.fi
5
_‘
E
‘
x
‘
h
_
¥
5
%
%
g
fi
*
N

 

THS.

OE? SQUAH SEE?) moaum

*E’E 3‘5"???" CF TE:

hast: Ea? 11:3 Dagrea cf M: 3;
MCHEGAN STAFF COLLEGE
.22.“ ,j. .2 _ _
Anus: EE Emrwy

2947

4 44 I .
' "mflw—w

 

H496

This is to certify that the

thesis entitled

71! M66, (3,! HP dew/an/io/r 9i
5‘qu J‘t’acf f/Oh/I'I]

presented by

’4 ll en 1?. Horny

has been accepted towards fulfillment
of the requirements for

At J: degree Mitt/7

 

(9U) _

‘ Major prof

Date ‘ ‘ 22, W¢77

 

D
.
-._.____~--_ _ _ D

A STUDY or THE DENATURATION
or SQUASH SEED GLOBULIN

by
Allan ll . Harvey

A THESIS
Submitted to the Graduate School of lichigan
State College of Agriculture and Applied
Science in partial fulfilment of the
requirements for the degree of

MASTER OF SCIENCE

Department of Ohemi atry
1947

FTC /2... ctr/5
#351}

194.383.;

ACKNOWLEDGMENT

The author of this thesis wishes to
express his sincere gratitude to
Professor 0. D. Ball for his

supervision and assistance.

INTRODUCTION
HISTORICAL

1. Isolation of squash seed globulin

CONTENTS

Page

Analyses of squash seed globulin .............

Properties of squash seed globulin
2. Methods of estimating sulfhydryl groups

in proteins 000000000000000000000000000.00

3. the influence of carbohydrates on the

coagulation of proteins ..................

EXPERIMENTAL

1. An attempt to liberate sulfhydryl groups

from squash seed globulin ........

. Nethod ..

'IIFJUOUJF’

00000.00

0 Reagentfl ooooeeeeeeeonOe

Iodine titration of squash seed

. Preliminary experiments ...............
Experiments on squash seed globulin ...

. Nitroprusside tests ...................

81°bu11n O0.0000000000000000000000

2. The effect of sugars on the coagulation of

squash seed globulin .....................

A. Coagulation by hydrochloric acid ......

BO coamlation by heat 00.000000000000000.

DISWSSION 000000000000
CONCLUSIONS ...........
BIBLIOGRAPHY .........

000000000000000000000.0000

00000000000

10

13
13
in
20
27
36

37

40
42
47
55

INTRODUCTION

Relatively little attention has been given the protein
squash seed globulin in Biochemical and Physiological _
experimentation although it was one of the first globulins
to be isolated and identified as a protein of this type.
The study of the globulin was resumed a few years ago
whenit received prominence as a substitute for edestin,
which was becoming less available as the restrictions on
the growing of hemp seed increased. A waste product of
the canning industry in some states, squash seed globulin
is always on the market at a price that is usually a small
fraction of that commanded by most cucurbit seeds. The
methods of isolation of the globulin are quite uninvolved
and the yield fairly high.

A study of the denaturation of squash seed globulin
was undertaken in order to record more data on the pro-
duction of sulfhydryl groups by various reagents, and to
determine the effect of sugars on the treatment of the
globulin with these reagents. As more information is
accumulated on the preperties and reactions of squash
seed globulin, a better understanding of the role it plays
in animal physiology will be realized, and the chemistry
of the proteins of both plants and animals will be brought
into closer relationship.

In the historical section of this thesis an attempt

was made to submit all the literature pertaining to the
protein squash seed globulin. In view of the nature of
the problem emphasis was placed on those parts of the
literature dealing with denaturation and the content of
sulfur and sulfur-containing amino acids in squash seed
globulin.

HISTORICAL

In the review of the literature pertaining to this
problem, consideration will be given to the following
topics:

1. The isolation, analyses, and.properties of squash
seed globulin.

2. Methods of estimating sulfhydryl groups in proteins.

3. The influence of carbohydrates on the coagulation

of proteins.

ISOLATION OF SQUASH SEED GLOBULIN

The main.protein isolated from the seeds of squash,
Ougurbita maxima, is known as squash seed globulin or
cueurbitin. It is soluble in saline solutions but insol-
uble in water, can be precipitated from saline extracts
of the seeds by dilution or by dialysis, and therefore
is termed a globulin protein according to Osborne's
classification (1).

Squash seed globulin was first prepared and analyzed
in 1878 by Barbieri (2) whose preparations were not,
however, crystalline. In 1881 Grdbler (3) first applied
the method of allowing warm sodium chloride solutions
saturated with squash seed globulin to cool in order to
obtain octahedral crystals. Chittenden and Hartwell (u)

later prepared the protein in the crystalline form and
published an analysis of it.

Iith the exception of Osborne's early work (5) on
seed globulins no attempt was made to improve the methods
of preparing squash seed globulin until 1939. Krishnan
and Krishnaswamy (6) used 10% sodium chloride solution
to remove the globulin, then diluted this extract with
15 volumes of water at 40° in order to bring about pre-
cipitation and crystallization.

The difficulty of obtaining hemp seed during the
last few years instigated a change over to squash seed
as a readily available source of seed globulin in place
of edestin. This substitution was reported by Vickery,
Smith, and Nolan (7) and later worked out in detail by
Vickery, Smith, Hubbell, and Nolan (8). Their method of
preparation of squash seed globulin included the use of
10% sodium chloride for extraction purposes, but specified
dilution of the extract with four volumes of water at
600 to obtain crystalline globulin. Preheating of the
squash seed extract to 750 and subsequent filtration
before dilution ensured the removal of any albumins present.
Slow cooling of the diluted solution from 600 made for
larger crystals, which could either be collected and dried,
or kept under 2% sodium chloride solution for some time.

ANALYSES OF SQUASH SEED GLOBULIN

Barbieri (2) made the first elementary analysis of
squash seed globulin on completion of its preparation.
His value for sulfur content of the amorphous protein
was 0.60%. Later Grabler (3), Chittenden.and Hartwell
(4), and Osborne (5) completed.analyses of the crystalline
protein with generally agreeing results. Their values
for percent sulfur were:

Grdbler 0.96
Chittenden and Hartwell 1.01
Osborne 0.89
Average sulfur content 0.95

Not until 1923 was further significant analysis done
on squash seed globulin. Jones and Gersdorff (9) isolated
the crystalline globulin from cantaloupe seed, Cucumis
Egggb and.made a comparative study of this protein with
the crystalline globulin of the squash seed. their com-
parison of the nitrogen distribution indicated a great
similarity between the two globulins. The analysis of
squash seed globulin yielded a sulfur content of 1.00%
on a moisture and ash-free basis. Analysis of the basic
amino acids of squash seed gldbulin yielded a value of
1.26% for cystine. No difference in the crystallographic
or cptical properties of the two globulins was detected.

4

Similar comparative work was done on the seed glob-
ulins from pumpkin, squash, watermelon, and cucumber by
Smith, Greene, and Bartner (10). Their tables of amino
acid composition indicated small differences in the
varieties of seed globulins, as well as differences
between edestin and the crystalline globulin of the
tobacco seed. '

Hess and Sullivan (11) investigated the cysteine,
cystine, and methionine content of squash seed globulin.
Determinations of the sulfhydryl groups present in the
unhydrolysed.protein.were compared with the cysteine
determined in the acid hydrolysate of the same protein
and the resulting values were found to agree. According
to Hess and Sullivan, this finding indicated that cysteine
complexes were present in the native protein. The cysteine
content of the unhydrolyzed protein in the native state
was determined by direct titration with iodine according
to the Okuda method for cysteine. The same value was
obtained whether the titration was carried out in the
presence of S millimoles of guanidine hydrochloride per
ml. of solution or in its absence, since the sulfhydryl
groups of proteins can be titrated by iodine whether the
protein is in the native or denatured state.

In squash seed globulin cysteine, cystine, and meth-
ionine accounted for 87% of the total sulfur of the

5
'unhydrolyzed proteins. The sulfur-containing amino acids,

however, accounted for all of the sulfur in the hydroly-
sates of squash seed globulin. About 13% of the sulfur
is lost during the hydrolysis of this protein in a form
not yet explained.

Using the following hydrolyzing agents, Hess and

Sullivan's results for squash seed globulin were:

Hydrolyzipg Aggy; figysteine metine
20%.301 and Urea o.u9 0.82
20% 301 - 0.17 1.11;
6H 112304 0.35 0.97
By direct titration with iodine: cysteine 0.51%
Total sulfur in unhydrolyzed protein: 0.99%
Cysteine, cystine, and methionine sulfur: 0.86%

Further work on squash seed globulin was published
by Hess and Sullivan (12), in which the phenylalanine
content of the protein was determined colorimetrically.
Hydrolyzing agents employed were: 7! H2804, 20% HCl, and
5H HaOH.

