A SEMI-MICRO DETERMINATION EOR PLANT STEROLS

Dick Waghorne

A THESIS
Submitted to the Graduate School of Michigan
State College of Agriculture and Applied
Science in p a r tia l fulfilm ent of the
requirements for the degree of
DOCTOR GE PHILOSOPHY
Department of Chemistry
1951

P roQ uest N um ber: 10008446

All rights reserved
IN FO R M A T IO N T O A LL U SER S
The q u a lity o f this reproduction is d e p e n d e n t upon the qua lity o f the copy subm itted.
In the unlikely even t th a t the a u th o r did not send a com plete m anuscrip t
and th e re are m issing pages, these w ill be noted. Also, if m aterial had to be rem oved,
a note w ill indicate the deletion.

uest
P roQ uest 10008446
P ublished by P roQ uest LLC (2016). C opyright of the D issertation is held by the A uthor.
All rights reserved.
This w o rk is protected aga inst unauthorized copying under Title 17, U nited S tates C ode
M icroform Edition © P roQ uest LLC.
P roQ uest LLC.
789 E ast E isenh ow er P arkw ay
P.O. Box 1346
A nn A rbor, Ml 4 8 1 0 6 - 1 3 4 6

I have appreciated the helpful assistance
of Professor C* D. Ball

CONTENTS

I.

Introduction ................................................................................1

II.

H is to ric a l

III*

Experimental ............................................................................ 9

................................................................................. 2

A.

Preparation of E sters

B.

Use of the Nephelometer .....................................16

C.

Hydrolysis of the Digitonide
followed by the Determination
of the Liberated ReducingSugars . . . .

D.

................

Oxidation of the Digitonide

IV.

Application of the Method

V.

Conclusions

Bibliography ..................

..................

11

25

..........................43

............................

71
79
81

1

I#

INTRODUCTION
The increasing in te re s t in the use of phyto-

s te r o ls as precursors for the synthesis of
hormones and a n tir a c h itic vitamins as well as
th e ir expanding use in other commercial f ie ld s ,
seemed to indicate th at a study of the development
of these s te ro ls in plants would be profitable#
In order to make such a study the need f o r a
method applicable to the estim ation of s te ro ls in
general, to include both saturated and unsaturated,
was very evident #
A survey of the lite r a tu r e revealed th at
there was no such method available#

The gravi­

metric methods lacking the required s e n s itiv ity
and the colorim etric methods lacking in th e ir
general ap p licab ility #
A determination of this type must meet two
primary requirements*

F ir s tly ,

the s te ro ls must

be separated from contaminating m aterials

on a

semi-micro basis and, secondly, the e n tire
quantity of s te ro ls

separated, without regard fo r

minor changes in s tru c tu re , must be accurately
estimated#

2

II.

HISTORICAL
Winda us (1), in 1909, showed th a t

ch o lestero l farmed a complex with d ig ito n in
which was insoluble in 95% ethanol*

Since th at

time the iso la tio n and p u rific a tio n of n atu rally
occurring ste ro ls has la rg e ly depended on th is
reaction*

In a l l cases stu died , whether of

animal or plant origin , the complex appears to
be formed from equimolecular amounts of dig iton in
and stero l*

Every s te r o l, with the exception of

calostexol from the juice of C alotrcpis gigantia
(2 ),

iso la te d from natural sources foims such a

complex and they a l l have lim ited s o l u b i l i t i e s .
In a c r i t i c a l investigation of th is pro­
cedure Schoenheimer and Dam (3) found that frou
95% ethanol the p re c ip ita tio n was not absolutely
q uan titativ e but varied according to the excess
d igito nin present*

In 1937, i t was reported by

Sperry (4) th a t Schoenhe imer had overcome th is
weakness of the method by the use of 80% ethanol,
in which medium the p re c ip ita tio n was q u an tita­
tiv e regardless of the excess of digitonin
present#
Many other

p re c ip ita tin g mediums have been

3

proposed, such as acetone
chloroform (6) and e th e r

(5), alcohol and
(7) e tc .

but the above

mentioned 80% ethanol appears to be the most
s a tis f a c to r y .
More recen tly Sobel e t a l

(8, 9) have

developed a method of p re c ip ita tin g ch o lestero l
by the use of pyridine and chlorosulfcnic acid,
re s u ltin g in the formation of an insoluble
pyridinium s u lf a te .

The value of th is method

for the p re c ip ita tio n of mixed s te ro ls of plant
origin has not as yet been in v estig ated .

Indeed,

whether or not a l l s te ro ls form an insoluble
cduplex under the conditions outlined has not
been ascertain ed .
Another p o s s ib ility fo r the p u rific a tio n of
the plant s te ro ls

lie s

graphic separation.

in the fie ld of ehromato**

To the present time the

work in th is fie ld has been lim ited and is ably
reviewed by Zechmeister (10).

The main lin e

of

approach has been the iso la tio n of one p a rtic u la r
s te r o l from a l l others or the separation of
several s te ro ls from a p u rified mixture.
Ladenberg(ll) made use of the colored
azobenzene-p*carboxylic e s te rs of the s te ro ls

4

and c a r r i e d
However, Ott
that

out t h e i r sep ara tion on alumina*
(12)

states,

the s e p a r a b i l i t y

"It

is new conceded

of s t e r o l s

on alumina

columns depends primarily upon d i f f e r e n c e s
between the number of double bonds in the
molecule, and not upon the p o s i t i o n of the
double bonds or the

structure

Wall and Kelley

of the

side chain."

(13) varied t h i s

approach

in t h a t they adsorbed i n t e r f e r i n g substances on
a magnesium oxide and Super-Cel column and ob­
tained the p a r t i a l l y p u r i f i e d s t e r o l s in the e l u a t e .
The chlorophylls and carotenoids, with the excep­
tion

of the carotenes,

this

p a r t i a l l y p u r i f i e d eluate i t

possible

were thus removed.

to p r e c i p i t a t e

From

was found

the s t e r o l s as the

dig iton id es and t o subsequently remove the

other

l i p i d s by washing with e t h e r .
Assuming the separation of the
contaminating substances there s t i l l
problem of t h e i r q u a n t i t a t i v e
this

sterols

from

remains the

estimation.

For

purpose many methods have been t r i e d with

g re a t e r or l e s s e r

success.

A number of colorimetric reactio n s have
been reported for the

s t e r o l s which have been

5

used for preliminary id e n tif ic a tio n and some for
q u an titativ e estimation*

The Salkowski re a c tio n

(14) depends on the color formed when a chloro­
form solution is layered with concentrated
su lfu ric acid and v a riatio n s of th is reactio n
using an alcoholic solution of the s te r o l (15)
and using the alcoholic s te r o l solution to which
an aldehyde has been added (16, 17)*
Bosenheim t e s t

The

(18) d iffe rs from these in th a t

90% tric h lo ra c e tic acid solution is used with
the free s te r o l or with a chloroform solution of
the stero l*

None of these reactions has been

adopted fo r q u an titativ e work.
In 1932, Bernoulli (19) determined
cholesterol by measuring the red color which
developed when a g la c ia l acetic acid solution
was heated with acetyl chloride and anhydrous
zinc chloride*

Thornton (20) made a c r i t i c a l

examination of th is reaction and found th at
there was a c h a ra c te ris tic transmission maxima
a t 440 mu fo r various s te ro ls but th a t the
extinction co efficien t was d iffe re n t fo r the
various s te r o ls .

I t is ,

th erefo re, apparent

that the use of th is method for the

6

determination of a mixture must be based on an
exact knowledge of the r a t i o of the various
s te ro ls present*
The Liebexmann - Burcbard reactio n (21* 22)
r e s u ltin g in the development of a blue color on
the addition of acetic

anhydride and su lfu ric

acid to solutions of s te ro ls has been the most
widely investigated of the color reactions#

In

1934, Schoenheimer and Sperry (7) developed th is
reactio n to the point where i t has been almost
so lely used for the estim ation of ch o lestero l in
animal tiss u e s and fluids#

Attempts have been

made to adapt th is reaction to plant ste ro ls
(13,

23) but again i t has been found that

d iffe re n t s te ro ls develop d iffe re n t colors#
Both the above authors used an empirical approach
to the problem by iso la tin g macro amounts of the
s te r o l mixture from a p a rtic u la r plant species
and using th is iso la te d m aterial for the
preparation of a working curve.

However, i t was

s t i l l necessary to make the assumption that the
r a t i o of the amounts of the various ste ro ls
present was the

same in a l l cases.

None of the color reactions reported is

7

applicable to saturated s te r o ls and therefore
they cannot be used fo r the q u an titativ e
estim ation of th is f ra c tio n .
Rappaport and Klapholz (24) p re c ip ita te d
the c h o lestero l by use of a standard solution
of d ig ito n in and then determined the excess of
d ig ito n in present by hydrolysing with su lfu ric
acid and q u an titativ ely estim ating the reducing
sugars lib e ra te d .
Windaus (1),

in h is o rig in a l method,

iso la te d the cholesterol d ig iton id e, dried and
weighed i t .

The amount of cholesterol present

was then determined by multiplying by an
appropriate fa c to r, depending on the r e la tiv e
molecular weights of ch olestero l and d ig ito n in ,
the factor for cholesterol being 0*242.

