotide to molecular oxygen is:
NADH --* Ea -w. Eb -<-Ec - 02.
However, camphor must be present before Eb reduction of P450 can occur.5
A ternary complex of reduced P450, 02 and camphor can be observed spec-
troscopically,5>6 but product formation will not occur unless putidare-
doxin is present. This complex could be formed independent of the
methods, either chemical or enzymatic, used to reduce the iron.
Attempts at forming oxy-P450 in the absence of camphor were unsuccessful.
The molecular weight of P450 is 45,000.7 Each molecule contains
one ferriprotoporphyrin IX group,7 shown in Figure II-1, and a
cavalently linked carbohydrate group was also found.5 Isoelectric
Figure II-1. The heme group. Exclusive
of the side chains, the ring has 4e-fold symmetry.
The iron has two sites of coordination remaining
(along the axial direction).
points of pH 4.55 for P450 and 4.67 for P450ca have been reported.8
The UV and visible absorption spectra have been published,5'8 The
spectrum of oxidized P450, in which form the enzyme is isolated, is
typical of low spin hemoproteins. Addition of camphor causes a blue
shift of the Soret peak (417 mm to 391 rmm); this has been assigned as
a shift of the iron from low spin to high spin. ESR9 and M~ssbauerl0
studies between liquid nitrogen and liquid helium temperatures have
also shown ferric P450 to be low spin and camphor completed P450 to
exist in a high spin/1ow spin mixture. The binding of carbon monoxide
to reduced P450 was found to occur at a slower rate in the presence
of camphor by stopped-flow studies.11 Thus the substrate camphor has
a pronounced effect upon the active site of the enzyme, believed to be
in a hydrophobic pocket.
As in the case of mammalian systems, denatured P450, i.e., P420,
can also be formed. The enzyme has been found to be quite labile at
room temperature, although presence of the substrate camphor or
certain sulfhydryls, such as cysteine, act as protecting agents. Treat-
ment with acid, acetone or guanidine result in conversion to P420.
Activity can be restored by sulfhydryls in certain cases, depending
upon time of exposure.12
A number of studies have been undertaken to discover the identity
of the axial ligands of the iron. ESR9'13 results and sulfhydryl
titrationll suggest that one ligand may be a sulfur atom from a cysteine
residue. It has been postulated that the unusual absorption of reduced
P450-CO is from coordination of sulfur and CO in the axial positions.15
NMR measurements of the relaxation rate of water protons in the bulk
solution, which is affected by the iron atom, have shown rapidly
exchanging protons within the coordination sphere of the paramagnetic
ion.14 The presence of an imidazole ring from a histidine residue, as
proposed from isoelectric focusing experiments,8 would be supported by
the NMR data. Protons could be exchanged at the 3-nitrogen position
while coordination to the iron would be through the 1-nitrogen. CO
complexes of the highly studied hemoproteins such as hemoglobin,
myoglobin and cytochrome c are known to have an imidazole nitrogen
from a histidine residue in the second axial site.
Nitric Oxide as a Spin Probe.
In recent years nitric oxide has been successfully used as a
spin probe of the heme environment of a number of hemoproteins. As
nitric oxide has an unpaired electron (2Hi), its complexes with ferrous
hemoproteins can be studied by ESR whereas the corresponding complexes
of carbon monoxide and oxygen cannot. Examination of the ESR of nitric
oxide complexes of cytochrome c,16 hemoglobinl7 and myoglobinl8 has
shawn hyperfine splitting which was assigned to both the nitrogen
nucleus of the NO molecule and the nitrogen of the second axial
ligand, known to be that of an imidazole group.
This work was undertaken to study the interaction of nitric oxide
with cytochrome P450 and P420 in an attempt to obtain information as
to the identity of the axial ligands of the iron in the porphyrin
Materials and Methods
Concentrations of P450 and putidaredoxin were determined spectro-
photometrically using published extinction coefficients.5 These
purified enzymes were generously provided by Dr. I. C. Gunsalus of the
University of Illinois.
Optical absorption, circular dichroism and ESR spectroscopies were
described in the previous section. Fourier transform infrared spectra
(Digilab) were taken by Dr. P. Callahan (University of Florida). Matched
0.025 mm pathlength Irtran 2 cells were used. Via computer interfacing,
a typical difference spectrum was obtained as follows. Sixty scans of the
sample were taken and summed. Sixty scans of the reference were also summed
and then subtracted from the sample. The conventional spectrum of % trans-
mittance vs wavenumbers was plotted.
The heme group of any hemoprotein can be removed by the acid ketone
method, providing it is not covalently bound. The method call for treat-
ment with acid followed by heme extraction with methylethylketone (MEK).
HC1 (0.1 N) is added to a solution of salt-free, ice-cold hemoprotein
to pH 2, followed by an equal volume of cold MEK. The solution is
swirled for several minutes then allowed to stand "in the cold. The
mixture will separate into an upper organic layer containing the heme
and a lower aqueous phase of protein. The procedure may be repeated
at this point to insure complete removal. The aqueous layer is then
dialyzed against water to remove dissolved ketone.
Hemin was added to apomyoglobin to insure that the reconstituted
protein had the same absorption characteristics as the native. Also
apomyoglobin could be spectrally titrated with hemin by observing the
increase at 408 nm to verify that one heme group per molecule was
Preparation of the Hemoprotein-Nitric Oxide Complexes
Nitric oxide gas was purchased froml Matheson and was purified
immediately before use by passage through a column (0.6 x 20 cm)
containing silica gel and potassium hydroxide. Sodium nitrite was
from MCB. 99.8% enriched 15N-sodium nitrite was obtained from Prochem.
The enzyme solutions were purged under an atmosphere of oxygen-free
argon or nitrogen prior to gentle bubbling with NO. Samples for
optical studies were sealed in Thunberg cuvettes. Samples for ESR
studies were placed in quartz tubes, sealed with parafilm and then
frozen in liquid nitrogen.
Results and Discussion
Figure II-2 shows the optical spectra of P450-N0 complexes.
Identical spectra of reduced P450-N0 could be obtained either by
adding nitric oxide to reduced P450 or by generating NO in solution
from sodium nitrite in the presence of excess dithionite. Note that
there is a difference in the spectra depending upon whether camphor
is present or absent. Absorption maxima and extinction coefficients
are given in Table II-1.
It has been shown that nitric oxide can act as a reducing agent
toward certain hemoproteins. Ferric hemoglobinl9 plus NO will give
Figure II-2. Optical absorption spectra of
complexes of cytochrome P450 and nitric oxide.
Solutions are 5.42 yM in heme and are buffered
in 50 mMi phosphate, pH7.
300 400 500 600