Smith and Greene (13) completed a more recent study
of the amino acid composition of squash seed globulin.
The following sulfur distribution in squash seed globulin

was reported:

Cystine 1.11%
Cystine sulfur 0.30%
Methionine sulfur 0.50%

Total sulfur (calculated) 0.80%
Total sulfur (analysis) 0.95%

PROPERTIES OF SQUASH SEED GLOBULIN

In his initial investigations of squash seed globulin
in 1892, Osborne (5) found that in distilled water, either
at 200 or 600 the protein, whether separated from solution
in sodium chloride brine by dialysis or by cooling, was
wholly insoluble. It was also insoluble in strong or
diluted glycerin, cool or warmed to 60°. In 10% salt
solution the crystals dissolved with the exception of a
mere trace, and the resulting solution was not precipitated
by saturating with sodium chloride, but was completely
precipitated by saturating with ammonium sulfate or mag-
nesium sulfate. In 0.l% sodium carbonate and 0.01%Epotash
solutions the protein dissolved readily and completely;
these solutions were not precipitated by sodium chloride.
lhen dissolved in 10% sodium chloride solution and heated
to coagulum, the protein behaves as follows: turbidity
at 87°; flocculation at 95°; and after filtering and
boiling, no coagulation. However, on adding a drop of

7
acetic acid, the filtered solution yielded a large pre-

cipitate.

Gaborne and Harris (14) reported their work on the
specific rotation of squash seed globulin - HEC- 38.730.
Jones and Gersdorff (9) determined the refractive index
of the globulin - N30: approx. 1.5fl5. An investigation
of the isoelectric points of acetylated and deaminated
cucurbitin was carried out by Kizel and Jurkevitsch (15).
The isoelectric point of cucurbitin was reported pH 6.3.

The majority of the information on squash seed glob-
ulin has been supplied by Osborne (1) in his “monograph
on seed globulins. In this publication the properties
of seed globulins in general were recorded on the assump-
tion that most proteins of this origin were similar in
physical and chemical behaviour. The properties that can
safely be attributed to squash seed globulin from Osborne's
work are listed below. Host of the globulins from seeds,
unlike those of animal origin, are incompletely coagulated
by heating their faintly acid solutions, and in some
cases are not coagulated at all. The degree of solubility
of many seed globulins in.sodium chloride solution depends
much on the temperature of the solution and increases
rapidly at 30°. Their solubility in salt solutions is
much affected by the amount of acid present; in general,

the greater the quantity of acid, the stronger the saline

8
solution necessary to dissolve the globulin. Precipita-
tion of seed globulins by acids may be due either to the
acid uniting with a base previously combined with the
protein to form a soluble compound, or by the formation
of a salt of the protein insoluble in water. Many of the
seed proteins appear to be little affected by long treat-
ment with alcohol, with little or no evidence of change
caused. Certain gldbulins pass into insoluble modifica-
tions, the proteans, on standing in contact with pure

water, or in the presence of hydrogen ions.

METHODS OF ESTIMATING SULFHYDRYL GROUPS IN PROTEINS

While the term 'denaturation'~has become firmly
implanted in the literature, the definition of this process
has been very loose and subject to many changes and re-
visions. In the present thesis the following definition
as proposed by Heurath, Greenstein, Putnam, and Erickson
(16) can appropriately be assigned to "denaturation”:
Denaturation is any non-proteolytic modification of the
unique structure of a native protein, giving rise to
definite changes in chemical, physical, or biological
properties. This definition excludes processes which
result in the hydrolysis of peptide bonds, i.e., chemical
or enzymatic degradation.

9
The literature on the liberation of sulfhydryl groups

during the denaturation of proteins and the occurrence of
sulfhydryl groups in.proteins has been amply covered by
Duddles (17) and by Hardt (18). Hardt has also reviewed
the work on the methods of estimating sulfhydryl groups
in proteins up to the work of Anson and Hirsky (19) (20)
(21), and of Anson (22) (2}).

Continuing his work with the sulfhydryl groups of
proteins, Anson (24) investigated both the nitr0prusside
and the ferricyanide reactions on egg albumin denatured
with.urea and with guanidine hydrochloride. He observed
that the amount of ferricyanide reduced to ferrocyanide
by the sulfhydryl groups of denatured egg albumin was,
within widely separated limits, independent of the ferri-
cyanide concentration. Sulfhydryl groups could be abolished
by reaction of the native form with iodine. Furthermore,
it was possible to oxidize all the sulfhydryl groups with
iodine without oxidizing many of them beyond the disulfide
stage.

lirsky (25) experimented with different denaturing
agents on egg albumin, using the nitr0prusside and iodo-
acetamide reactions, as well as the ferricyanide reaction
to estimate the number of sulfhydryl groups formed upon
denaturation. He noted that in neutral medium, ferricyanide
appeared to react only with the sulfhydryl groups of egg

10
albumin; therefore the quantity of ferrocyanide formed
could be considered the equivalent of the number of sulf-
hydryl groupsrbgg albumin detectable with nitr0prusside.
In solutions of urea, guanidine hydrochloride, and Duponol
PC (in excess) the same number of sulfhydryl groups were

formed.

THE INFLUENCE OF CARBOHYDRATES ON THE COAGULATION
OF PROTEINS

The interaction between carbohydrates and.proteins
has been suspected as the underlying common basis for a
wide variety of naturally occurring phenomena, many of
which play important roles both in biological and in
industrial processes. Investigators have attempted to
gain.more knowledge of this interaction by studying the
denaturation of proteins as influenced by carbohydrates.

In 1929, Beilinnson (26) reported the inhibition of
the heat coagulation of proteins by sucrose. Egg albumin
was found to be thermostable at 75° in.a solution saturated
with sucrose. Glycerol also provided.protection from heat
coagulation although the effect was not as great as from
sucrose. Newton and Brown (27) later showed that sucrose
and glucose prevented the denaturation or coagulation of
plant sap proteins by freezing.

(28)
Hrosteaux and Eriiksson—QuenselAfound that when the

11
proteins hemoglobin, serum albumin, serum globulin, edestin,
phycoerythrin, erythrocruorins, and hemocyanins were dried
in vacuum at room temperature, they were more or less
denatured, and did not assume their former state of dis-
persion when placed in water. The addition of sugars,
sodium chloride, alanine, glycine, or gelatin to the
proteins before drying protected them to some extent from
denaturation, the effectiveness of the substances decreas-
ing in the order given. Urea was practically ineffective,
while lactose was the most effective. The minimal quan-
tities found by experiment to prevent denaturation com-
pletely closely approached the quantities calculated
necessary to form a unimolecular layer on the surface of
the protein micelles.

Ball, Hardt, and Duddles (29) found that D-glucose,
D-fructose, D-mannose, L-arabinose, D-xylose, D-mannitol,
and sucrose had a preventative effect on the heat coag-
ulation of egg albumin. The inhibiting effect did not
increase with the increase of time of contact of the agent
with the egg albumin even at a high pH. The influence of
sugars and mannitol on the liberation of sulfhydryl groups
from egg albumin was determined.by the use of‘both the
iodoacetic acid method and the phenolindo-2,6-dichloro-
phenol titration method for the estimation of liberated
sulfhydryl groups. The influence of sugars on the heat

12
coagulation of proteins was studied by determining the
amount of nitrogen left in the filtrate from the heat
coagulated.protein solution.

Using electrophoretic procedures, Hardt, Huddleson,
and Ball (30) found that D-glucose, D-galactose, L-arabinose,
sucrose, D-fructose, lactose, D-mannitol, and D-xylose all
showed some degree of inhibiting action against the C
component formation in heated bovine serum. The 0 component
was formed by heating an albumin—alpha-globulin fraction
of bovine serum.

Fischer (31) discovered that D-ribose, L-arabinose,
ascorbic acid, and digitoxose prevented the heat coagu-
lation of bovine serum at a concentration of 5% by volume.
Glucose, sorbose, galactose, fructose, glucosamine, inos-
itol, maltose, lactose, sucrose, raffinose, glycogen, and
soluble starch had no effect on bovine serum at the same
concentration. The preventative action of ascorbic acid
appears to be the result of its acidity since its sodium
salt has no inhibiting effect.

13
EXPERIMENTAL

The experimental work on this problem was separated
into two parts:
1. An.attempt to liberate sulfhydryl groups from squash
seed globulin
A. Method
B Reagents
0 Preliminary experiments
D. Experiments on squash seed globulin
I. NitrOprusside tests
I. Iodine titration of squash seed globulin"
2. The effect of sugars on the coagulation of squash
seed globulin
A. Coagulation by hydrochloric acid
B. Coagulation by heat.

AN ATTEMPT TO LIBERATE SULFHYDRYL GROUPS FROM SQUASH
SEED GLOBULIN

' A. METHOD:

Protein sulfhydryl groups are estimated by means of
their reaction with ferricyanide, as a result of which
they are oxidized to disulfide groups and ferrocyanide
is formed (25):

2 protein SH + 2 ferricyanide : protein 8-8 + 2 ferrocyanide.
An excess of ferricyanide is added and the quantity of

In
ferrocyanide is estimated. This is done by adding ferric
sulfate which reacts with ferrocyanide to form Prussian
blue which is estimated in a photoelectric colorimeter.
The intensity of the blue color is a measure of the number
of active, protein sulfhydryl groups.

For relatively simple sulfhydryl compounds such as
cysteine and glutathione, the reaction with ferricyanide
proceeds stoichiometrically (32). Conditions under which
the sulfhydryl groups have reacted with ferricyanide in
a precise and definite manner have been found (25). The
protein should be dissolved in approximately neutral sol-
utions of the denaturing agent. Under these conditions
the reaction proceeds rapidly. It is completed in less
than one minute; no more ferricyanide is reduced in 60
minutes than in one minute. Nor within wide limits do the
concentration of ferricyanide or the temperature affect

the quantity of ferricyanide reduced.