This

procedure has been adapted for use with plant
s te r o ls by Wall and Kelley (13).

The facto r

used in th is case was 0*253 based on the average
molecular weight of the various plant sterols*
By such a means the saturated and unsaturated
s te ro ls can be estimated together,

then the

saturated s te ro ls can be determined through the
removal of the unsaturated ones by the method

8

of Anderson and Nabenhauer (13, S3).

By

difference the unsaturated s te ro ls can be e s t i ­
mated*
Muhlbock and his coworkers (£6) estimated
ch o lestero l by is o la tin g the d igito nid e,
redissolving i t

in a mixture of methyl and ethyl

alcohol and re p re c ip ita tin g i t

by the addition

of water and recording the tu rb id ity of the
re s u ltin g m ixture•
Szent-Gyorgyi (£7), in 19£3, and Okey (£8),
in 1930, developed methods for the q u an titativ e
estim ation of cholesterol which depended on the
diehrornate oxidation of the digitonide followed
by an iodometric t i t r a t i o n
die hr ornate*

of the excess

These methods approached the

required s e n s itiv ity but were time consuming and
somewhat lacking in precision*

Later attempts

by various workers (6, £9, 30, 31) resu lted in
only moderate improvement*

MacLachlan (3£)

d ire c tly extrapolated Boyd’s oxidative method to
germinating soybeans, but estimated th a t the
r e s u lts were probably about 10$ low*

9

III.

EXPERIMENTAL
I t was thought th a t an in v estig atio n at

four d iffe re n t types of procedures might r e s u lt
in an adequate method*
1*

Some reaction involving the hydroxyl group,
which is common to a l l the s te r o ls , as the
reac tiv e centre;

e . g . - preparation of an

e s te r followed by hydrolysis and determina­
tio n of the
2*

lib e ra te d acid*

The use of nephelometry to determine the
ra te and density of the p re c ip ita t ion of the
digitonides*

3*

Hydrolysis of the digitonide followed by the
estimation of the lib e rated reducing sugars.

4*

Total oxidation of the digitonide.
Preliminary studies were made on these

methods in the hope tb a t one or more might yield
encouraging resu lts*
Throughout the work the following s te ro ls
were used as standard prep aratio n s:1.

Cholesterol - Difco Laboratories Cholesterol;
M.F. 148-149°C used as
purchased*

2*

Stigm asterol - prepared frcm crude soybean

10

s te r o ls and re c ry s ta lliz e d a
to t a l of fourteen times as the
acetate and free s te r o l; M.F*
166-167°C•
3*

S ito s te r o l -

N u tritio n al Biochemicals
Corporation S ito s te ro l,
re c ry s ta lliz e d once from
absolute ethanol; M*P*
136♦5-137*5°C*

4.

Corn s te ro ls - extracted from a sample of
"soap stock", obtained from
the Canada Starch Co*, with
hexane, evaporating the e x tra c t
to dryness and re c ry s ta lliz in g
twice from aqueous alcohol;
M#P. 137*5-139 °C•

11

A.

Preparation of E sters
Acetylation - Because of the ease of

removing the excess reagent, acety l chloride was
used throughout as the

ac ety la tin g reagent*

The

isomerism re su ltin g from the lib e ra te d hydro­
chloric acid was disregarded because only the
t o t a l s te ro ls were to be determined.
to id en tify

No attempt

individual ste ro ls was an ticip ated .

The standard methods of determining acety l
groups were tr ie d but i t was found that creeping
of the reagents occurred in the usual apparatus
with acid creeping mechanically into the
receiving fla sk during d i s t i l l a t i o n .

Also the

d ire c t saponification of the e s te r with alcoholic
a lk a li followed by t i t r a t i o n
a lk a li was u n sa tisfa c to ry .

of the excess
In the l a t t e r case

no s a tis fa c to ry end point could be reached because
there appeared to be a gradual, e r r a tic release
of a lk a li over a period of several days.
The most sa tis fa c to ry r e s u lts using th is
method were obtained as follows:Samples containing zero, twenty-five and
f i f t y mgoa. of cholesterol were weighed into short
necked micro Kjeldahl flasks f i t t e d with 24/40

standard taper necks.
was

Two ml. of ac ety l chloride

added to each sample and they were then

refluxed for one hour.

The excess acety l

chloride was removed by means of a water
a s p ira to r a t room temperature followed by drying
in a vacuum over a t 80°C for twenty-four hours.
The aeetylated samples were then refluxed
for one hour with f i f t y percent su lfu ric

acid

and the lib e ra te d acetic acid was d i s t i l l e d ,
using a sp ecially designed d i s t i l l a t i o n head,
Figure 1*

This equipment was designed to prevent

the creeping of the

su lfu ric acid which occurs

in the standard apparatus.
ml. of d i s t i l l a t e was

Approximately f i f t y

collected and i t was

immediately t i t r a t e d w ith 0.5 N sodium hydroxide
using phenol red as an in tern al in d ic a to r.
Typical re s u lts obtained by th is method are
ill u s tr a te d in Table 1.
As can be seen frcm these figures the
r e s u lts are variable and are not adequately
reproducible even when using r e la tiv e ly large
samples.

I t was not believed feasib le to improve

both the precision and s e n s itiv ity of such a
method for the required use*

2 4 /4 0

S

STEAM
\lN L E T

APPARATUS FOR DISTILLJNG
TITRATION

FIG. I

HAc FOR

14

TABLE 1

CHOLESTEROL DETERMINED BY ACETYLATION AND TITRATION
OF LIBERATED ACETIC ACID

Mgm, Cholesterol
Present_______

Ml,

,05 N NaQH

Mg, Cholesterol
Found_______

0

0.06

0.0

25

1.26

23.2

50

2.28

42.8

0

0.06

0.0

25

1.28

23.3

50

2.81

51.9

0

0.04

0.0

25

0.58

11.2

50

1.40

27.1

I t was thought th a t i f a highly colored
e s te r could be prepared and iso la te d that a
colorim etric method might be developed which did
not depend on the r e la tiv e stru c tu re s of the
various s te r o ls .

However, attempts to prepare

and purify semi-micro amounts of the
p-phenylazobenzoyl e s te rs q u a n tita tiv e ly
according to Ladenbergfs procedure (11) f a ile d .
Attempts were also made to prepare the
glycyl e s te r of the s te ro ls with a view to
determining the nitrcgen present in the e s te r
by means of a micro Kjeldahl technique.
attempts were unsuccessful.

These

B.

Use of the Nephelometer
Preliminary in v estig atio n with cholesterol

indicated th a t the flo ccu latio n of the digitonide
occurred f a i r l y rapidly and th a t th is
flo c cu la tio n greatly affected the r e s u lts
obtained.

Search for a s ta b iliz e r resu lted in

the use of a gum g h a tti solution for th is purpose*
During the preliminary in vestig ation using puri­
fied solutions of s te ro ls the tu rb id ity was read
d ire c tly

on the p re c ip ita tin g d ig ito n id e.

Experimental
Preparation of stock solutions
Digitonin s o lu tio n :- 600 mgm. of d ig ito n in
(Merck) was

dissolved in BOO ml. of 95% ethanol

by heating, 800 ml. of d i s t i l l e d water was
added; to th is mixture 50 ml. of 2% gum g h atti
solution was added.

The whole solution was

thoroughly mixed.
S terol s o lu tio n s:- A solution of cholesterol was
prepared in 95% ethanol so th at B ml. would
contain 0*1, 0.2, 0.3, 0.4, 0.5 mgm. of the
c h o le ste ro l.

Similar solutions of erg o stero l

and s tig a a s te ro l were prepared.

Procedure
Two ml. of the s te r o l so lu tio n was

pipetted

into a nephelometer cuvette and fiv e ml. of
d ig ito n in solution added.

The mixture was

shaken vigorously fo r t h i r t y seconds and then
allowed to stand u n t i l read.

A ll readings were

made on a Coleman Photo-Nephelometer ♦
Results
Preliminary

studies

without the addition

of gum g h a tti resu lted in a rapid change in
nephelometer readings as shown in Table 2.

The

addition of the gxm gh atti resu lted in a much
more stable

condition over f a ir ly long time

in terv als as I s

illu s tr a te d in Table 3.

From the r e s u lts of Table 3 each of the
following samples was allowed to stand one hour
p rio r to reading.

Using th is

time the values

for ch o le stero l, ergosterol and stigm asterol are
to be found in Th ble 4 and represented graphi­
c a lly in Figure 2, using 0.1 mg. of ch o lestero l
as an instrumental standard.

Fran Table 4 and

Figure 2 i t can be read ily seen that the r a te ,
volume and probably the p a rtic le

size of the

p re c ip ita te s vary over a wide range.

18

TABLE 2

CHANGE IN NEPHELOMETER READINGS WITH TIME
NO GUM GHATTI

Mg. Cholesterol

Nepkelometer Readings
1 Min.

15 Min.

0.1

1.85

2.30

0.2

4.18

4.45

0.4

7.79

8.38

0.5

9.50

10.24

0.1*

2.00*

2.15

0.2

3.90

4.18

0.4

7.85

8.50

0.5

9.70

10.50

* Used to standardize instrument.

