PART I

A MODIFIED PHYSICAL-CHEMICAL

METHOD FOR THIS DETERMINATION

OF VITAMINS D IN FISH LIVER OILS
I’AR'I‘ 2

STUDIES IN THE. ICHROMATOCRAPHY
OF IRRADIATEI: RRGOSTEROL ON
ACTIVATED SUPERFILTROL

1 TIxesis for the Degxuc of M. S.
g MICHIGAN STATE COLLEGE

Jacob MltcheII Hag:
$943

pt...

11"

!l..j§\njflla.lfl1 .I . I, . I I ‘

1 a
«.I‘ ~9'.‘
tI ,“ I

. o'.‘
4‘ m
.71....

.40: 3' 44 a

y

2‘ UL

. A". I.” 3
‘ “.«.‘Lu‘.‘“_,h;r. ‘
'3 if; “ ‘

. ..

3.:r - . . —u l .l '. '. '1 ‘r ‘. a , ' ’ ‘
J. “ -. I . “ .p ,1 .‘1-‘ _ _ ’..‘, .1 ~..' 'J .1 . I ' . .‘ .. a ..,, .' .
1 ‘ I . . .. . ‘ ‘

. r ‘ ‘.
. ‘.r‘ I,» .f,“ “.“‘:.‘ ‘.n. _
~f‘ I. 'kl‘ 3.'\.-.‘ "1; \

. ’I.‘ fl "t

I

~ I . 3/1 I.
Ia“ :, _3:-“1t-'.r\.
a > ' “ '

~47..- v...

w.

I ,3”
! I‘D"; "

-4‘ . .1.
1.,- -".5’.I','
' "I I r.-" '-‘ .|r J‘ '._
. ' L-' ' I. I - .
.od'. 1‘ 3-.11’ 1"". ”“6 I. ,.
much”. 1). , . ‘9'
‘~'. ‘...-. ‘ .I’ig ”I "1
H. V .i*}‘ . ] Vial “"4: "I ‘(v
4'9 u“.-:'LL":J,'.~
".L 0". ‘ . ‘

I 4.1. c]. .3‘
I" i‘iI‘*'¢;f's

"’3? .
{ is." .I; ‘-

u ‘ - ~‘

\ .“v‘M’ ‘ _.
\t' 9 :..
1‘ ."x:-‘.

N. .) . 3...! ‘x' . ..- . . k ‘ A , ' . .' ~ ".I ‘ "y 1 4." .' ‘ ..)."‘ v A-v-‘L’;..‘-‘ .
In .. ',.-_ " ‘ _ g ‘ ". ‘.W' "1"5‘9- .r'.':‘ 3 }-'.’-'- *"7;‘ ‘21“ ‘.'
.‘z 1% - '. .' I I ,' .. '« {:3 ‘.| 1’1“ J. I. V - ‘ . : .' ’lv ' I ’.~ I. J. ' ‘6' .4. "If. t‘ ' I L 4 ‘ . ' ' I l t". “'.~‘*)'141A‘ ..
. . «Q . ‘ ' ~. , ‘ C w‘ ‘.l« I I. .7 ‘- 3 . ' . o ‘ ".' -‘ ' I "I." ‘3' . "'1 “ 1' ' . 3. ‘ H ‘h' " .1 '"

MTVMJI ‘-,-_‘ 1‘ '. . " ‘I ..' ‘. .' of“: W‘" ' .7 " ' '.1 ' 3' .w'. ' . " - " " «#V ' 3 ‘ . -'

.' '1'" . j ‘ ‘ :a I ' '~ ' 1 ‘ . ' J"... I "r‘ ‘ ‘ -. .‘ ' .- I . . ' ""1 ' ’ ‘ V. "2“,: “3”- '

I. . .. ‘ , ‘ .‘ ‘2' 0‘ ‘ , ' .L . 1‘ . . . .. ‘ -‘.'1 . I. .‘ a . ‘ ‘ r

\ '.-,;"'. ‘ " ‘ 3. '1' ' a 0"»,- - ' - ' ' ‘ ' ' " I ‘ . I #

w r- U i ' ‘ .C 5 ‘ I .. A . b . ’ ’

”2L
.

n"
. 4,1 ‘ J"- * H

\ - > , L’ . V . ‘ -.

_1~‘ f .‘l.‘~$:"’[:§'."‘ (‘2 t v,‘.