B. REAGENTS:

The experimental work was based on Anson's procedure
(24), and an attempt made to outline a method for the
estimation of sulfhydryl groups produced on the denatur-
ation of squash seed globulin using the reagents listed
below.

Protein: Finely ground squash seeds, Cucurbita maxima,

were used as a source of squash seed globulin. Three

15
different preparations of the globulin, all isolated accor-
ding to the method of Vickery, Smith, Hubbell, and Nolan
(8), were employed in this problem. All preparations
were crystallized three times. The first preparation,
isolated in 19u5, was kept in a dry condition‘until used
subsequently in 10% sodium chloride solution. Difficulty
was encountered in putting this dried globulin into 10%
sodium chloride solution. About one quarter of the amount
used would not dissolve and appeared to be denatured,
probably due to the methods of washing and drying the
protein during isolation. The second and third prepar-
ations were isolated currently and left in the refrigerator
as crystals under 2% sodium chloride solution and under
toluene. This globulin went back into saline solution
with little or no evidence of precipitated, denatured
globulin. Before making nitrogen determinations all glob-
ulin solutions were filtered slowly through paper pulp.
The following squash seed globulin solutions, all 10% in
respect to sodium chloride, were used in this problem:
Protein I: made up from the dried globulin prepared in

19n5. Protein content: 11.61 mg./ml.
Protein II: made up from the second protein.preparation
kept as crystals under 2% sodium chloride.
Protein content: 3.79 mg./ml.
Protein.III: made up from the third protein preparation
kept under 2% sodium chloride. Protein

16
content: 9.79 mg/ml.

Protein II was used for most of the qualitative side
experiments in the problem. The nitrogen content of each
protein solution was determined in triplicate by Micro-
kjeldahl procedure (33). Two ml. aliquots of each solu-
tion were used and the boric acid distillates titrated
with 0.0205M hydrochloric acid, using one dr0p of methyl-
ene blue and eight drops of methyl red as an indicator.
From the nitrogen content, the protein concentration of
each solution was determined , using the value 17.05%
for the nitrogen content of squash seed globulin (9).
Sodium Chloride: The protein was kept in 10% (or 1.67M)
sodium chloride solution to prevent.its precipitation
in the crystalline form. However, at 200 it will stay
in solution under 5% sodium chloride. The reacting sol-
utions were maintained at 10% sodium chloride concentra-
tion until after ferricyanide was added; then the salt
. content is irrelevant since denaturation is complete.
Cyteine Hydrochloride: Cysteine was chosen as a suitable
sulfhydryl reagent to test the validity of the method
proposed for the liberation of sulfhydryl groups from
squash seed globulin. A stock solution of cysteine hydro-
chloride was made up by dissolving 157.5 mg. of cysteine
hydrochloride in 10 ml. of 5M hydrochloric acid, and dil-
uting to 100 ml. to make the solution 0.01M. It was

17
stored in the cold and no longer used after 5 or 6 hours.
To standardize the solution the hydrochloric acid was
neutralized by adding standard sodium hydroxide (using
congo red as an indicator) and the neutral solution
titrated with 0.01H potassium ferricyanide until a dis-
tinct yellow color was producedm(32).

72529: For the denaturation of egg albumin and the pro-
duction of sulfhydryl groups both Anson (24) and Mirsky
(25) used solutions 8.3M in respect to urea. In his exper-
iments with edestin and egg albumin, Burk (34) reported
the minimal effective concentration for urea in aqueous
solutions as 6.06M for 2% egg albumin, and 1.5M for 2%
edestin solutions. In view of these figures, and also in
view of the fact that at 1.67M sodium chloride concentra-
tion there is inhibition of sulfhydryl production below
approximately 2.8M urea concentration for edestin (34),
the urea concentration for the production of sulfhydryl
groups in squash seed globulin was arbitrarily set at 4M.
This value is loosely based on the assumption that squash
seed globulin and edestin are similar in this respect.
Cuanidine Hydrochloride: The denaturant guanidine hydro-
chloride was prepared from guanidine carbonate as directed
by Anson (24), dried.under vacuum, and kept in an air-
tight container until used. Anson used between 1.4 and
2.4 gm. of guanidine hydrochloride per m1. of 2% egg albu-
min. An excess of guanidine hydrochloride was used in the

18
work on squash seed globulin in order to ensure denatur-
ation.

Sodium Laugyl Sulfate: To take the place of Duponol PC,

a denaturant used by Anson (24) and by Mirsky (25), a 10%
solution of sodium lauryl sulfate was used. Duponol PC

is a commercial product containing a mixture of C10 - C18
compounds of the series CH3(CH2)n-0320803Na. An excess of
this denaturant was also used on squash seed globulin.
Hydrochloric Agig: Anson (24) carried out the denaturation
of egg albumin with urea in acid medium, because it en-
hanced sulfhydryl production. Burk (34) confirmed this
procedure but noticed that for lactalbumin a hydrochloric
acid medium inhibited sulfhydryl production at low urea
concentrations. Burk gave no figures for edestin with
hydrochloric acid. * According to Bawden and Pirie (35)
urea can shift the pH values of proteins in solution.

They also noted that the pH of protein solutions would
increase 0.8 to 1.0 units on the addition of urea. In
alkaline medium, other groups than sulfhydryl may reduce
ferricyanide (25). In view of these facts the liberation
of sulfhydryl groups from squash seed globulin was carried
out in neutral or in slightly acid mediums.

Buffer Solutions: Buffer solutions were used in order to
standardize the pH of experimental solutions. A M phosphate
buffer'pH 7.0 was made up by mixing equal parts of M K23204
and M KH2P04 solutions (25). A M phosphate buffer pH 6.4

19
was prepared by mixing equal parts of M NaEHPou and M
NaHZPOn solutions (24). Both solutions of buffers were
checked on the Beckman pH meter. Buffer solutions from
pH 4.0 to 6.2 were made up using the potassium acid phtha-
late - sodium hydroxide system (36).
Potassium Ferrigyanide: All standard ferricyanide solu-
tions were made up daily from a 0.1M stock solution, and
kept in the cold. To oxidize sulfhydryl groups to disul-
fide stoichiometrically, Anson (24) used 0.05 millimolss
of ferricyanide per 0.001 millimolss of ferrocyanide pro-
duced. Mirsky (25) used 0.003 millimolss of ferricyanide
per 0.001 millimolss of ferrocyanide produced by egg albu-
min. In the oxidation of cysteine hydrochloride, Anson
(22) obtained best results using 0.002 millimolss of ferri-
cyanide for every 0.001 millimolss of ferrocyanide. For
this work, in order to ensure an excess of ferricyanide,
0.01 millimolss were used in each experimental solution
to oxidize any sulfhydryl groups from squash seed globulin.
Potassium Ferrocyanide: Standard ferrocyanide solutions
were prepared daily from a 0.1M stock solution of potassium
ferrocyanide kept in the cold.
Ferric Sulfate: To develop the Prussian blue color from
the ferrocyanide formed, a solution of ferric sulfate in
gum arabic was made up according to Folin and Halmros (37).
Sum arabic was substituted for gum ghatti, which was un-
available. The effect of the gum is to prevent the

20
precipitation of the Prussian blue.
Sulfuric Agig; M sulfuric acid was added after the reaction
of ferricyanide with sulfhydryl in order'to prevent any
further ferricyanide reduction by non-sulfhydryl groups.
in the mixture.
C. PRELIMINARY EXPERIMENTS
1. Standardization with Ferrocyanide:

For the first part of this problem the Lumetron
colorimeter was used in order to estimate Prussian blue
from ferrocyanide produced by the oxidation of sulfhydryl
groups. The Lumetron was standardized in terms of cysteine
according to the procedures of Mason (32) and of Anson (22).
From the 0.01M stock solution, standard solutions of
cysteine hydrochloride were prepared containing from 0.0015
to 0.0005 millimolss of cysteine. To each was added suf-
ficient standard sodium hydroxide to neutralize the hydro-
chloric acid present, 2 ml. of 0.001M ferricyanide, 4
drops of phOSphate buffer pH 7.0, and the mixture kept in
a water bath at 37° for 5 minutes. After the reaction with
ferricyanide, to each solution.was added 0.5 ml. of ferric
sulfate solution and enough distilled water to bring the
volume of the mixture to 10 ml. The mixture was allowed
to stand 5 minutes before reading the percent transmission
in the Lumetron oolorimeter. Readings were taken using
both the yellow-green (530 mu wavelength) and the red (650

mu) filters. Control solutions containing no cysteine

21
were also run. Prussian blue was formed and the percent
transmission determined under the same conditions with
standard solutions containing from 0.0015 to 0.0005 milli-

moles of ferrocyanide.