19

ol
col
CO!
oil

©

o
o
cO

Q
O
©

©
©

13
d

F
h

Ol
041

i
§
rH

9-*
e

E
h
E
h

©
.d

LO
o♦
to

o
o
•
to

lO
04
•
CO

o
o
•
o
rH

o
G
O
•

IO
(O
•
to

o
c•
to

o
to
•
c-

o
o
•
o

rH

o
04
•

rH

©
+*

to

lO
04
•

EH

°
col

to
CO
•

rH

rH
to
CO
•
CO

o
GO
•
to

o
04
•
o

o
o
•
o

to
CO
•
to

to
CO
•
to

to
CO
•
IN

o
o
•
o

eD*
to

to
00
•
to

to
04
*
tr­

*H

©

S25

c!j

°l
Oil

04
O*
•

rH
to

ml

*

rH

o
to
*
to

to
cO
*
to

rH

io
D•
o*

O
o
•
o
1—
1
o
o
♦
o

rH

+»
d

©

F
h

+»
©
d

©

IS
J
«d
<rH

F-4
©

O
m

©
+»
©

©

r—I

o
.d
o
ai

§

+»
©

rH

*
o

CM
•

o

to
«
o

•
o

*
in
•
o

o

+>

*d
©

©
r=>

20

TABUS 4
NEPHELOMETER READINGS FOR CHOLESTEROL,
ERGOSTEROL AND STIGMASTERDL
M«au S te ro l

Cholesterol

Ergosterol

Stigm asterol

0.1*

2.00*

1.28

1.85

0.2

4.52

2.78

3.81

to
•
o
*

Nephelometer Readings

6.72

4.50

5.88

0.4

9.24

5.92

8.05

0.5

12.54

7.50

10.46

Used to standardize the instrument for a l l
readings*

A = CHOLESTEROL
B = STIGMASTEROL

NEPHELOMETER

READINGS

ERGOSTEROL

MG.

STEROL

MEASUREMENT OF TURBIDITY USING O.l MG. OF
CHOLESTEROL AS STANDARD

FIG. 2

zz

In order to obviate the n ecessity of using
c h o le ste ro l as a standard for the other s te ro ls
these s e rie s were repeated using the 0*1 mgm.
le v e l of each s te r o l as the
fo r th a t p a rtic u la r s e r i e s •

instrum ental standard
These r e s u lts are

shown in Table 5 and fig u re 3.
I t becomes evident

from these r e s u lts th a t

such a determination might

be e n tire ly feasib le

for so lu tio n s of pure s te ro ls but that i f a
mixture of s te r o ls is present th a t an intimate
knowledge of the individual s te ro ls present
would be required*
I t would appear from these r e s u lts , however9
th a t the ra te and density of p re c ip ita tio n
increases with the decreasing of the number of
double bonds*

23

tabu; 5

NEPHELOMETER READINGS FOR CHOLESTEROL,
ERGGSTEROL AND STIGMASTEROL
Mam. S te ro l

Nephelometer Readings
Cholesterol

Ergosterol

St igraasteiol

0.1*

2.00*

2.00*

2.00*

0.2

4.32

4.26

4.11

0.3

6.72

6.65

6.33

0.4

9.24

9.06

8.52

0.5

12.54

11.39

10.82

* Instrument standardized by 0*1 mg. le v e l of
each s te r o l for i t s own s e r ie s .

A = CHOLESTEROL
ERGOSTEROL

NEPHELOMETER

READINGS

STIGMASTEROL

0.2
MG.

MEASUREMENT

0 .3
STERO L

OF TURBIDITY USING

EACH STE R O L AS ITS

0 .4

0.1 MG. OF

OWN STANDARD

FIG. 3

0 .5

25

C*

Hydrolysis of the Digitonide Followed by the
Determination of the Liberated Reducing
Sugars
Rappaport and Klapholz (24) p re c ip ita te d

c h o le ste ro l, hydrolysed the excess d ig ito n in in
the m other-liquor and then q u an tita tiv ely
determined the reducing sugars liberated*

It

was thought th a t the m other-liquor from the
p re c ip ita tio n of the s te ro ls from p a r tia lly
p u rifie d plant e x trac ts might contain a v ariable
amount of other substances capable of simulating
sugars in th e ir reducing a b i l i t y .

This is

e sp e c ia lly tru e when the excess d igiton in in the
mother-liquor is subjected to hydrolysis in the
presence of a strong acid
for a period of two hours*

(4.0 N s u lfu ric acid)
In order to overcome

th is d i f f i c u l t y the digitonide was iso la te d ,
p u rifie d ,

dried and then subjected to hydrolysis#

Preliminary

inv estig ation s indicated th at

i t was impossible to tra n s fe r the p re c ip ita te
from a centrifuge tube a f te r p u rific a tio n to a
fla s k for hydrolysis.

Due to th is fa ct a

centrifuge tube was desigied which permitted
c e n trifu g in g , washing, hydrolysis and d ilu tio n
to standard volume to be accomplished without

26

transfer.

This

was then found

tube i s

shown in Figure 4.

th at during

required f o r the hydrolysis

the active b o i l i n g
the p r e c i p i t a t e

was c a r r i e d up onto the walls
stuck t h e r e .

It

of the

tube and

In t h i s manner i t was removed from

the r eaction mixture and did not undergo
hydrolysis.

This was overcome by adding a few

drops of n-amyl alcohol to the tube.

As the

alcohol recondensed during hydrolysis i t
down the s i d e s of the
excessive f r o t h i n g .
that

washed

tube and also prevented
Another advantage found was

the s t e r o l and the aglucone of the d i g i t o n i n

were soluble

in the

alcohol layer whereas the

sugars l i b e r a t e d were soluble

in the water l a y e r .

Thus the completion of hydrolysis could be
determined by the disappearance of the
precipitat e .
The biggest d i f f i c u l t y
control of the

rate

of b o ilin g during h y d ro l y s is .

Many anti-bump agents,
carborundum,
fritted
success.

teflon,

g la s s ,

was found to be t h e

including boiling tubes,

beads,

small pieces of

e t c . , were t r i e d without

too much

The boiling tube was perhaps the most

successful and was used for most of t h is work.

27

f

IO 3 0 $

8 ML.

TUBE USED FOR CENTRIFUGING, WASHING,
HYDROLYSING, AND DILUTION TO VOLUME

FIG. 4

28

In addition to an adequate lack of control these
b o ilin g tubes were found hard to rinse free of
a l l so lu tio n .
Preparation of Reagents
Digitonin so lu tio n :- one grn. of digitonin
(Merck) was dissolved in one l i t r e of water and
then reduced in volume to 500 ml.
Cholesterol so lu tio n s:- solutions of
ch o lesterol in 95$ ethanol were prepared, con­
tainin g 0.25,

0.50, 0.75, 1.00, 1.25 mgm. per

five ml*
4 N Sulfuric acid.
4 N Sodium hydroxide.
n-Amyl alcohol ( r e d i s t i l l e d ) .
Alkaline copper and phosphomolybdic acid
solutions were prepared according to Folin and
W
u (33).
Exp e rim e n t a l

Five ml. aliquots of each of the solutions
of cholesterol were pipetted into individual
centrifuge tubes of special design and 2.5 ml.
of digitonin solution added.

The contents were

mixed and heated to coagulate the p re c ip itate
and l e f t

overnight to complete the p r e c ip ita tio n .

29

The following day the tubes were centrifuged and
the supernatant liq u id decanted.

The

p r e c ip ita te s were washed by centrifu g atio n and
decantation.

After washing,

one ml. of 4 N

su lfu ric acid was added and the tubes refluxed
fo r four hours;

Figure 4.

The contents were

n eu tralized by the addition of one ml. of 4 N
sodium, hydroxide and diluted to eight ml.

The

amyl alcohol layer was removed with a fine
c a p illa ry and suction and two ml. aliquots were
pipetted into Folin-Wu sugar tubes.
the alkaline copper reagent was

Four ml. of

added and the

tubes were then placed in a rapidly boiling
water bath for twenty minutes.

After heating

the tubes were cooled in running water without
a g ita tio n and four ml. of the phosphomolybdic
acid solution was added to each.

The contents

were mixed, allowed to stand u n t i l a l l evolution
of gas ceased,
thoroughly.

diluted to 25 ml. and mixed

All color readings were made in a

Coleman Universal spectrophotometer a t 420 mu.
The color developed by th is method was
found to be unstable with time as i s
in Table 6•

illu stra te d

30

TABLE 6

CHANGE OF COLOR WITH TIME USING FOLIN-WU REAGINT

Mg, Cholesterol

Spectrophotometer Readings 420 mu
0 Time 30 min. 150 min, 5 h r s . 2 1 h r s .

0.00

100

100

100

100

100

0.25

75.5

77.5

82.0

84.0

81.0

0150

56.0

57.5

64.0

69 .0

69.0

0.75

47.5

47.5

54.0

59 .0

61 .5

1.00

35 .5

35.0

37.5

42.5

50.0

1.25

33.5

33.0

34.5

39 .0

47.0

31

Because of t h i s

u n s ta b ility of color

developed by the Folin-Wu reagent, Nelson’s (34)
modification of the Somogyi reaction was
in vestigated.

The color developed with th is

reagent was much more stable and more concordant
r e s u l t s were obtained.