. r F‘D'N' 5‘ ‘1‘,
‘ I .
.v 1.\~ “ ‘r‘. H
i‘hf J.' , x
j“, \ "-
v" I
'4 - l

’7
I I
'\

‘1
11¢.

. 1 _ _,I_ '. l'
:l . 8&1. >‘.'.{.". 3‘.
_- «M. q. -‘.
' w

'-u .
"n‘

’34)) I \"-;'a‘

 

PART I
A MODIFIED PHYSICAL-CHEMICAL METHOD FOR THE
DETERMINATION OF VITAMINS D IN FISH LIVER OILS

PART II
STUDIES IN THE CHROMATOGRAPHY OF IRRADIATED
ERGOSTEROL ON ACTIVATED SUPERFILTROL

by
Jacob Mitchell Hage

A THESIS

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

MASTER OF SCIENCE

Department of Chemistry
1943

7’5’4/
H/4-I

'3. ‘4’" 0"

. U

ACKNOWLEDGEMENT

The writer Wishes to eXpress his appreciation to
Dr. D. T. Ewing for his guidance and counsel during the
course of this investigation, and to Parke, Davis and Co.,
Detroit, Michigan, for the fellowship grant which made
this work possible.

PART I

A Modified Physical-Chemical Method
for the Determination of Vitamins D in Fish Liver Oils

A physical-chemical method for determining vitamins D
in fish liver oils has been.deve10ped by Ewing and Kings-
ley (5). The authors based their procedure upon a two-step
chromatographic treatment where E(l percent, 1 cm.) is de-
termined first for the combined vitamins D and sterols, and
second for the separated sterols. By difference, the value
for the vitamins D is obtained.

The second step in the above chromatographic treatment
not only requires a considerable amount of time, but unless
the procedure is very carefully followed, the E(l percent,
1 cm.) for the sterols is likely to be high. From a criti-
cal study of this second chromatographic treatment, it has
been concluded that errors are due to rapidity of drawing
solution through column and also cracking of the packed
superfiltrol. This would tend to cause incomplete removal
of the vitamins D from sterols thereby giving high values
for the E(l percent, 1 cm.) of the sterols.

To avoid the above difficulties and to simplify and
shorten the entire procedure, the following treatment is
suggested to replace the original second chromatographic
treatment:

Referring to the procedure of Ewing and Kingsley (5),

page 303, column two, at the end of paragraph five, we have

(2)

the following sentence: uThe sample at this point contains
vitamins D and sterols". For the procedure after that sen-
tence, substitute the following:

Procedure.--To correct for the absorption at 500 7'34
due to the sterols present, another 1 ml. aliquot of the
chloroform solution of vitamins D and sterols is evaporated
to dryness in a 125 ml. erlenmeyer and then taken up in
25 m1. of benzene-skellysolve (2:1). To this is added di-
rectly a definite quantity of superfiltrol (a 5/8'' test tube
filled to a depth of one inch) and the mixture is allowed to
stand for thirty minutes with frequent swirling. The super-
filtrol is then filtered off, collecting the filtrate in a
125 ml. erlenmeyer. The flask should be rinsed twice with
10 ml. portions of benzene-skellysolve, adding each portion
after the preceding portion has gone through. These portions
should be added so as to wash down the sides of the filter
paper. The filtrate is then evaporated to dryness under re-
duced pressure, and the residue taken up in 1 ml. of chloro-
form. To this is added 10 ml. of the antimony trichloride re-
agent. Using a 2 cm. absorption cell, the extinction is de-
termined at 500 7”,! exactly three minutes after mixing.

From the two extinction values, the E(l percent, 1 cm.)
is calculated for vitamins D and sterols combined, and for
sterols alone. The difference between these two values gives
the E(l percent, 1 cm.) for the vitamins D in the original
sample. This value multiplied by the factor 19300 gives the

(5)

potency in U. S. P. Units of vitamins D per gram of oil.

Experimental Determination of Time Necessary for Como
plete Separation of Vitamins D from Sterols.-It has been
found that Vitamins D, and D, are very rapidly and com-
pletely adsorbed by activated superfiltrol from a solution
in benzene-skellysolve (2:1), whereas the sterols which are
present in natural fish liver oils in small amounts are ad-
sorbed very slowly and to a small extent. To measure the
time necessary for complete removal of vitamins D from
sterols in a benzene-skellysolve C2:1) solution with acti-
vated superfiltrol, reference oil 47761 was used. A sample
of this oil was run through the procedure up to the point
where the modification for determination of E(l percent, 1 cm.)
for sterols present replaces the older second chromatographic
treatment. We have at this stage 10 ml. of chloroform solution
containing vitamins D and sterols. A series of 1 ml. aliquots
of this solution are taken and run through the above modified
procedure. However, instead of swirling frequently for thirty
minutes, time intervals of one, five, ten, twenty, thirty, and
sixty minutes were used. They were then filtered and evapora-
ted to dryness under reduced pressure. The extinction for
each sample was then determined as above. The results are
shown in Table I. These values when plotted, Fig. 1, show
very clearly the rapid removal of vitamins D and the leveling
out of the curve at an extinction value due to the sterols

present.