TABLE I
Standardization of cysteine hydrochloride solutions with
ferrocyanide solutions of the same strength under the same
conditions, using two wavelengths on the Lumetron color-
imeter for reading percent transmission of the Prussian

blue deve10pedt

Percent Transmission

Cysteine mM Ferrogy mM gysteiggo Egrrogy gysteiggo Egrrocy
0.0015 0.0015 29.5 34.0 _ 9.0 7.0
0.0014 0.0014 33.0 36.5 12.0 9.0
0.0013 0.0013 31.5 40.5 16.0 10.5
0.0012 0.0012 40.0 42.5 16.5 13.0
0.0011 0.0011 49.5 45.0 30.0 16.5
0.0010 0.0010 50.0 49.5 26.0 20.0
0.0009 0.0009 53.5 54.0 32.5 29.5
0.0008 0.0008 60.0 59.0 36.5 34.5
0.0007 0.0007 64.0 71.0 48.0 48.0
0.0006 0.0006 71.0 76.5 61.0 56.0
0.0005 0.0005 76.5 78.0 68.0 60.5

22

The results in Table I indicate that the red filter
(650 mu) is impractical for the measurement of color in-
tensity. Using the yellow-green filter (530 mu), the
results show conformity to the equation:
2 sulfhydryl +-2 ferricyanide ' disulfide + 2 ferrocyanide
only around a concentration of 0.001 millimolss of cysteine.
Furthermore, the values for each determination varied as
much as 4% transmission in four consecutive readings.
Although the method appeared to be substantiated for a
certain concentration of cysteine by this evidence, the
degree of possible error was too high to warrant further
use of the Lumetron colorimeter for purposes of estimating
the Prussian blue color. Moreover, in view of subsequent
work, the need of standardization curves for more precise
determinations of the sulfhydryl groups from squash seed
globulin was not yet indicated.

2. Oxidation of Cysteine Hydrochloride:

From the results of the above work it was decided that
preliminary experiments with cysteine hydrochloride to
determine the effect of reagents on the oxidation of sulf-
hydryl groups should be carried out at aIconcentration of
0.001 millimolss of cysteine. The procedure of Anson (24)
for the production of sulfhydryl groups in egg albumin with
urea was followed as closely as possible, with those modi-

fications presumed necessary for the production of sulfhydryl

23
groups in squash seed globulin.

The procedure outlined below was considered suitable
for cysteine determinations under these conditions. To
0.5 m1. of 0.002M cysteine hydrochloride solution were added
2 m1. of 25% sodium chloride solution and 2.5 m1. of 8M
urea solution, thus making an acid solution, 10% in respect
to sodium chloride and 4M in reapect to urea. The mixture
was held at 370 in a water bath for 5 minutes then 0.5 m1.
of 0.02M ferricyanide solution were added" The ferricyanide
was allowed to react for one minute at 37°, then 5 m1. of
0.0979M sodium hydroxide and 4 drops of phosphate buffer
pH 7.0 were added to bring the pH to neutrality. After
one minute 0.5 m1. of M sulfuric acid, 2.3 m1. of dis-
tilled water, and 1.0 m1. of ferric sulfate were added,
bringing the total volume of the mixture to 10 m1. Thirty
minutes after the addition of ferric sulfate the Prussian
blue developed was read in values of percent transmission
in the Junior Coleman photometer at 530 mu wavelength.

The Junior Coleman photometer was considered more
accurate for this type of work than the Lumetron color-
imeter because of its narrower band width -- 35 mu. The
above procedure deviated from that of Anson (24) in that
the oxidation of sulfhydryl groups was carried out in 10%
sodium chloride concentration, 4M urea concentration, and
in an acidic medium. This acid.medium was held for one

minute before neutralization was effected. In those

24
solutions in which the sulfhydryl oxidation was carried out
in neutral medium, very little Prussian blue was developed.
The presence of 10% sodium chloride in contact with urea
may account for the rise of the pH of the mixture above
neutrality, which in turn would prevent the oxidation of
sulfhydryl groups. The ability of urea to raise the pH of
solutions was suggested by Bawden and Pirie (35).

To establish a relationship between intensity of color
and quantity of ferrocyanide, Prussian blue was formed in
solutions containing known amounts of ferrocyanide under
precisely the same conditions as when cysteine hydrochloride
was used. To 0.5 m1. of 0.002M potassium ferrocyanide
solution were added the same reagents as above, and the
results campared with those of the above procedure in Table
II.

In order to observe the effect on.the development of
Prussian blue of the additional reagents, namely sodium
chloride, urea, ferricyanide, sodium hydroxide, and sul-
furic acid, used in the method already preposed for the
liberation of sulfhydryl groups, the following procedure
was adopted. Only the minimal reagents necessary to
develop the Prussian blue from ferrocyanide solutions
were employed. To 1 m1. of 0.001M potassium ferrocyanide
solutions were added 0.2 ml. of phosphate buffer pH 7.0,
0.5 ml. of ferric sulfate, and enough water to bring the

volume of the mixture to 10 ml. The Prussian blue was

25
estimated in the Junior Coleman.phot0meter 30 minutes
after adding the ferric sulfate.

TABLE II
Effect of the reagents on the oxidation of sulfhydryl
groups and the deve10pment of the Prussian blue color
in the proposed method for the liberation of sulfhydryl

groups from squash seed globulin.

 

Added Added Reagents to form Percent
fixateine Ferrogy, ggussiap blug Transmission
1 mM -- complete 51.5
-- 1 mM complete 53.0
.. 1 mM minimal 39 . 5

The results in Table II show that the complete reagents
necessary to liberate sulfhydryl groups from squash seed
globulin in urea hindered the development of Prussian blue,
but did not interfere to any great extent in the oxidation
of sulfhydryl groups. However, thirty minutes after the
addition of ferric sulfate, the Brussian blue started to
precipitate from all solutions. Up to this point the
intensity gradually increased, the greater'part of the
color being formed at the end of fifteen minutes.

26
3. Liberation of Sulfhydryl Groups from Egg Albumin:

The preliminary experiments with cysteine proved
essentially that any sulfhydryl groups liberated from
squash seed globulin could be oxidized by ferricyanide,
and Prussian blue developed using a modified method.

This method was tried out on a 1.0% egg albumin solution
10% in reSpect to sodium chloride, prepared according to
Anson's method (24). The same procedure was followed in
detail, using 1 ml. of the egg albumin solution and 4 m1.
of 10% sodium chloride solution to bring the total volume
of the protein solution to 5 m1., and the total content

of egg albumin to 10 mg. To this was added 2 drops of M
hydrochloric acid, 2.6 ml. of 12M urea solution, and the
mixture held at 370 for 30 minutes in a water bath. To
oxidize sulfhydryl groups, 0.5 m1. of 0.02M ferricyanide
was added and the reaction allowed to proceed for 5 min-
utes. The mixture was neutralized by adding 2 drops of

M sodium hydroxide and 4 dr0ps of phosphate buffer pH 7.0.
After one minute oxidation was stopped by adding 0.5 m1.
of M sulfuric acid. The precipitated protein was filtered
off and 1 m1. of ferric sulfate solution added to the
filtrate, bringing the total volume of the solution to

10 ml. The Prussian.blue deve10ped was read in 30 minutes
as percentage transmission in the Junior Coleman photo-
meter at 530 mu wavelength. A control solution containing

no egg albumin was used to correct for the color contributed

27
by the reagents.

The Prussian blue developed in the filtrate showed
58.0% transmission intan average of triplicate runs.
Anson (24) found that 0.001 millimolss of ferrocyanide
were formed when ferricyanide was reduced by 10 mg. of
denatured egg albumin in urea solution. In view of this
value and the results of previous experiments with ferro-
cyanide in this problem, the value 58.0% is in close
agreement with the work on cysteine hydrochloride. The
higher transmission may be due to the lower concentration
of urea used in this method, or to the presence of sodium
chloride, both of which differ from the reagents used in
Anson's procedures. However, the chief purpose of this
preliminary experiment was to point out that the modified
method, for the liberation of sulfhydryl groups in squash
seed globulin, was valid for a protein that has been known

to liberate sulfhydryl groups in urea solution.

D. EXPERIMENTS ON SQUASH SEED GLOBULIN

l. Urea Denaturation:

The pr0posed method was carried out on a solution
containing 11.61 mg. of squash seed globulin (Protein I)
per m1., and 10% in respect to sodium chloride. 0n the
basis of a cysteine content of 0.51% for squash seed glob-
ulin (ll), 5 m1. of this solution would contain 0.0024
millimolss of cysteine. The addition of hydrochloric acid

28
to the protein solution produced a slight turbidity that
did not change after 30 minutes in 4M urea solution at 37°.
The addition of ferricyanide caused an increase in tur-
bidity; this condition persisted after neutralization of
the mixture. A white flocculent precipitate was formed on
adding sulfuric acid, which, after filtering, left a clear,
light yellow solution. The average transmission of four
identical runs was 98%. The same results were obtained
with a solution containing 3.79 mg. of squash seed globu-
lin (Protein II) per m1., and 10% in reapect to sodium
chloride. From this evidence it would appear that no
sulfhydryl groups were produced in an expoSed, active posi-
tion by this method.

In the following experiments the different factors
and reagents involved in this method were changed and
revised in an effort to liberate sulfhydryl groups from
the globulin. The results of each change in procedure
were taken into consideration in the ensuing experiments.