The color development in

the high ranges of s te r o l concentration with
th is reagent, was too intense for p ra c tic a l use
and a new se ries

of cholesterol standards was

prepared containirg 0.1,

0.2,

0.3, 0.4, 0.5 mgm.

of cholesterol per five ml. respectively.
Using these substitutions the r e s u lts of
duplicate serie s A cuad B shown in Table 7 were
obtained.
Again the color developed seemed to be too
intense for accuracy so the solutions were
diluted to f i f t y ml. before reading and these
re su lts,

duplicate se rie s C and D, are shown in

Table 8.
These r e s u lts seemed to be of the desired
accuracy and a working curve was prepared by
averaging the r e s u l t s of these two s e r ie s ,
Figure 5.
Solutions of stigraasterol containing

32

TABLE 7

COLOR DEVELOPMENT USING NELSON’S REAGENT

Mgm. Cholesterol Spectrophotometer Readings 420 mu.
Series A
100

100

o
•
H
o

0*00

Seri es B

58*5

59*0

0*20

34*0

35*0

0 . 30

21*0

22.0

0.40

12*0

13*5

0 *50

8*0

8*5

33

TABLE 8

COLOR DEVELOPMENT USING NELSON’S REAGENT DILUTION
TO 50 ml.

Mam. Cholesterol Spectrophotometer Readings 420 mu.
Series C
100

100

0.10

78.5

O
.
o
00

0.20

61.0

63.0

0.30

49.0

to
•
o

0.00

Series D

0.40

38.0

39.0

0.50

31.0

32.0

co

cr

LU

CM

WORKING

z<t m
o
UJ
o

O O

N

10

in

o

CURVE

MG. CHOLESTEROL

in

USING NELSON’S REAGENT

34

rO

CM

N0ISSIWSNVW1

lN 3 0 8 3 d

FIG. 5

35

0*1, 0,2,

0,3,

0,4,

0,5 mgm. per 5 ml* were

prepared and the s te r o l content was determined
using the cholesterol working curve.

Four

r e p lic a te s A, B, C and D of each level of the
s t e r o l were determined individually*

The error

was calculated to see whether or not reproduci­
b i l i t y was being obtained, Table 9.
The r e s u l t s with stigma s te r o l showed some
promise and so solutions of mixed soybean
ste ro ls,

supplied by Prof* C. D. Ball, were

prepared containing equivalent amounts to the
previously used ch o lestero l and stigmasterol
so lu tio n s.

The r e s u l t s obtained with these

solutions are shown in Table 10.

Again four

re p lic a tio n s A, B, C and D were determined
in dividually.
The r e s u lts with soybean s te r o ls
appeared to be very low.

(Table 10)

However, i f th is

mixture of s te r o ls were to contain 20% impurities
the percentage errors shown in Table 11 would
have re su lte d .
Further se rie s of determinations resulted
in erro rs of approximately the same magnitude
and i t was found that while reasonable duplication

36

F
h

STiaiASTEROL

DETEIMINED

FROM CHOLESTEROL WORKING CURVE

o

n

w

in o to 10
•

•

•

•

to* o• <o
o*
o»
•
«
O O H rl N in

•

O D- r -l O O to

H

+ + +
tO o o

O- c\3

r~ i

H i—
If—
I
+++■
+
•I

9 f
o

to o o o o

o

C\2 GO

in

O W O O H

cam

10 10

i—I C\J CO i—I C\2

o o o o o

o o o o o

+ +■+•+ I

Pi

0
o

lOo o o o

O 1—
<CMCOCO
<4 o
C
O
♦H
>
+3 •atH
CO
u
a>
CO
IS

o
to

E-t

to o o o o

O- CM rH CO CM
O O O] O Cft 00

to

•

O O*

o o o o
« •

•

03
to
I
D
•H
P
i
<D
co

O ' CM lO t o to
O i—I CM CO H OO r - | CM CO
^

O
•

cn o> OJk CM

o o in^ coco

i—
i

LO t o o

•

•

•

•

O tO C9 tO IN CO
O O' LO ^ COCO

I—
I

4*}
Pi
a>

G
O
a>

fi
tH

+»

ca

o o o o o o
O r-» CM CO

to

o o o o o o
O

I—

I CM t o

^

LO

<
D
0
53
•H

§

o

37

b
u

O iD W O W
•

•

*

•

LO
CO
O• 03•
•
•
OO^D WHH

•

O tOO Oi—
IpH
i—
I
1++ • I

p«

w

i—
1
+ 111

o to co^ 'O
r-H O o lO
rHO O O O

tO i—I i—I ^ c 0
O O O O lO

O

o o oo o

b
u

Iu

I

n
+>

8
o

o»

ta£ o
*
H C
O
-«
R
CO

+

+

I

r-H rH <o
o a> o O CO
o• o• 03%•
COCO• •
o

O rH 02 03 to ^

<D
•H

•H

Pi

a>

CO

E-*
to

OC
O03cO^
OOHOiCO^

C
O

<D

Pi
©
co

o

+ 111

I

• • o• O• o•
O i 03 a> o lO
C- «o
CO
lO O

O
O
rH

o• to
o o• o*
• •
o a> o o o io
o c> cOto ^ to
I—
I

e-«

-R

<D
C
O
<J>

fi

e*
fl
iH
-R

CO

o oo o o o

O rH 02 to

LQ

o o o o o o
O rH 03 to ^

lO

38

F
-t
O
m
&

to
o o o wo
# • • • •
o o* c» o eo oo

O• to• O• to• CO
•
O CMf>- <0 CMGO
CMrH r-H CMC
M

rH rH CM02 C
M

I

I I I I I

'C
f
©

I I
+»

O
rH

8
•F
h

o

•

o
*
©—
»d
+»

3
o

O

03

o

•

o

lO
0*

o

♦ *

o
rH
*

i l l

to

t

1

tO CMO lO O
O CO COttfi o to
O O rH CMCO tO

©
©
tH
fH

GO O

to ’M
*o>
o O o rH
»

♦

«

•

1 1 t

1

ffj

00 to CM O <o
O ES <o to r H to
O O r H CM CO tO

©

o

•

♦

•

•

•

•

©

-H

Fh

©
03

03

O

*

S

to

o

©

3

CM
CM
O

o
f

03

GO O
tO tO

O• o• to• to• to•
O CMGO <0 CO CM
O GO 55 lO ^

o

to o o o

to

O t o GOCO

•

o

•

«

•

•

O CD tDlO ^ ^

Eh
*£

Ft
©

©

©

£

S
Fh
©

+>

03

o

03

o o o o o o
O rH CM to ^ lO

o o o o o o

O rH CMCO -Cjt to

o

39

Pi

O
Pi
f-t

8
v li

s

Pi

<—1
o

Pi

©
+» pj
CO 2
o

>»Prn

o

co

3
S3

•

o

+ I I I I

CM CM CO <0 CO
60 CM
O O
O O O rH rH

H * rH CM CO H«
CM H * CM IS rH
O O O O »H

1 I I I I

+ •111

co ao *>■ ^

o

CM CM rH rH CM

t I

H
*o»C
Os- <o

*>-

O«0M0 0>0>
O O rH CM CM CO

©
©

fl

O cm in o - cm ao
O rHrHrHCO CO

CO

©
Pi
©

•r-l
Pi
©

CO

CO

3
&
in

O^OHOON

EO •—I CM CM 02

I 1

o
o

IO O CM GO
• • • •

OWr^HtOO

&

+»

lO

O• O• O• k)* «0•

in o o o
» • • •

in

O lOifl t>o o»
co <£> m in co

O "I
O O

o

o

o o o o o o
O H CM CO
in

o
o

•

•

in o o o
•

•

*

COO (O o
eo in

•

H1

E -i

-p
PI
©

OQ

©

Pi
P-i
rH
P
ft
©
43
CO
>.
o

CO

O

o

o

o

o

rHCMCO H1 lO

o

40

TABLE 11

PERCENT ERROR FOR SOYBEAN STEROLS ASSUMING

Series A

Series B

Series C

+ 3*00

- 2*00

-12*00

+ 1*00

+ 2.50

+ 9.00

0.50

0.00

+ 4*00

- 1.00

+ 9.00

- 3.25

- 2.50

- 6.50

+ 1.25

I
00
•
o
o

20# CONTAMINATION

Series D

- 8.80

- 0.60

+ 2*80

+44

41

could b© expected on samples carried through
side by side that duplication from day to day
was much more uncertain.

In order to investigate

t h i s point several s e r ie s of known amounts of
c h o le stero l were run on successive days and the
r e s u l t s plotted in Figure 6.
In view of th is v a ria tio n , which appears to
be in the hydrolysis and which up to the present
has not been controlled,

th is method was

temporarily abandoned and investigation of the
oxidation of the digitonide begun.

o

oc

LU

h-

D CO
LU
_J
0
01

o

CD

co

O

cm

to

CD

LlI

O

o

o

o

a>

oo

h-

cd

to

CD

ro

o

CM

N0ISSIIAISNVH1 lN 3 0 « 3 d

FIG. 6

VARIATIONS

n

DUE TO POOR CONTROL

OF HYDROLYSIS

42

43

D. Oxidation of the Digitonide
Reagents

( a l l reagents a n a ly tic a l grade)

80% Ethanol - 95% ethanol diluted with
d i s t i l l e d water*
1% Digitonin solution - 1 ga,

of digitonin

(Merck) dissolved in 100 ml* of 80%
ethanol with warming.