Log Io/I

 

 

 

0.30

 

 

 

 

0.20

 

 

 

 

 

0.10 \

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

10 20 30 40 50 60
Time in minutes
Fig.1.—~Adsorption of reference oil 47761 from a solution

in benzene-skellysolve (2:1) upon activated superfiltrol.

<4)

Table I

Adsorption of Reference 011 47761 upon Activated

Superfiltrol from Benzene-Skellysolve (2:1)
Time (Min.) 1 5 10 20 so 40 60

Log IO/I .32 .29 .21 .13 012 012 008

Since an extinction value in the neighborhood of .14
has been considered correct for the sterols present in
reference oil #47761, the time necessary for complete re-
moval of vitamins D from sterols in ordinary fish liver
oils has been set at thirty minutes. If, instead of swirling
by hand, a device was used whereby the swirling could be done
mechanically, this time period could be shortened.

EXperimental Results.-The method was applied to the
following types of fish liver oils (Tables II and III):

A High vitamin D fish liver oils
B Vitamin D liver oil concentrates

C Vitamin D liver oil distillates

The results in Tables II and III are self eXplanatory and
show that the modified procedure is capable of reproducibility,
if the details are followed exactly.

A fairly good agreement is indicated between the modified
physical-chemical procedure and the bioassay. The modified

(5)

procedure also agrees very well with the original procedure.
For those results which are somewhat lower than the results
of Kingsley and Ewing (5), it is believed that a slight dew
teriorating of the oil has taken place. No oils of a po-
tency less than 10,000 U. S. P. Units per gram were run.

The time saved in using this modifification on a set of four
runs is approximately two hours. This makes it possible to
carry through such a set of runs in an eight-hour day.

Peroxides in Ether.-Ether as obtained commercially very
often contains peroxides. Among the simpler of these sub-
stances are ethyl hydrogen peroxide, C235-0o0oH, a colorless
liquid; diethyl peroxide, CzHg-O-Ooczfig, a liquid boiling at
65°C; acetone peroxide,(033603)z, boiling at 132°C and tri-
acetone peroxide, (03H602)3, melting at 97°C. Many of these
compounds are explosive.

Ether is used in the above procedure for making the de-
ve10per, skellysolve-ether-ethyl alcohol (50-10-1), and for
eluting. For pr0per deve10ping of the columns, it is very
important that the ether be free of peroxides or the bands
will be dull in color and not very sharp.

Peroxides normally present in ordinary ether give a de-
finite color reaction with antimony trichloride reagent. If
the concentration of these peroxides is fairly high, they
will be carried through the column of activated superfiltrol
and will remain with the vitamins D and sterols. This is

especially true in the case of the 25 mls. of ether used for

(6)

eluting after the first chromatograph. This ether passes
through a column of activated superfiltrol which is only a
few centimeters long and, therefore, any peroxides present
will be carried through. The results in case this happens
will invariably be high. It is recommended that highly
purified ether be used throughout and that it be tested
daily by dissolving a few crystals of potassium iodide in

1% starch solution and adding a few milliliters of the ether.
There will be a starchpiodine reaction if peroxides are pre-
sent. No reliance can be placed on the superfiltrol to re-
move peroxides. It has been shown that peroxides were present
in the solutions with vitamins D and sterols after the first
chromatograph when using impure ether.