To prevent the immediate precipitation of the globu-
lin a.procedure was tried involving the omission of hydro-
chloric acid. Buffer solutions of pH's 7.0, 6.4, 6.0, 5.6 ,
and 5.2 were used in order to establish different levels
of pH for sulfhydryl liberation and oxidation. In prelim-
inary experiments with Protein I it was found that below
a pH of 5.2 buffer solutions caused a flocculation of the
protein. To 5 m1. of the protein solution were added 2.7

29

m1. of 12 M urea solution and 6 dr0ps of buffer solution.
After holding the mixture at 37° for 30 minutes, 0.5 m1.

of 0.02M ferricyanide was added and.any reaction allowed
to proceed for 5 minutes. After adding 0.5 ml. of M sul-
furic acid, the precipitate was filtered off and 1 ml. of
ferric sulfate added to bring the total volume of the
mixture to 10 ml. After 30 minutes no Prussian blue was
formed in any of the solutions, although precipitation of
the flocculent protein was prevented at all levels of pH
until sulfuric acid was added. Ferric sulfate was poured
over the residues of protein on the filter paper from each
mixture; but after 60 minutes no color was produced, show-
ing that no ferrocyanide existed in an occluded form on the
precipitated protein.

Although the preceding experiment showed no evidence
of ferricyanide oxidation of sulfhydryl groups, there
remained the question of whether the protein was denatured
by urea while it was in the unprecipitated form. To inves-
tigate this question qualitatively, Protein I was made 4M
in respect to urea and held for 15 minutes at 37° in a
water bath. To the resulting clear solution was added 4
parts by volume of water in an attempt to precipitate the
crystalline globulin. A gummy, amorphous protein.precip-
itate was formed on dilution.that would not go back into
solution on addition of an excess of 10% sodium chloride,

which would ordinarily effect solution. The same results

30
were obtained with solutions of guanidine hydrochloride

and sodium lauryl sulfate. When hydrochloric acid was
added to the unprecipitated protein solutions after 15
minutes contact with these denaturants, the resulting pre-
cipitates were less flocculent than without denaturants
present.

With 5 m1. of Protein I solution the foregoing pro-
cedure was carried out at pH 5.6 with an increase from 30
minutes to 2 hours in the time allowed for sulfhydryl
liberation in urea solution. A slight turbidity was
noticed in each solution at the end of 2 hours, however,
no Prussian blue was developed.

In an attempt to eliminate any protein precipitation
and to determine the effect of having all reagents present
in the solution at the same time, the following procedure
was employed. To 5 m1. of Protein I were added 2.7 ml. of
12M urea solution.and 6 drops of buffer solution pH 5.6.
The mixture was held at 370 for 2 hours, then 0.5 m1. of
0.02M ferricyanide added. After 5 minutes 1 m1. of speci-
ally prepared neutral ferric sulfate solution was added
and the mixture allowed to stand 30 minutes. Although no
precipitate was formed, no Prussian blue was deve10ped.

The effect of an excess of protein was studied by
adding 50 mg. of the dried squash seed globulin to each of
‘ three test tubes. To each was added 2 m1. of 25% sodium
chloride solution, 1 gm. of urea, 0.5 ml. of M hydrochloric

31
acid, 0.5 m1. of pH 5.0 buffer solution, and the mixture
held at 37° for one hour. One ml. of 0.02M ferricyanide

was added.and the reaction allowed to proceed for 10 min-
- utes. The precipitated protein from the first test tube
was filtered off, the filtrate made to 9 ml. with water,
and 1 ml. of ferric sulfate added. The precipitate was
also removed from the second test tube and the filtrate
brought up to neutrality by adding 0.5 m1. of M sodium
hydroxide and 0.5 m1. of pH 7.0 phoSphate buffer. This
mixture was made to 9 ml. with water and 1 m1. of ferric
sulfate added. To the third test tube was added 1 m1. of
ferric sulfate, with no filtration of the precipitate.
After one hour in urea and sodium chloride solutions,
approximately one half of the protein in all test tubes
had changed from the crystalline state to a flocculent,
amorphous mass. Ten minutes after adding ferric sulfate
no color was evident in any solution. At the end of one
hour the precipitated protein in the third test tube showed
evidence of Prussian blue occluded to it, while the other
two solutions exhibited no ~color. 0n acidifying the
filtrate from the second test tube a deep Prussian blue
color was developed.

To investigate the possibility that on long time

exposure of squash seed globulin to urea there was caused
a liberation of sulfhydryl groups in the preceding exper-

iment, the procedure was repeated with an increase from

32

l to 15 hours in the time of holding the protein solution
at 37° in a water bath. No Prussian blue was developed in
any solution 15 minutes after adding ferric sulfate. The
third test tube, containing the precipitated protein, again
gave evidence of Prussian blue occluded to the precipitate
after standing one hour.

The effect of sodium chloride on the reactions of the
above procedure was studied by omitting the addition of
2 ml. of 25% sodium chloride solution, and reducing the
holding time at 37° to 2 hours. No protein appeared to
go into solution, but rather the crystalline globulin
turned into a partially gummy, denatured mass. The results
were the same as in the foregoing work, including the
appearance of Prussian blue formation after long time
standing of the solution containing the precipitated pro-
tein.

2. Hydrochloric Acid Denaturation:

In preliminary experiments with hydrochloric acid as
a denaturant it was found that the flocculent precipitate
produced when hydrochloric acid was added to a solution
of squash seed globulin in 10% sodium chloride could be
dissolved by adding sodium hydroxide immediately after
precipitation. However, after 15 minutes standing in acid
medium, no dissolution of the precipitated.protein could
be effected by adding even concentrated sodium hydroxide.

33
The pr0posed method for the liberation of sulfhydryl

groups from squash seed globulin by urea was carried out
on 5 ml of Protein I, using 2.6 m1. of 0.01M hydrochloric
acid in place of urea. No Prussian blue was formed in any
of several runs made.

The same procedure was repeated with the ferricyanide
added to the protein solution before holding the mixture
at 37° for 30 minutes. No.0hange in results was shown.

The effect of an excess of protein was studied by
adding 4 dr0ps of M hydrochloric acid and 2 m1. of water
to 50 mg. of the dried protein. The mixture was held at
37° for 2 hours and 0.5 m1. of 0.02M ferricyanide added.
After 5 minutes 1 m1. of ferric sulfate was added and the
color deve10pment estimated after 30 minutes. A slight
amount of green color was noticed on the precipitated
protein. As in the experiments with urea, part of this
protein was in the form of a flocculent precipitate, the

remainder a gummy mass after denaturation.

3. Sodium Lauryl Sulfate Denaturation:

A 10% solution of sodium lauryl sulfate was employed
as a denaturing agent for squash seed globulin, with the
omission of hydrochloric acid from the procedure in order
to keep the protein in solution in the presence of ferri-
cyanide. To 5 ml. of Protein III were added 6 drops of
buffer solution pH 6.0, 1 ml. of 10% sodium lauryl sulfate,

34
and the mixture held at 37° for 15 minutes, followed by
the addition of 0.5 ml. of 0.02M ferricyanide solution.
After 2 minutes the mixture was made to 9 ml. with water
and 1 m1. of ferric sulfate added. The protein precipi-
tated on the addition of ferric sulfate, although the
sodium lauryl sulfate content was increased in an effort
to prevent flocculation of the protein. After a period
of 30 minutes no Prussian blue was noticed in.any solution.

The foregoing procedure was used on 50 mg. of the
dried globulin with no sodium chloride present, and the
denaturation allowed to proceed for 2 hours. No Prussian
blue was deve10ped. When 4 drOps of hydrochloric acid
were added to the mixtures before holding at 37°, the
precipitated protein became more flocculent, but no color
was produced. The same results were obtained on increasing

the holding time from 2 to 15 hours.

4. Guanidine Hydrochloride Denaturation:

Guanidine hydrochloride was employed in the solid
form as a denaturant on squash seed globulin because of
the instability of its aqueous solutions. The denaturation
was carried out at different levels of pH, namely 7.0,
6.4, 5.0, and 4.0, using identical procedures for solutions
containing 50 mg. of the dried globulin, and for solutions
containing 4 ml. of Protein III. To each protein solution
were added 0.2 gm. of guanidine hydrochloride, 0.5 ml. of
buffer solution, 2 ml. of water, and the mixtures held at

35
37° for 60 minutes. One ml. of 0.02M ferricyanide was

added and the mixtures allowed to stand 10 minutes before
adding 1 m1. of ferric sulfate. Protein precipitates
formed in the solutions containing Protein III when ferric
sulfate was added, but no color was formed. The solid
protein did not appear to dissolve to any extent; however,
Prussian blue began to form on the precipitate after 30
minutes standing.

The above work was repeated with the ferricyanide in
the presence of the protein during denaturation. Although
the amount of Prussian blue occluded to each precipitate
was slightly increased, this color did not appear until
the solutions had been standing 30 minutes.

5. Heat Denaturation:

To study the effect of heat denaturation on the liber-
ation of sulfhydryl groups from squash seed globulin, 4 m1.
of Protein III were heated for 30 minutes in a hot water
bath at 1000 in the presence and in the absence of 1 m1.
of 0.01M ferricyanide. Different levels of pH were main-
tained during denaturation by adding buffer solutions of
pH's 7.0, 6.4, 5.0, and 4.0. To all solutions were added
1 ml. of 0.01M ferricyanide and 1 m1. of ferric sulfate.