No v a r ia b il ity

was noted i n r e s u lts a f t e r two months
usage i f

stored in a glass stoppered

flask*
Absolute ethanol - Canadian Industries
Limited absolute s p i r i t s .
Potassium dichromate - 29*422 gms . of dried
pot as si tarn dichromate dissolved in
100 ml* of d i s t i l l e d water and made up
to 1 l i t r e with concentrated sulfuric
acid*

The use of Nicloux’s reagent is

not necessary.
Ferrous ammonium sulfate - 39*2 gms. ferrous
ammonium sulfate dissolved in 1 M
sulfuric

acid and made up to 1 l i t r e

with 1 M sulfuric acid.
Proc edure
The ex traction of the plant material and i t s

44

subsequent p a r t i a l p u r i f i c a t i o n fo r p r e c i p i t a t i o n
was clone by the use of Wall and K e l l e y ’s
procedure
S p irits

(13),

s u b s t i t u t i n g Canadian Mineral

"Hexane", B.P.

150~158°F., for

Skellysolve B.
Following the

adsorption on magnesium oxide

and Super-Gel of a sample of p lant e x t r a c t
equivalent to 0.25 - 1.0 gm. of a i r dry plant
material

(depending on the

amount o f s t e r o l about

s t e r o l content optimum

0.25 - 1.0 mg. s t e r o l

although amounts ranging from 0.15 - 2.0 mg. can
be determined under these
is evaporated j u s t

conditions)

to dryness on a well

v e n t i l a t e d steam bath.

The residue

taken up with hexane and f i l t e r e d
a 15 ml.
it

centrifuge

tube*

is

cooled,

directly

into

For rou tine analysis

is advantageous to adsorb one l a r g e r sample,

evaporate to dryness,
volumetric
aliquots

filte r

into a 25 ml.

f l a s k , make up to volume, mix and take

of the r e qu ir e d volume,

The solvent

then evaporated once again to dryness.
to

the eluate

obviate the n e c e ssity

this

In order

of using b o i l i n g tubes

evaporation was carried out

of rapidly

is

boiling 95^ ethanol.

in

the vapors

Under these

4:5

conditions evaporation is rapid and no active
t o i l i n g occurs*
by th is treatment

I f a l l the hexane i s not removed
the tubes are transferred to a

steam bath to remove the f i n a l traces of hexane*
It

is e s s e n tia l th a t a l l the solvent be removed

a t t h i s time, or

i t w ill in te r fe re with the sub­

sequent precipitation*
P re c ip ita tio n of the digitonides is
accomplished by suspending the tubes in the
vapors of boiling ethanol as shown in Figure 7.
Five ml* of absolute ethanol are added, taking
care to wash down thoroughly the sides of the
tubes*

Nothing fu rth er is added u n t i l a l l the

residue has dissolved.

Hi is i s followed by the

addition of 2 ml. of dig itonin solution and 1*25
ml. of d i s t i l l e d water.

The contents are

thoroughly mixed by means of a glass thread
which is in turn rinsed with 80$ ethanol.

The

tubes are heated for about a minute longer and
then placed in pint preserving ja rs with a screw
top and sealed (7).
quarter

The ja r s have about one

inch of 80$ ethanol in the bottom.

In

t h i s manner p re c ip ita tio n occurs overnight with
a minimum of evaporation.

CONDENSER

CORK

SHIM

SAMPLE

TUBE

ETH A N O L

SAND

REFLUX FOR

BATH

EVAPORATION OF HEXANE

AND PRECIPITATION OF

FIG. 7

DIGITONIDES

47

A fter overnight p r e c ip ita tio n , any
p r e c ip it a te sticking to the walls of the tube is
loosened by rubbing w ith a glass thread and
rin sin g down with 80% ethanol*

The tubes are

then centrifuged a t 3000+ r*p*m* for one-*half
hour and the

supernatant liq uid

is removed by

means of a c a p illa ry tube and gentle suction*
Decantation is not sa tis fa c to ry with a l l samples
because the p re c ip ita te does not pack s u f f i c i ­
e n tly ,

th is

samples*

is p a rtic u la rly true of the smaller

In some cases a flo a tin g gelatinous

p r e c ip ita te

i s present which may range from a

f a in t cloudiness to large discreet particles*
This i s removed with the supernatant liquid and
does not appear to contain any of the sterols*
The p re c ip ita te i s washed once with 80% ethanol,
using about 3 ml* and thoroughly mixing the
p r e c ip ita te with a glass thread.

I f the concen­

t r a t i o n of ethanol is correct the thread may be
read ily rinsed free of p re c ip ita te with 80%
ethanol,

i f the ethanol concentration is much

le s s than 80% t h i s

is impossible*

The tubes are

again centrifuged and the supernatant liquid
removed a s before.

This washing removes the

48

excess d ig ito n in .

The p re cip itate

is then washed

twice, a t l e a s t , with ethyl ether using the above
technique,

in order to remove other l i p i d s .

After the f in a l removal of the e th e r,

the

tubes are allowed to stand fo r about two hours,
in p ra ctice they are allowed to stand overnight,
in a covered beaJaer a t room temperature.

Then

drying i s completed by placing them in an oven
a t 90-95°C. for about one hour.

By t h i s means

a l l danger of spattering and loss of p re c ip ita te
is avoided.
The tubes are then cooled and a measured
quantity of sulfuric acid - potassium dichrornate
reagent
added i s

i s added.

The exact amount of reagent

immaterial providing th a t there i s an

adequate excess (25%) and that the amount can be
accurately duplicated.

In p ractice,

the authors

have been using j’ust over 4 ml. dispensed frcm a
sp e c ially designed p ipette

(Figure 8) modified

from MacDougall and Biggs (35), which has been
very s a tis f a c to r y as shown by Table 12.
Immediately a f t e r the reagent has been
added, the tubes are suspended in a steam bath
to a depth approximately one quarter inch above

49

r ^ \

AUTOMATIC

ZERO

OVERFLOW

CAPILLARY
T IP

SUCTIO N

REAGENT

AUTOMATIC P I P E T T E

FOR

K 2 Cr2 0 ? -

REAGENT

FIG. 8

50

tab u : 12

STABILITY OF K2CR207 - H2S04 REAGENT AND
REPRODUCIBILITY OF MEASUREMENT WITH PIPETTE

Sample No.

Date

A
B
C
D
E
F
G

May 21
” 22
"2 5
"2 8
"2 9
"3 0
June 8

Ml. Fe++ Sol'n
23.15
23.15
23.15
23.15
23.20
23.15
23.15

51

the surface of the reagent.

This i s accomplished

by slipping pieces of rubber tubing over the
tubes to support them on the steam bath cover.
The tubes are l e f t stationary for three and
one-half hours a f t e r which time they are
inspected and gently rotated in a sloping
p o sitio n .

I f any p re cip itate

sides of the tube i t
a g ita tio n .

is clinging to the

is rinsed down by gentle

Heating is then continued for

another one-half hour.
The contents of the

tubes are then

q u an titatively rinsed into beakers using 40-50
ml. of d i s t i l l e d water,

1 drop of ortho-phenan-

throline ferrous complex indicator

(36) is added

and the excess die hr ornate is then t i t r a t e d by
means of 0*1 N ferrous ammonium sulfate solution*
The end point is a very d i s t i n c t change from
blue to red and is

sensitive to less than one-

half drop of 0.1 N ferrous solu tion.
S t a b i l i t y of Reagents
Sulfuric acid - potassium die hr ornate:-

The

reagent appears to be stable indefinitely i f
stored under the

conditions used.

This fa c t i s

I l l u s t r a t e d by the t i t r a t i o n of measured a liq u o ts

52

as shown in Table 12 and is substantiated by
Williams and Reese (37)*
On heating the reagent
there

is a d e fin ite and reproducible loss of

oxidizing power as is
It

in a steam bath

i l l u s t r a t e d in Table 13*

is also apparent in the working curves for

the s te r o ls

(Figure 9) because a t zero concen­

t r a t i o n of s te r o l the curves do not in te rs e c t
the ordinate at the zero point.
Ferrous ammonium s u l f a t e : -

This reagent

was in d ire c tly s ta b iliz e d by means of a lead
reductor described by Duke (38).

When the

reductor was f i r s t prepared the ferrous ammonium
su lfa te

solution was circulated through i t

times, t o t a l volume about 15 l i t r e s ,
ensure thorough mixings

three

in order to

During the f i r s t month

of use there was an apparent increase in the
reducing power of th is
to approximately 1$.

solution which amounted
After this period the

solution remained stable throughout the course
of th i s work.
Preparation of Working Curve
In e a r lie r

Investigations i t has been

assumed th a t the oxidation proceeds with the

53

TABLE 13

LOSS OF OXIDIZING STRENGTH OF K2CR207 - H2S04
ON HEATING

Time of Heating
Hours
Ml. Fe++ Sol*n. % Recovery CrpO?0
1
2
3
4

23.10
22.97
22.88
22.73
22.65

100
99.43
99.05
98.33
98.03

54

UJ
JlL

or
o

20.0
UJ

o

Q

UJ
CO

z
o

15.0

I—
CD

CHOLESTEROL

Q

CV
J UJ
o

jE

CM

10.0

u.

o
41
o

CO
*■

"O)
LL.