Ethyl Alcohol Present in Chloroform.-Ethyl alcohol is
used as a preservative in chloroform. It was, therefore, be-
lieved worth while to check the effect of alcohol in the
chloroform used for the preparation of antimony trichloride
reagent. For this purpose, chloroform which had been washed
ten times with water to remove alcohol, dried over anhydrous
sodium sulfate and doubly distilled was used. Antimony tri-
chloride reagent was prepared using this chloroform and vari-
ous amounts of ethyl alcohol was added to 25 ml. portions of
this reagent. Results show that ethyl alcohol in the reagent
tends to lower the extinction value for vitamins D as well as
sterols. These results are shown in Fig. II for reference

oil #47761. The aliquots taken were run using the above

Log 1-'o/I

 

 

1.30

 

 

1.10

 

 

 

0.90 %

 

 

 

\ V M4» ter ls

 

0.70
*\

 

 

 

0.50

 

 

0.30

 

 

 

 

 

 

0.10

 

F

 

 

 

 

 

 

 

 

 

 

 

 

t——Y—_ Stérol
L l

 

 

 

0.1 dTfifl‘S 0.7 0.9 1.1 1.3

Ml. of ethyl alcohol per 25 ml. standard reagent
Fig. 2.--affect of alcohol in the antimony trichloride
reagent upon the extinction for vitamins D plus sterols

and for sterols in reference 011 47761.

(7)

prepared reagents. In reagent containing ethyl alcohol, hy-
drogen chloride is liberated indicating the following reaction:
0 O V
N H
CHa-C-Cl + 023503 ——9 CH3-C-OC2H5 + HCl

Decomposed Chloroform.-Chloroform which does not contain
alcohol tends to decompose, forming phosgene. The presence of
phosgene can be detected by odor or testing with starch-iodide
solution. Shaking with activated charcoal tends to remove
phosgene from the chloroform. This treatment must, therefore,
be used daily. Contaminated chloroform of this type tends to
give low results for both sterols and vitamins D. No extensive
study has been made of this.

modified Reagent.--A reagent which has been treated with
metallic zinc, tin or antimony to remove pentavalent antimony
has been suggested by Nield, Russell and Zimmerli (7). This
reagent is claimed to shorten the initial reaction period and
to cause less variation in color. Use of a reagent treated
with zinc upon samples of cholesterol tends to confirm this
claim. No directions for the preparation of this reagent was
given so the following is suggested: Dissolve 18 g. CP anti-
mony trichloride in 100 mls. of freshly purified chloroform.
After completely dissolved, add 5 g. of zinc dust. Allow to
stand for thirty minutes, shaking at intervals. The solup
tion must be protected from the air. Filter into reagent
bottle, add 2 mls. of acetyl chloride and allow to stand

thirty minutes before use.

(8)

Discussion.-The chromatogram of the non-saponifiable
portion of fish liver oils upon activated superfiltrol from a
solution in skellysolve-ether-ethyl alcohol (SO-lO-l) is very
interesting. A few seconds after adding the solution, a deep
blue band develops about one-half centimeter from the surface
of the superfiltrol. As the solution continues to pass through
the column, a green band forms directly under the blue band and
is separated from it by a very dark, narrow, greenish-blue
band. The intensity of color and degree of banding depends,
somewhat, on the nature of the fish liver oil, but they are all
of the same general character. In the case of concentrates and
distillates, the banding is not as intense, although they are
similar to those of the natural fish liver oils. Upon deve10p-
ing the above chromatogram with skellysolve-ether-ethyl alco-
hol (50-10-1), the blue and green portions change to a deep
gray and the narrow, greenish-blue band proceeds slowly down
the column. The blue eventually fades out, leaving the band
olive green. This band is formed whether Sudan III is added
or not. No cases have been observed where it does not develOp.
It is, therefore, believed that the use of Sudan III in the de-
termination of natural fish liver oils, concentrates and dis-
tillates by the above procedure is unnecessary.

It has been shown that peroxides in the ether will cause
the E(l percent, 1 cm.) of the vitamins D plus sterols as well
as that of the sterols to be high. Since a difference is taken,

it often happens that the final result comes out very well.

(9)

However, the amount of peroxide in the aliquot taken for the
sterol correction may be lowered by the succeeding treatment
and, in that case, the error will not cancel out. EXperi-
ments by R. B. Young (9/22/42, 9/8/42, 10/5/42, etc.) Show
the effects of peroxides in the ether. He, however, main-
tains in all cases that this is due to moisture from incom-
plete drying or poor antimony trichloride reagent. Time
studies show that after fifteen minutes of drying at 50°C and
under reduced pressure all the residues obtained in the above
procedure show no change in E(l percent, 1 cm.) upon further
drying.

Reagent may cause some change in E(l percent, 1 cm.)
values from day to day, but these would probably be small
fluctuations and not consistently high as is the case with
peroxide contamination of the ether.