Ho color was produced until the solutions had stood for

30 minutes. In all solutions a precipitate was formed after
heating which was less flocculent than that produced by
hydrochloric acid. The supernatant liquid above each

36
precipitate was found to be quite turbid, and from it a
flocculent precipitate could be formed by adding 2 drape
of M hydrochloric acid.

E. NITROPRUSSIDE TESTS

The reaction of squash seed globulin to the nitro-
prusside test was taken as a criterion of the validity
of the ferricyanide method used in this problem. The
results of this test, using Protein III, were compared
with those of a 1.5% solution of egg albumin, 10% in
reapect to sodium chloride, and a 0.001u solution of
cysteine hydrochloride under the same conditions. To 4
m1. of the protein or amino acid solution were added 0.2
ml. of M hydrochloric acid and 1 gm. of urea, and the
mixture held at 37° for 15 minutes. On completion of the
denaturation 4 drcps of 5% sodium nitr0prusside solution
and 2 drops of concentrated ammonium hydroxide were added.
The egg albumin and cysteine hydrochloride solutions
immediately turned reddish purple, indicating positive
tests for sulfhydryl groups. The squash seed globulin
solution remained a yellow-brown in color. Using the same
method of denaturation, both egg albumin and cysteine
hydrochloride gave positive tests for ferricyanide while
squash seed globulin showed a negative test. In both tests
the color developed in aqueous solutions of egg albumin

was greater than that deve10ped in sodium chloride solutions,

37

in which a slight precipitation was noticed. However,
the alternative tests carried out without the addition
of hydrochloric acid, and thus without the precipitation
of either squash seed globulin or of egg albumin, gave

identical results as when a precipitate was present.

F. IODINE TITRATION OF SQUASH SEED GLOBULIN

An.attempt was made to duplicate the work of Hess and
Sullivan (11) on the cysteine content of native squash
seed globulin. By direct titration with iodine according
to the Okuda method for cysteine these workers found a
cysteine content of 0.51% in squash seed globulin.

A protein solution containing 11.61 mg. of globulin
' per ml. in 10% sodium chloride solution was made up from
the dried material prepared in 1945. Ten m1. of the
protein solution were made 2%:with reopect to hydrochloric
acid by adding 10 m1. of 4% acid. Five ml. of 4% hydro-
chloric acid and 5 m1. of 5% aqueous potassium iodide
were added. The solution was cooled to-20o and titrated
to a permanent yellow color with M/600 potassium iodate
(in 2%’H01) that had been standardized against cysteine
hydrochloride similarly treated.

The globulin precipitated in an amorphous, flocculent
mass as soon as the first drop of acid was added. The end-

point of the titration was reached with the first dr0p of

38
iodate. The end-point was also reached with the first
dr0p when M/1200 potassium iodate was used, although
theoretically 0.75 ml. of iodate should have been used
on the basis of a cysteine content of 0.51% in squash
seed globulin. The addition of 2 m1. of 1% starch solution
in order to provide a clearer end-point, gave the same
results. In further titrations 10 ml. of 2M glucose in
4% hydrochloric acid solution were used in order to make
the protein 2% with respect to hydrochloric acid but no
difference in results was evident.

An effort was made to prevent precipitation of the
. globulin by changing the sodium chloride concentration
to 4%. At 20° precipitation of the crystalline globulin
is Just prevented in this concentration of sodium chloride.
It was found possible to add more acid to this solution
before precipitation of the amorphous globulin was effected.
However, before the concentration of acid reached 2% the
protein was again.precipitated as a flocculent mass.

To keep the protein in solution it was decided to omit
all hydrochloric acid additions except that present in the
potassium iodate. To 10 ml. of squash seed glObulin solu-
tion in 10% sodium chloride were added 5 m1. of aqueous
potassium iodide, 10 ml. of phosphate buffer pH 6.4, and 2
m1. of 1% starch solution. Four m1. of M/1200 iodate were
used in the titration. The same value was obtained when

no starch was added, but in both cases the endepOint of

39
the titration was quite vague. The value obtained repre-

sents a cysteine content of 2.72% for squash seed globulin.
However, on the addition of the required amount of hydrochloric
acid to these titrations, the depth of the color immediately
produced indicated that the end—point of the titration

had been reached and exceeded to some extent.

40
THE EFFECT or SUGARS ON THE COAGULATION 0F SQUASH SEED
GLOBULIN

A. Coagulation by Hydrochloric Acid:

To study the effect of sugars on the coagulation of
squash seed globulin by hydrochloric acid, the following
procedure was drawn up using Protein III, containing 1.67
mg. of nitrogen per m1., and in.a solution 10% in respect
to sodium chloride. To 4.5 ml. of the protein solution
were added 5 ml. of sugar solution and the mixture held
at 37° in a water bath for 15 minutes in order to allow
any combinations between sugar and protein to take place.
Four sugar solutions were used: 2M D-glucose, M D-glucose,
M D-galactose, and M D-levulose, all of which were made
10% in respect to sodium chloride in order to prevent
protein.precipitation as crystals. A11 sugars were of
Pfanstiehl C. P. anhydrous quality from the following
lots: D(+)glucose, 'P‘u #1655; D(-)1evulose, -"": #
1675; D(+)ga1actose, -e-u #1601. Coagulation of the
protein was effected by adding 0.5 ml. of M hydrochloric
acid and the solution allowed to stand 2 minutes at 37°.
*The coagulated protein was removed by filtration, using
ordinary qualitative filter paper, and the amount of
coagulation was determined by analyzing the filtrate for
nitrogen by MicrokJeldahl method (33). All values in

41
Table III represent the average nitrogen value of tripli-
cate determinations. A control solution containing no
protein was run on each sugar solution and a correction
made for nitrogen from this source. The pH of all solu-

tions was found to be 1.2 on the Beckman.pH meter.

TABLE.III
Effect of sugars on the coagulation of Equash seed glob-
ulin.by hydrochloric acid,\using"2 ml. of the filtrate
for nitrogen determinations. Original nitrogen in 2
m1; of the filtrate =,l.50 mg.

Mg. H in 2 % of original
Sugar Conc. ml. of Filtrate H in ziltrate
M D-glucose . 0.03 5.3
0.5M D-glucose 0.00 0.0
No D-glucose 0.05 3.3
M D-galactose 0.05 3.3
No D-galactose 0.05 3.3
n D-levulose 0.00 0.0

Ho D-levulese 0.05 3.3

42
B. Coagulation by Heat:

Protein III was again used to study the effect of the
same sugars at the same concentrations on the heat coagu-
latioh.of squash seed globulin. A phthalate - sodium
hydroxide buffer of pH 6.0 WSS‘UBBd to establish the pH of
the solutions. To 4.5 m1. of the protein solution were
added 5 ml. of sugar solution, 0.5 m1. of buffer solution
pH 6.0, and the mixture held for 30 minutes at 80° in.a
water bath. The coagulated protein was filtered off, using
quantitative filter paper, and a 2 m1. portion of the
filtrate used for nitrogen determinations by Microkjeldahl
method (33). Determinations were run in triplicate and
correction was made for the nitrogen contributed by the
sugar and by the reagents. The results are given in Table
IV.

The coagulate from heat denaturation differed from
that formed by denaturation with hydrochloric acid in that
the former was less flocculent, and left a turbid filtrate
that could neither be filtered off by quantitative filter
paper, nor centrifuged off in 60 minutes.

43
TABLE IV
Effect of sugars on the coagulation of squash seed glObu-
lin by heating 30 minutes at 80°, and at a pH of 6.0.
Original nitrogen content of 2 ml. of the filtrate used
in nitrogen determinations - 1.50 mg.

Mg. H in 2 % of original
Sugar Conc, m1. of filtrate N in filtrate
M D-glucose 1.13 75.3
0.5M D-glucose 1.09 72.7
No D-glucose 1.16 77.3
M D-galactose 1.03 68.7
No D-galactose 1.16 77.3
M D-levulsse 1.21 80.7

No D-levulose 1.16 77.3

44
From Table IV it was concluded that at 80° coagu-
lation was not sufficiently complete to study the effect
of the sugars on coagulation of the globulin. Further-
more, the presence of the buffer solution was thought to
have an effect on the coagulation, and therefore was
omitted in the next experiment, in which the procedure
is essentially the same as the preceding one. Centrifu-
gation of the coagulated protein mixture was substituted
for filtration with quantitative paper in order to slim-
inate any error from adsorption on the paper, although
the filtrates in both cases were almost the same in nitro-
gen content as revealed by Microkjeldahl determinations
on identical solutions. To 5 ml. of the Protein III
solution were added 5 ml. of the sugar solution, and the
mixture placed in a water bath. The bath was allowed to
heat up slowly from 25°, taking 5 minutes to reach boiling
temperature, and held at 100° for 15 minutes. According
to Osborne (1) this slow rise of temperature enhances the
coagulation of the globulin at lower temperatules during
a given period of time. The mixture was centrifuged for
5 minutes after coagulation, and 2 ml. of the supernatant
liquid used for nitrogen determinations by MicrokJeldahl
method. The results shown in Table V represent the average

of triplicate determinations at each sugar concentration.

1*5

TABLE V
Effect of sugars on the coagulation of squash seed globu-
lin by heating at 1000 for 15 minutes. Original nitrogen
content in 2 ml. of the filtrate used for nitrogen deter-

minations - 1.67 mg.