IZ
O
I—
<
o

STIG M A STER 0L
5 .0

X

o
0 .5
MG.

WORKI NG
STEROL

STEROL

CU RV ES FOR

CH0LE

AND S T I G M A S T E R 0 L

FIG. 9

55

exclusive formation of carbon dioxide and water
and the cholesterol content has been calculated
on t h i s b a sis.

Okey (28) assumed the formula

for ch o lestero l digit cnide to be C82H140°29*
M. W
* 1585, and calculated th at 10*62 ml* of 0*1
N potassium die hr ornate were equivalent

to 1 mg*

of cholesterol, whereas Mcnasterio (6), reporting
no formula, calculated th at 10.48 ml. of 0.1 N
sodium th io su lfa te were equivalent to 1 mg* of
cholesterol as the digitonide*
Feeling that t h i s assumption may not be
sound, and c e rta in ly i t

did not hold fo r the

conditions outlined in the present work, the
author f e l t

i t b e tte r to consider the detem in a-

tio n as being empirical and so prepared a
working curve by p lo ttin g the mg. of s te r o l
present against the ml. of ferrous solution
equivalent to the die hr ornate used f o r the
oxidation (Figure 9).

The t i t r a t i o n values on

which Figure 9 is based are to be found in
Table 14.
It

is apparent,

th at a t very low concen­

tr a tio n s of stigmasterol the straight

lin e

re la tio n sh ip does not hold and determinations in

56

§
•H
+*
d
M
O
d
O
*<
+H
»d
i Td
pH <
D
O 09
co 0
+ 1
+ DQ
>O
pt|
03
fH
♦O
-1
O 1

d
<D
p

§
d
f43H
d
s
O
I l O O CM O J r H 0 0 I O
I E N r-IC tftv O C O O -O

|

• • • • • • • •

I cvjuf>c^a>cM<
sH«oa>
rHi—
Ii—
| i—
|

Si

09

SI

CD

p
H
o

• It
Od
•• O
03
I—
IP
o i
COo
+ 09
+ 09
<D<D
O
•w
C
D
H

O to o

CO a ) <3> 03 t o H

CQlOWOlOWo>lDW
C
Q000 CQH00
WNHHHH

a

o

8
&
a

M
W
O

rH ♦
O
d SP
<
D
43
oioomoiooioo
C
Q-6 OC\JlO£>003moO
D
C
DC
rH
C
O O O O O H H H rlW
OC
D
jd d
o d«

57

o
+»

tH

fltf

•H

y
o

•
*4
••
rH
O

o
<*H
*CJ
CD

CO 03

0

+»

a
o
PS

H
O
<D
P>
CO
a
S
•rH
CQ

+
H- H
CD O P«M O
CM
• U
o
3

• a
Do
rH CM
O JH
ca O
+■ CO
O
H
* C
<r> CD
Ph o
M
* 0>
i—1

■—(
o *
JH tt>
07 d
■P
10 -H
g a>
GiQ CO
•rH 0
-P*
c o PH

1 O 'D ®

W O O lO

Ii ^co
ofcC
• •
• •\2tQ
«<
•aa>
•C
«'i cva^toc^rHcQioc-

10
CM

0> tO 0> CM lO lO O
lO CM 0 > O * <& CO lO

to o oo

to rH C» E M f)

CM CVI rH rH i—t

oiooiooinomo
O C M iO C -O C M lO O -O

•

• • • • • • • *

O O d 3 O i—I rH rH rH CM

58

t h i s range (0-0*1 mg. s te r o l) are inaccurate.
This lack of l in e a r i ty may be a ttrib u te d to the
f a c t t h a t stigm asterol digitonide i s more soluble
in 80$ ethanol than is cholesterol dig itonide.
I f the stra ig h t

lin e portion of the stigmasterol

curve i s extrapolated to zero concentration, the
difference between th is point of in te rse ctio n
and the reagent blank should give a measure of
the s o l u b ility of stigmasterol digitonide.

This

value l i e s in the neighbourhood of 30 pgm. per
8.25 ml. of 80$ ethanol.

On the same basis, the

ch olesterol curve when extrapolated to zero
concentration in te rs e c ts the ordinate a t a point
which coincides with the reagent blank, therefore,
c h o le stero l digitonide is not appreciably soluble
in 80$ ethanol.
In addition,

the slope of the curve for

stigm asterol is about 6$ less than th a t for
c h o le stero l.

This may be accounted for by the

fa ct th at the molecular weight of stigmasterol
is about 6$ greater than th a t of cholesterol so
le ss digitonide p re c ip ita te would be expected
for equal weights of the two.

59

Reproducibility with other Plant S te ro ls
In order to check: the recovery or other
plant s t e r o l s as compared with stigm asterol,
standard solutions of s i t o s t e r o l , mixed
s i t o s t e r o l and stigm asterol and iso la te d com
s te r o ls were prepared.

Measured aliquots of

these solutions were used and the recovery of
the s t e r o l s , as stigm asterol, was noted
(Tables 16, 17, 18).
These r e s u lts would indicate th a t the
s o l u b i l ity of the digit on ides of these various
plant s t e r o ls i s p r a c tic a lly id e n tic a l and that
a working curve prepared from one, i . e .
stig m asterol, can be used f o r the determination
of the others.

However, d iscretion should be

used in extrapolating to other mixtures.
V alidity of Procedure
A se rie s of samples of corn le a f tissue
obtained during various stages of growth were
extracted and the free s te r o ls detemined.
Replications of each sample were run separately
a f t e r the original extraction with hexane.

From

Table 19 i t can be seen th a t the accuracy of the
determination should be not less than + 3/&.

60

&
©
>
O
o
©
PS

l
OO(OWO(0^0
« ! • « • • • •
^OGOCftOCOcOtN

a>Oa>oJOo>o>o>
rH
i—
I

O
U
a>
■
P
co S
3
° 2
-PfO
H
at|
03

®OlOC^OO0tn

Q O O tfolO C fttQ iO
H C Q C O tO lO 0O 'sf<

• • • • • • • •

O O O O O O r H r H

tO

E*H

o
CD +3

+>
8

0
0
0

-P
■ft f-t
CO PH

O O O O O O O O
O O l Q O O O l O O

O
2CQCO^0ISOlO
♦ • • * • • • •
O O O O O O r H r H

tiD

o

£3
0

,©
EH

Ph H

REGOVERT

OF MIXED STIGMASTEROL AND SITOSTEROL

(5 0 -5 0 )

FROM HEXANE SOLUTION

61

<
D
>
O
O
<
P

00^01^00)10
OOO>C^rHOl>C0

O O o > G > O O a iO >

•—I i—4

rH i—I

03

«H

O
OCV2COo>I>-OtfDOO
<
DC
S OO^QOOOWtN
C•MC
OcOcOlOtNO^
-H$
• • • • • • •
f-4

co o

CDCDCDCDCDCDHH

d)

C
Q
1-1
8

O O O O O O O O

g O O L O O O O lO O
<
DC
Q Cs3tOtO^<lOC>OlO

4* C
D
CO F-»

P*

d>

o

a>
.a
EH

O O CD O CD CD rH rH

62

©

OCDtDiOWO>wO

CD

rHOOOc5>0>O0>
i—
If—
IrHrH
rH

sO

O I—| <—I CQ CJ> 00 CJ 03

PS

CQ

00
rH

M
©S
S
+> 2
C
Oo
PR
ti

OO iO rH tO C-O tO

s

o

CO

cO o> H

lO

C \J a O

CT> ''3*

O O O O O O O rH

I
$
gH
CQ

rH

s0 <D
s

+»
CQ

CQ
©

M
•P
M

W

o
la
©

,Q

O lO O O O lO O O

oiNOvoocvaoo
H03W^©^0»10
O O O O O O O rH

63

TABLE 19

PRECISION OF DATA OBTAINED FROM CORN LEAF TISSUE
Mg* Sterol
Found

%

Deviati
frccL Me«

19

0.665
0.66 7
0.677
0.670
0.669
0.667

-0.60
-0.30
+1.20
+0.15
0.00
-0.30

13

0.440
0.457
0.454
0.457

-2.65
+1.11
+0.44
+1.11

7

0.736
0.730

+0.41
-0.41

9

0.660
0.660

0.00
o
•
o
o

Sample No*

11

0.600
0.59 0

+0.84
-0.84

15

0.422
0.422

0.00
0.00

17

0.690
0.675

+1.00
-1.00

20

0.494
0.485

+0.80
-0.80

8

0.518
0.496

+2.20
-2.20

10

0.383
0.393

+1.35
-1.35

Because these re p lic a tio n s were run side by
side on the same day, a further single determina'
tion of four o f the above samples was run.
These four check determinations were run on the
same day but about three weeks a f t e r the l a s t of
the re p lic a tio n s

in Table 19*

The figures in

Table 20 would indicate th a t the r e s u l t s are
consistent from week to week and th a t there i s
no appreciable change in the strength of the
reagents.

This fa c t i s also substantiated by

the r e s u lts of Table 21 which shows the recovery
of known amounts cf stigmasterol one month a f te r
the preparation of the working curve.
Recovery of Added Stigmasterol to Com Extracts
In order to check the recovery of added
ste ro l,

two se rie s of recovery checks were made.

Series A in which the stigmasterol was added
following the adsorption of the sample and
Series B in which the stigmasterol was added
before the adsorption (Table 22).
I n te rfe rin g Substances
DigitOnin has been reported to pre cip itate
phosphatides from petroleum ether solution
and to p re c ip itate alcohols other than the

(39)

65

TABLE 20

PRECISION OF REPLICATIONS RUN AFTER A 3 WEEK
TIME INTERVAL

Sample No.
7
8
9
10

Mg. Sterol
Found
0.770
0.520
0.648
0.405

Deviation from
Values in Table 19

aJo

+5.0
+2.6
-1.8
+4.4

66

TABLE 21

RECOVER! OF STIGMASTEROL FROM HEXANE SOLUTION
ONE MONTH AFTER THE PREPARATION OF THE WORKING
CURVE

Tub© No.
H
V
Y
J
W
D
R
I

Mg. Stigmasterol
Present
0.075
0.100
0.300
0.400
0.600
1.040
1.095
1.350

Mg. S terol
Found
0.080
0.106
0.306
0.405
0.607
1.040
1.095
1.375

$ Recovery
106.6
106.0
102.0
101.2
101.2
100.0
100 .0
101.8

67

<
D o• 00
o «o CQ D*- rH
• • • • • « •
l>
rH IN O i
O
03 CD lO
o 8 a» a» cn O o o>a>
i—1i—1
<
d rH

*0

a>
3
Pj
•H
4->

§

TA
BLE 22

o
T3
05
Tl
TJ
-4
t—
I
o
JH
a>
4*

o o o o o o o o
LO to o o o to to
to
02• 02• CQ
to to to c - IN
• • • » • *
Oo o o o o o o

CO

'H
o
P
=
H
H
o
p«
(D

OcOOt>tOCOlOW
^o in w o o ^ ^ n
tD
C«OlO
lO
LO
'^lO
'^
•
•
•
•
•
•
•

o o o o o o o o

-P
CO

<D

r—
1

OiOWCOtO^HlO
<— I r -1

C
O

r H i—1 r—I

68

s

M
©

ca

>
o

o

o
(D
PS

LO to to to
• • • • •
O to o o» (D
o o> o a> o »
rH
rH

TA
B
LE 22 (Coat'd)

o
1—J
PM
PS
o
CQ

nH
©
o
■^l

w
e
P4
pq
pq

r—1
O
©
4^
CQ

O o
LO o
03 in
• •
o o

o
o

o
o
LO o
* •
o rH

o
to
to
•
o

to
co
to
•
o

o
o
o

•

1—I

a
pq
«
*C*

ps

o

S
O
£*4

pq
E-i

i—I

3
S=S

8
©
+*
CQ

CO
o
to
•
o

a> o
to o»
tO
• •
o o

CQ
P«4
O

*
O
£S

ps

M
p»
o
o

©
i—1
a

s

CQ

©

co
rH H

a> o
rH rH 03

69

s te r o ls

(40, 41).

I t was found that under the conditions used
in th is work fo r the digit onin p re c ip ita tio n no
p re c ip ita te would form with le c ith in u n til more
than 5 mg* were present*
and 5 mg. of le c ith in
tissu e ,

The addition of 1 mg*

to samples of corn le af

hoth "before and a fte r adsorption,

resu lted in no increase in the amount of
digitonide p r e c ip ita te .
Using oleyl alcohol and ste a ry l alcohol as
representative members of the higher alcohols i t
was found th a t no p re c ip ita tio n occurred under
the above conditions u n t i l more than 16 mg• of
these alcohols were present.

The r e s u lts

obtained with added le c ith in and the alcohols
are found in Table 24.
In view of these r e s u lts and the size of
the aliq u o ts used in the determination i t s e l f i t
would appear that interference from these
sources i s unlikely in normal plant

tissu e •

70

TABLE 24

THE EFFECT OF ADDED IMPURITIES OH THE RECOVERY
OF STEROLS
Mg. S tero l in
Sample

Added
Impurity

Mg. Sterol
Found

Hone

5 mg. Lecithin

None

10 mg. Lecithin

0.053

0.665

1 mg. Lecithin

0.669

0.667

5 mg. Lecithin

0.667

None

4 mg. Alcohols *

None

None

16 mg. Alcohols *

None

None

0.485

4 mg. Alcohols *

0.483

0.485

16 mg. Alcohols *

0.488

* 50-50 mixture o f

s te a ry l and oleyl alcohols

71

IV

APPLICATION OP THE METHOD
During the summer

of 1950Golden Bantam

sweet corn was grown in order to obtain tissue
samples of the various

parts of

throughout the growing

season.

the

plant

The p lo ts

were

located a t Hespeler, Ontario, Canada, in a lig h t
sandy loam s o i l which was seeded with winter rye
in the f a l l

of 1948 and which was allowed to

mature in 1949, was then cut and ploughed under
while in the swath.
1950#

Seeding was done on June 3rd,

At approximately eight day intervals

samples were harvested u n til the se t of seed.
As the plants developed they were divided into
various fra c tio n s ,
cob husks, e t c .

i.e .

ta s s e ll, silk ,

s ta lk ,

At one stage of development,

July 26 and 27, samples were obtained at four
hour intervals

in order to study any diurnal

variations*
The percent moisture recorded was obtained
by weighing the fresh sample when harvested and
subsequently weighing the a i r dried m aterial.
The l i p i d and s te r o l contents were calculated on
th is a i r dry weight and not cn a moisture free
basis.

The percent li p i d was determined by the

72

difference in weight of the a i r dry sample before
and a f t e r exhaustive hexane extraction in a
soxhlet type e x tra c to r.

After extraction the

solvent was removed from the samples by drying
in an oven a t 63-65°C. for 48 hours and then
allowing them to come to equilibrium with the
a ir

(usually required 4 - 7

days).

The s te r o l

analyses were carried out on the hexane extracts
by the oxidative method outlined and the re s u lts
are tabulated in
A number of

Tables 25 and 26.
in te re s tin g observations can be

made from Table 25.

In the f i r s t place, the

growth curve of the corn leaves follows the well
known form as is seen in Figure 10.
the t o t a l lip id s

Secondly,

reach a peak rath er early in

the development of the corn and then f a l l
markedly.

very

This fa c t is i l l u s t r a t e d in Figure 11.

As th is decrease in t o t a l lip id s coincides with
the very rapid growth of the young cobs i t might
be postulated th a t they are reentering the
metabolic cycle in some form, perhaps converted
to carbohydrate, and are then u t i l i z e d for the
development of these cobs.

Thirdly, the ste ro ls

are retained in the le a f for a much longer

73

o
U HfJ
<D S=1

-p cd
cn r—
i

to

O
•

PM

o

rH

•
o

in

os

rH

t>
•

to•
in

o

•
rH

02

to
o
o•

in
rH
c»•

CO in
CO 03
to• 00•

•

iL
B) <D
£f Oh

— to

02

o
•
02
to

to

o

in•
o
02

o
02
in»

to
cn
to

02
0
to0•

o
to
to•

a»
to
to»

00
•

00
*

i i
rH

O

»

23,3

rH

r~H
o
a>

HP ♦
c o d <D

•

t(0 3
w

to

f-\

& IDtH

O o
in
03•
•
1—
1 o

s

rH

o

o

o

o

as

to
•
02
02

o

o

o

o

CO

in
•

cn

rH

0.490

u

PMEh

Q

to
o•
o

in in
i—
i c-•
*
o
o

o
a*
in
•
*
•—
i to

rH
•

02

•
to

O ©H

in

to

CO

02
•

IN

CO

to

•
in

in

IN

to

tO

in

rH

o

o

o

o

o

o

02

o

in•

o

CO

a»

oo

00

in

vH

in

in
in

«

•

•

•

•

•

•
rH

1

Q rCj

o

•

o

as

as

co

<D*h eg

u a>s
F
h ciJ

<o
+H+»
a> S S

E*QrH aj
< |P h P

to

«
as
00

3*

rH

to

03

o
•—i

o

O

o

•

rH
rH

O

ca t*0 *

•

*
IN

o
rH

as

•

02
CO

in
•
8

02

i>
rH

•

CD

o

O

•
00
IN
to

02

•

*

in

as

*
as
00

in

o
02

00

1—1

in

co

to

02

02

CO

o

O

o

o

o

inm

in•

in•

OS
CO
1—1

CO
Os

02
00

om

o•

o*

os

in

IN

o

as

CO

•

in

o

•

•

to

rH

rH

rH

O

O

CN

Os

03

as

c-

in

to

to

•

»—1

o

•

»