 

Mg. N in 2 % of original
Sugar Cone, ml of Filtrate N in filtrate
H D-glucose 1.08 6#.7
0.5M D-glucose 0.99 59.3
No D-glucose 1.04 62.3
I D-galactose 0.91 5#.5
0.5M D-galactose 1.08 64.7
No D-galactose 1.04 62.3
M D-levulose 0.88 52.7
0.53 D-levulose 0.87 52.1
No D-levulose 1.0“ 62.3

45
Because of an apparent difference in the turbidity

of the supernatant liquids from the solutions used in
Table V, the percentage transmission of each solution was
determined in the Junior Coleman photometer at 550 mu
wavelength. To complete Table VI, readings were also made
on 0.25! and 0.1M solutions of the same sugars, as well '
as solutions containing no sugar, under identical proce-
dures of denaturation and centrifugation. The instrument
was set to read 100% transmission with a control solution

containing 10 ml. of 10% sodium chloride solution.

TABLE VI
Percentage transmission of the supernatants from squash
seed globulin solutions containing 9.79 mg. of protein
'per m1., heated at 100° for 15 minutes. Readings made
on the Junior Coleman.phot0meter at 550 mu wavelength.

Total volume of each solution = 10 ml.

 

 

CONCENTRATIONS
.§28§£ 311‘ SLILJ! SLEELHL 0'1 M 4§Egg£
D-glucose 38.5 24.2 15.3 7.0 7.5
D-galactose 51.3 #2.0 11.8 8.5 7.5

D¥levulose 37.5 19.5 16.0 7.5 7.5

“7
DISCUSSION

with respect to the methods employed in this problem
for the liberation and estimation of sulfhydryl groups,
the protein squash seed globulin did not produce sulf-
hydryl groups in an active, exposed form on denaturation
by urea, guanidine hydrochloride, sodium lauryl sulfate,
hydrochloric acid, or by heat. Neither was there evidence
of the presence of sulfhydryl groups in the native protein
when the same methods of estimation were used. According
to the protein classifications set forth by Neurath,
Greenstein, Putnam, and Erickson (16) this characteristic
would put squash seed globulin in that class of proteins,
such as amandin and insulin, which neither in the native
state nor after treatment with concentrated guanidine
hydrochloride give tests for sulfhydryl. Although there
is a similarity between squash seed globulin and a-andin
in that they are both seed globulins, two other seed glob-
ulins, edestin and exoelsin, belong to that class of pro-
teins which in the native state do not give tests for
sulfhydryl, but do so after treatment with concentrated
guanidine hydrochloride (16). However, as Gortner points
out (38), the classification of proteins as globulins
represents a very poorly defined group.

Hess and Sullivan (11) concluded from their work on

squash seed globulin that cysteine complexes were present

#8
in the native protein, and substantiated their claim by
reporting values for the iodine titration of the native
protein according to the Okuda method for cysteine. An
attempt to duplicate these values in the problem failed
because of the persistent appearance of a flocculent
protein precipitate on the addition of the acid necessary
to carry out the titration. Hess and Sullivan stated that
aqueous or dilute salt solutions of the proteins were used
in their experiments, and from this it was assumed that
a sodium chloride solution was used, since the protein is
practically insoluble in water. The sodium chloride
content of the solutions was varied over a wide range,
but with no prevention of precipitation on the addition of
hydrochloric acid. A high value for cysteine was obtained
by omitting the required acid, but eXperiments with cysteine
hydrochloride pointed out that the conditions of the titra-
tion must be followed closely in order to secure accurate
results.

If cysteine complexes are present in squash seed
globulin, then this protein can be placed in the same
classification as edestin and excelsin, in which it was
assumed that the cystine-cysteine sulfur was composed of
cystine sulfur plus cysteine sulfur (16). This grouping
is not in agreement with the previous grouping of squash
seed globulin with amandin and insulin, in which the

cystine-cysteine sulfur is composed entirely of cystine

1&9
sulfur (16). It follows then that the methods of liberating
and estimating sulfhydryl groups of squash seed globulin
in this problem would be considered inadequate if the
grouping with edestin and excelsin is accepted, unless
the inherent properties of the protein are such that liber-
ated sulfhydryl groups are not produced in an active form
on denaturation with the above mentioned agents.

The method prqposed in this work for the liberation
and estimation of sulfhydryl groups from squash seed glob-
ulin was found to be effective for cysteine hydrochloride
and for egg albumin solutions, but gave negative results
on the globulin itself. The presence of precipitated
protein during the exposure of the globulin to ferricyanide
was thought to prevent the liberation of sulfhydryl groups.
However, on elimination of this precipitate by omitting
the addition of hydrochloric acid and using buffer solu-
tions to establish different levels of acidity for sulf-
hydryl oxidation, no Prussian blue was developed. The
time period of exposure of the protein to the denaturing
agents was increased to two hours with no change in results.

The appearance of Prussian blue occluded to the pre-
cipitated protein in the experiments with excess protein
in the solid state came as a result of long time standing
of the precipitated.protein in the presence of sodium
chloride, urea, ferricyanide, and ferric sulfate. Since

it takes only one minute for any reactive sulfhydryl groups

50
to reduce ferricyanide (2%), and since the greater part of
PrusSian blue deve10pment is achieved in five minutes from
the time of adding ferric sulfate, this subsequent color
was presumed to be the result of ferricyanide reduction by
other than sulfhydryl groups. Furthermore, an increase in
the time of exposure to the denaturing agent to 15 hours
produced no difference in the length of time elapsed before
the appearance of Prussian blue on the precipitate, which
showed that sulfhydryl groups were not liberated by long
time standing of the protein in contact with the denaturant.

In those solutions in which acidification of the final
mixture brought about a development of Prussian blue, the
color was attributed to the existence of protein and ferri-
cyanide in a neutral or slightly alkaline medium, caused
by the presence of urea in the solutiOn (35), which would
enhance the ability of non-sulfhydryl groups on the protein
to reduce ferricyanide (25). In order to confirm this
reduction of ferricyanide by non-sulfhydryl groups, prelim-
inary experiments were run on.squash seed globulin using
hydrochloric acid as a denaturant. On completion of the
denaturation of the protein, ferricyanide was added and the
mixture brought up to neutrality by the addition of M
sodium hydroxide and phosphate buffer solution of pH 7.0.
Four drOps of M sodium hydroxide were added and the mix-
ture held at room temperature for 5 minutes before acidi-

fying with 6 drOps of M hydrochloric acid and adding ferric

51
sulfate. The presence of ferricyanide in contact with the
protein in alkaline medium produced a deep Prussian blue
color on acidification of the final mixture.

When sodium chloride was omitted from the procedure
by adding the desired reagents to the dried protein, a
similar appearance of Prussian.blue occluded to the precip-
itated.protein was noticed after long time standing of the
final mixture. If any of the protein had gone into solu-
tion, this would indicate that the presence of sodium
chloride alone did not inhibit the production of sulfhydryl
groups from squash seed globulin. 0n the other hand, if
none of the protein had dissolved, the presence of Prussian
blue in the mixture could definitely be ascribed to reduc-
tion of ferricyanide by other than sulfhydryl groups.
Lastly, the preliminary experiments with cysteine hydro-
chloride showed that the presence of sodium chloride did
not inhibit the oxidation of active sulfhydryl groups by
ferricyanide.

The comparative results of the nitr0prusside test
on squash seed globulin, egg albumin, and cysteine hydro-
chloride confirmed those of the ferricyanide reaction in
that no sulfhydryl groups were liberated from squash seed
globulin when denatured by the above mentioned agents.

In the preliminary experiments on the denaturation
of squash seed globulin by urea, guanidine hydrochloride,

and sodium lauryl sulfate, the criterion of denaturation

52
was taken as the appearance of a gummy, insoluble precip-
itate on dilution of the denatured protein solution with
water, and the subsequent failure of this precipitate to
dissolve in 10% sodium chloride solution. This state of
denaturation was reached by holding the sodium chloride
solutions of the protein in contact with the denaturant
for 15 minutes. No investigation was completed on the
minimal concentration of denaturant necessary to effect
denaturation. In the case of hydrochloric acid, the fail-
ure of the acid precipitated protein to dissolve in alkaline
medium was taken as the criterion of denaturation. This
condition was attained by holding the acid precipitated
protein at room temperature for approximately 15 minutes.
The appearance of a gummy coagulate when squash seed globu-
lin solutions were heated at 1000 for 15 minutes was con-
sidered a reaction of denaturation.

D-glucose, D-galactose, and D-levulose appeared to
have little or no effect on the coagulation of squash
seed globulin by hydrochloric acid, when the degree of
coagulation was indicated by the relative amount of nitro-
gen remaining in the filtrate. The results in Table III
are apt to show an inhibition of protein coagulation in
the presence of I glucose and an increase in the presence
of 0.5M glucose. However, in view of the completeness
of coagulation and the resulting low nitrogen content of
the filtrates, these results are misleading. A differ-

ence of 0.03 mg. in the amount of nitrogen in 2 ml. of the

53
filtrate corresponds to an acid titration value of 0.10

ml. This value is quite insignificant when consideration
is given to the amount of acid necessary to correct for the
reagents themselves and for the nitrogen contributed by
the sugar. These amounts were respectively 0.20 and 0.21
ml. Furthermore, the degree of experimental error is
greatly increased by the large amounts of sugar to be
digested in the licrokjeldahl method of determining nitro-
gen in the filtrates.