02

o
•
02
02

o

•

to

00
in

•

•

o

OS

o

•
os
to
to

o•
02

in

to

to
in

in
cO

to
o

O

o

•

o-

03

CO

oo

70,0

<>

C
l
+P
>J
^ «>

3

E
h

03
•

143.5

Eh

•

r —1

PM

•xJ
•H
C
M

02

to

to

485.0

L
O

in

•
to

0.4

E-«

•

47.7

*

>
»<
d eg
h
p <g

a» to
co

as

74

co
co
M
E
h
ft

o
<DCl

+> <a

CO

B

rH

Ph

F
h

tO
•

to
•

o

a*
rH

1— 1

GO
•
rH
rH

D-

l>

cm
rH

O
4

4

4

o
rH

o
r— 1

co
i>*

<
M
CM

HQ (D

ft
ft

O
o

o

El
©
+»
co

CO lO
Cft
in*
•
O o
£>

in

•

*

o

CM

tO

tO

GO

CM

CM
CM

CM

to4

GO

rH

rH

rH

to

to

to

D -

e -

o

to

CM

in

in

4

O

•

o

CM
CM

4

O+»
§

tO

•

o

4

rH

ft
ft

•

CM

in

•

CM
CM

tO
CM

CM

o
to•
o

CM

rH

CM

4

to
GO

o
00

in

4

4

4

to

4

4

to

4

to

4

4

4

o

•

CM

GO

00

co

00

in

in

in

in

in

o• o
CM to

o

o4

o

rH

S ft
ft

El
ft ft
H
O•
Ph -H ffl
■H d) Q

2=C
5

rH
rH

+»

o
ffl -HtQ a»
tO
P
<1> S
ft C
5 i>
t» H
O
aJ ri
*& *
H
**
9O
*
C
O*
H

ft ft

° 3
© CO

B

*rl f t

£HO

rH

o
to

r—I

rH
rH

rH

rH
rH

rH

o

O

o

O

in
tO

in

o

4

4

h

rH

CO

4

o
oo

•ft

4

a»

to
to

o
to

4

H<

00

4

4

4

O
O
rH

4

o*

4

4

4

4

4

B•

B•

B4

B
ft
H
O
aJ
■rH
S3 O
tC*

4

o

**

©

o
o

S3

O
m
m o
oo

ft

O o
O
o
•«
•«
co

75

CD

30

WEIGHT

i
i
i

20

TO

AGE IN DAYS

GROWTH CURVE OF CORN

FIG. 10

LEAVES

7

2000

100

80 ^C9
COB WEIGHT —

60

800

40

400

20

AGE

COMPARISON

60

40
IN

100

DAYS

OF TOTAL

WITH AGE AND

80

COB

FIG. II

LIPID OF LEAVES
WEIGHT

COB

LIPID IN LEAVES
TOTAL

20

WEIGHT

PER

1200

PLANT

LIPID WEIGHT

OF

ONE

PLANT

1600

77

period but f i n a l l y they too begin to disappear
as i s seen in Figure 12.
little

Fourthly,

there can be

dcubt th at the corn le a f has the a b i l i t y

to synthesize ste ro ls up to about the 75th day
of growth but that following th is

period the

synthesis is d e f in ite ly limited and does not
p a r a l l e l t h e i r destruction.
of synthesis of the

F if th ly ,

the r a t e

ste ro ls may not quite keep

pace with the rate of growth of the corn le a f
because there seems to be a slig h t decrease in
the percentage value.
From Table 26 i t would appear th a t there is
little

v ariatio n in the s te r o l levels in corn

leaves over a 24 hour period.

40

o’

CORN

LEAVES

20

WEIGHT

OF STEROLS

30

AGE

DEVELOPMENT

IN

DAYS

OF FREE STEROLS

F I G . 12

WITH AGE

79

V
1*

CONCLUSIONS
The q u an titativ e estimation of mixed ste ro ls
through reactions involving the hydroxyl
group does not seem feasible on a micro or
semi'-micro basis •

2*

The quantitative estimation of mixed s te r o ls
by nephelometry does not appear feasible
under the conditions studied unless an
intimate knowledge of the individual s te r o ls
present is av ailable.

3*

Nephelometry might easily be adapted for a
rapid micro estimation of individual s te r o ls .

4.

Estimation of the reducing sugars liberated
by the hydrolysis of the digitonides yields
r e s u lts of the desired s e n s itiv ity but lacks
the required precision.

The lack of pre­

cision appears to be due to poor control of
the boiling ra te during hydrolysis.
5.

The oxidation of the digitonide by dichromate
followed by the quantitative estimation of
the excess dichromate yields consistent
r e s u lts of good precision and s e n s i t i v i t y .

6.

The use of Nieloux's reagent for the oxida­
tio n is not necessary*

80

7*

Under the conditiaas used the oxidation does
not proceed to the exclusive formation of
carbon dioxide and water and cannot be
considered a stoichometric reaction,

thus the

determination is considered to be purely
empir i c a l .
8•

The precision of the method i s b e tte r than
+ 3% and samples containing as l i t t l e

as

0*15 mg. c£ s te r o l can be used*
9*

The t o t a l l i p i d fra ctio n of the corn leaves
reached a maximum in about 65 days and then
decreased very markedly*

This decrease just

precedes the very rapid development of the
cobs *
10.

The s te r o l fractio n of the com leaves
reached a maximum in about 75 days and th eir
subsequent decrease was not as marked as that
of the t o t a l lip id fraction*

11*

These r e s u lts are based on only one y ea r’s
crop and should be fu rth e r substantiated.

81

BIBLIOGRAPHY
(1)

Windaus, A., Ber., 42, 238(1909).

(2)

Basu, K. , and Nath, M.C., Biochem. Z.,
1561(1934).

(3)

Schoenhe imer , R., and Dam, H., Z. Physiol.
Chem., 215. 29(1933).

(4)

Sperry, W.M. , J . Biol. Chem., 118. 377(1937).

(5)

TalaYant, E., Coll. Czechoslov. Chan.
Communs., 15^, 232(1950).

(6)

Monasterio, G., Biochem. Z. ,

(7)

Schoenheimer, R., and Sperry, W.M., J . Biol.
Chem., 106, 745(1934).

(8)

Sobel, A.S., Drekter, I . J . , and Natelson, S .,
J . Biol. Chem., 115, 381(1936).

(9)

Sobel, A.E., Goodman, J . , and Blau, M.,
Anal. Chem., 23, 516(1951).

28,

265. 444(1933).

(10)

Zechmeister, L., Progress in Chromatography.
John Wiley and Sons In c ., New York, p. 173
(1950) .

(11)

Ladenberg, K., Femholz, E. , and Wallis,
E .S ., J . Org. Chem., 3., 294(1938).

(12)

Ott, A.C., Ph.D Thesis, Michigan State
College, p. 53(1943).

(13)

Wall, M#e ., and Kelley, E.G., Anal. Chem.,
19, 677(1947).

(14)

Salkowski, E ., Z. Physiol. Chem., 57^, 523
(1908).

(15)

Woker, G. , and Antener,
Acta., 22, 13 09(1939 ).

(16)

S cherrer, I . ,
(1939).

I . , Helv. Chim.

Helv. Chim. Acta., 22, 1329

82

(17)

Woker, G., and Antener, I . , Helv. Chim,
Acta., 22, 511(1959).

(18)

Rosenheim, 0 ., Biochem. J . ,

(19)

Bernoulli, A.L., Helv. Chim. Acta.,
274(1932).

(20)

Thor nton,M,//., Private Communication.

(21)

Liebermann, C., Ber., 18, 1803(1885).

(22)

Burchard, 0 .,

(23)

King, L.C., and Ball, C.D., Private
Communi cati on•

Chem. Zentr.,

23, 47(1929).
15,

61^(1), 25(1890).

(24)

Rappaport, F r ., and KLapholz, R. , Biochem.
Z., 258. 467(1937).

(25)

Anderson, R . J . , and Nabenhauer, F. P.,
J . Am. Chon. Soc . , 46, 1957(1924).

(26)

Muhlbock, 0 ., Kaufmann, C#, and Wolff, H.,
Biochem. Z., 246. 229(1932) seen in abstract
only, Chem. Abs ., 26., 2536(1932).

(27)

Szent-Gyorgyi, A., Biochem. Z.,
(1923).

(28)

Okey, R. , J. Biol. Chem., 88, 367(1930).

(29)

Yasuda, M. , «T. Biol. Chem., £2, 303(1931).

(30)

Turner, M.S.,

e l.

136. 112

Biol. Chem., 92, 495(1931).

(31)

Boyd, E.M., J .

Biol. Chem., 101. 323(1933).

(32)

MacLachlan, P.L., J. Biol. Chem., 113, 197
(1936).

(33)

Hawk, P.B., Oser, B.L., and Summerson, W.H.,
P ra ctical Physiological Chemistry. 12th ed.,
The Blakiston Co., Toronto, p.520(1947).

(34)

Nelson, N., J. Biol. Chem., 153. 375(1944).

(35)

MacDougall, D., and Biggs, D., Private
Communicati on.

83

(36)

Pieroe, W.C., and Haenisch, E.L.,
Quantitative Analysis, End e&., John Wiley
and Sons I n c ., Mew York, p*219(1940)*

(37)

Y/illiams, M.B., and Reese, H.D., Anal# Ghem. ,
22, 1556(1950).

(38)

Duke, P.R., Ind. & Eng. Chem., Anal. Ed.,
17. 530(1945).

(39)

McEwen, H.D., and MacLachlan, P.L., Proc .
Soc. Exp. Biol. Med., 48, 195(1941).

(40)

Windaus, A., and Weinhold, R., Z. Physiol.
Chem., 126, 299(1923).

(41)

Melanoff, I . S . , Am. J . Pharm., 9£, 390(1927).