The effect of D-glucose, D-galactose, and D-levulose
on the heat coagulation of squash seed globulin at 800
and at a.pH of 6.0 appeared to be small or non-existent.
The use of qualitative filter paper for the filtration of
the coagulated protein solution in this experiment and in
the experiment with hydrochloric acid coagulation might
have accounted for a certain amount of error in results.
In the experiment with heat coagulation at 100°, any error
due to adsorption of the protein on the filter'paper or
to incomplete filtration was eliminated by centrifugation
of the coagulated mixture.

The effect of D-glucoee, D-galactose, and D-levulose
on the heat coagulation of squash seed globulin at 1000
in an unbuffered solution was found to be the same as at
80°. The variation in percentages of original nitrogen
in the supernatant solutions after coagulation was suf-

ficiently small to be attributed to the experimental error

5n

encountered in the Hicrokjeldahl method of determining
nitrogen. However, the differences in percentage trans-
mission of the filtrates from various concentrations of
each sugar indicated that the presence of sugars during
the heat coagulation of squash seed globulin tended to
keep the protein in a more dispersed state. This infer-
ence was based on the observation that the amount of
nitrogen in each supernatant solution, as determined by
Microkjeldahl analysis, was approximately the same.

The apparent anomaly between the results Obtained
in the nitrogen determinations of the supernatant solu-
tion and those obtained by percentage transmission read-
ings might be explained as due to the finite difference
in the degree of diapersion of the protein as influenced

by sugars.

55

CONCLUSIONS

Sulfhydryl groups, as estimated by the ferricyanide
method, are not liberated from squash seed globulin

in an active form by urea, guanidine hydrochloride,
sodium lauryl sulfate, hydrochloric acid, or heat.

0n the basis of solubility tests, squash seed globu-
lin can be irreversibly denatured by urea, guanidine
hydrochloride, sodium lauryl sulfate, hydrochloric

acid, and heat.

D-glucose, D-galactose, and D-levulose have no effect
on the coagulation of squash seed globulin by hydro-
chloric acid or by heat.

The degree of diapersion.of squash seed globulin on
heating in sodium chloride solution is intensified
by the presence of D-glucoss, D-galactose, and

D-levulose.

10.

BIBLIOGRAPHY

Osborne, T. B., The Vegetable Proteins, 2nd ed., pp.
51-67. Longmans, Green & 00., New York (1924).
Barbieri, J., Ueber die Eiweissubstanz der Kurbissamen

J. prakt. Chem., 1g, 102 (1878).

8

Grubler, 0., Usher ein Krystallinisches Eiweiss der
Kurbissamen. J. prakt. Chem., g1, 97 (1881).

Chittenden, R. H., and Hartwell, J. A., Crystalline
Globulin and Globuloses, or Vitelloses. J.
Physiol., ;_1_, 435 (1890).

Osborne, T. B., Crystallyzed Vegetable Proteids.
Am. Chem. J.,.14, 662 (1892).

Krishnan, P. 8., and Krishnaswamy, T. K., Proteins
and other Nitrogenous Constituents of Iater
Melon Seeds. Biochem. J., 11, 1284 (1939).

Vickery, H. B., Smith, E. L., and Nolan, L. 8., A
Substitute for edestin. Science,‘22, 317 (1940).

Vickery, H. B., Smith, E. L., Hubbell, R. B., and
Nolan, L. 8., Isolation of Cucurbit Seed Glob-
ulins. J. Biol. Chem., liq, 613 (1941).

Jones, D. B., and Gersdorff, C. E. F., Proteins of
the Cantaloupe Seed, Cucumis‘mglg, J. Biol.
Chem.,.§§, 79 (1923)-

Smith, I. L., Greene, R. D., and Bartner, E., The
Amino Acid Composition of Seed Globulins. J.
3101. Chem., 1_6_lg, 159 (1946).

11.

12.

13.

.14.

15.

16.

17.

18.

57
Hess, W. 0., and Sullivan, M. 1., The cysteine,

Cystine, and Methionine Content of Proteins.
J. Biol. Chem., _1_5_l_, 635 (1943).

Hess, W. 0., and Sullivan, M. X., The Determination
of Phenylalanine in Proteins. Arch. Biochem.,
.5. 165 (1944).

Smith, E. L., and Greene, R. D., Further Studies on
the Amino Acid Composition of Seed Globulins.
J. Biol. Chem., lél. S33 (191m.

Osborne, T. B., and Harris, I. F., The Specific
Rotation of Some Vegetable Proteins. J. Am.
Chem. Soc.,.g5, 842 (1903).

Kizel, A., and Jurkevitsch, V., Isoelectric Points
of Acetylated and Deaminated Cucurbitin.
Biokhimiya,Ԥ, 276 (1941). Brit. Chem. Abstracts,
A III, 921, (1941).

Neurath, B., Greenstein, J. P., Putnam, F. W., and
Erickson, J. 0., The Chemistry of Protein
Denaturation. Chem. Rev.,‘14, 157 (1944).

Duddles, l. J., The Effect of Sugars and Hannitol
on the Coagulation of Egg Albumin. Thesis for
the Degree of I. 8., Michigan State College,
East Lansing, Mich., (1932).

Hardt, C. B., Sulfhydryl Groups and the Coagulation
of Proteins as influenced by Reducing Sugars.
Thesis for the Degree of M. 8., Michigan State
College, East Lansing, Mich., (1941).

19.

20.

21.

22.

2}.

24.

25.

26.

27.

58‘

Mirsky, A. E., and Anson, M. L., Sulfhydryl and
Disulfide Groups of Proteins. J. Gen. Physiol.,
12. 1#27 (1936).

Mirsky, A. E., and Anson, H. L., Sulfhydryl and
Disulfide Groups of Proteins. J. Gen. Physiol.,
22» 439 (1935)-

Mirsky, A. E., and Anson, M. L., The Reducing
Groups of Proteins. J. Gen. Physiol., l9,
451 (1916).

Anson, M. L., The Reactions of Denatured Egg Albu-
min with Ferricyanide. J. Gen. Physiol., 21,
247 (1939-1940).

Anson, M. L., Reactions of Iodine and Iodoacet-
amide with Native Egg Albumin. J. Gen. Physiol.,
31; 321 (1939-1940).

Anson, M. L., The Sulfhydryl Groups of Egg Albumin.
J. Gen. Physiol., 21. 399 (1940-1941).

Hirsky, A. E., Sulfhydryl Groups of Egg Albumin
in Different Denaturing Agents. J. Gen. Physiol.,
25,, 709 (19110-1941).

Beilinnson, A., Thermostabilisation der Eiweiss-
13sungen mit Rohrzucker und Glycerin. Biochem.
2°: fill» 399 (1929)-

Newton, B., and Brown, I. B., Frost Preclpitation

of Proteins of Plant Juice. Can. J. Research,

5. 37 (1931) .

28.

29.

30.

31.

‘ 32.

33.

34.

35-

59

Brosteaux, J., and Eriksson-Quensel, I., Effects of
Drying on Proteins. Arch. Phys. Bio1.~, 121,

209 (1935). c. A., 12. 2997 (1936).

Ball, 0. D., Hardt, C. B., and Duddles, I. J., The
Influence of Sugars on the Formation of Sulf-
Hydryl Groups in Heat Denaturation and Heat
Coagulation of Egg Albumin. J. Biol. Chem,,
1.5.1. 163 (1943).

Hardt, C. B., Huddleson, I. P., and Ball, 0. D.,

An Electr0phoretic Analysis of Changes produced
in Blood Serum and Plasma Proteins by Heat in
the Presence of Sugars. J. Biol. Chem., 1:55
211 (1946).

Fischer, R., Prevention of Heat Coagulation of Serum
Proteins by Sugar. Experientia,‘3, 29 (1947).

Mason, H. L., A study of Glutathione. J. Biol.
Chem., go, 623 (1930).

Microkjeldahl Nitrogen Method, J. Assoc. Offic. Agr.
Chemists, 97-100 (1941).

Burk, n. F., Effects of Inorganic Electrolytes on
the Liberation of Sulfhydryl Groups in Proteins.
J. Phys. Chem., 41, 104 (1943).

Bawden, F. 0., and Pirie, N. I., The inactivation of
Some Plant Viruses by Urea. Biochem. J.,,fifi,
125a (1940).

60

36. Morrow, C. A., ‘Bigghgmical Laboratory Methods, p. 99.
John Wiley & Sons, Inc., New York (1927).

37. Folin, 0., and Malmros, H., An improved Form of
Folin's Micro Method for Blood Sugar Determin-
ations. J. Biol. Chem.,‘éfi, 115 (1929).

38. Gortner, R. A., Outlines g£_Biochemist;y, 2nd ed.,
pp. 417-424. John_Wiley & Sons, Inc., New
York (1938).

 

 

 

 

 

 

 

,11 "MWIHY 'UUWE‘.

fr612.015
5 H341 Harvey

1943 63

. \|\\|\\\\\|\\\\II\“I“HM“\|\|\WW\lHWHUIHN

 

3 1293 024467