It is of interest to also point out that R. B. Young
(7/7/42, 8/5/42, 9/9/42) frequently encountered E(l percent,
1 cm.) values for the sterol correction using the old second
chromatographic procedure which were high. He maintained
again that this was either due to moisture because of income
plete drying or poor reagent. It is much more probable that
vitamins D were coming through the column as explained above
causing high sterol corrections.

If E(l percent, 1 cm.) values are high for vitamins D
plus sterols, as well as sterols, it must be assumed that some

contaminent (peroxides) is causing the trouble. If, on the

(10)

other hand, the E(l percent, 1 cm.) is normal for vitamins D
plus sterols, and high for sterols, it must be assumed that

vitamins D are not being removed completely from the sterols.

Sample
No.

47761

p6846

(11)

Table II

Comparative Vitamins D Values of Fish Liver Oils

by Physical—Chemical and Biological Methods

height

of

Sample,

Grams

l

.8

1% Biological

E Calcd. Method

1 cm. U.S.P. U.S.P. Diff. Ewing &

soohuc Units/g. Units/g. % Kingsley (5)
A. High Vitamin D Fish Liver Oil

0.79 15,250

0.80 15,450

0.77 14,875

0.78 15,050

0.77 14,875

0.77 14,875

0.78 15,050

0.77 14,875

0.76 14,675

0.77 14,875

0.77 14,875

0.78 15,050

0.77 14,875
Av. 0.775 14,975 15,000 -0.17% 15,000

0.99 19,100

0.99 19,100

1.05 19,875

1.02 19,675
Av. 1.01 19,500 20,000 -2.5% 20,000

(12)

Table II (cont'd.)

Weight 1% Biological
of E Calcd. Method
Sample Sample, 1 cm. U.S.P. U.S.P. Diff. Ewing &
No . Grams 500 be Units/g. Units/g. % Kingsley (5)
41860 1 0.79 15,250
0.77 14,875
Av. 0.78 15,000 16,000 -6.2% 15,400
57481 1 0.70 15,500
0.75 14,100
Av. 0.715 15,725 16,500 —16.7% 14,800
65221 1 0.71 15,700
0.71 15,700
Av. 0.71 15,700 14,000 -2.1% 15,550
76892 1 0.80 15,450
0.81 15,625
Av. 0.805 15,525 16,000 -5.0% 15,450
28285 .8 0.60 11,575
0.62 11,975
Av. 0.61 11,775 11,500 -+2.4%
29265 .8 0.67 12,925
0.65 12,150
Av. 0.85 12,550 15,500 -7.7%

f.

u‘
\c
H‘
15
\
. .
| . ‘
u C
I‘ .\ !\
u .L r
0 .
..
\ .4
O I
D A-
-.. ¢\
7 G
0 ‘ w!;
I

.7.
J

 

3

(15)

Table II (cont'd.)

Weight 1% Biological
of E Calcd. Method
Sample Sample, 1 cm. U.S.P. U.S.P. Diff. Ewing &
No. Grams 500l94: Units/g. Units/g. % Kingsley (5)
21275 .8 0.65 12,550

0.60 11,575

Av. 0.825 12,050 12,000 -+0.42%

29655 .8 0.47 9,075
0.48 9,275

Av. 0.475 9,175 11,500 -21.8%

Hg 1 .8 0.85 18,400
0.82 15,825
0.85 18,400
0.81 15,825
0.81 15,825

65251 1 0.77 14,875 14,000 -+6.2% 15,000

E. Vitamin D Fish Liver Oil Concentrates
44470 .07 10.7 206,500
11.0 212,500

Av. 10.85 209,400 240,000 -125 255,500

26695 .5 2.58 45,925
2.27 45,800

Av. 2.52 44,775 45,000 —0.5%

(14)

Table II (cont'd.)

Weight 1% Biological
of E Calcd. Method
Sample .Sample, 1 cm. U.S.P. U.S.P. Diff.
No. Grams 500 31,2 Units/g. Units/g. %

C. Vitamin D Fish Liver Oil Distillates
44080 .5 1.54 29,725
1.52 29,525

Av. 105?) 29,525 51,000 "5%

Ewing &
Kingsley (5)

29,700

(15)

Table III

Results of D. H. Baker Using Modified Procedure

Weight 1% Biological
of E Caled. Method
Sample Sample, 1 cm. U.S.P. U.S.P. Diff.
No. Grams 500.8u2 Units/g. Units/g. %

47761 1 0.748 14,400
0.757 14,200
0.771 14,900
Hg 1 .8 0.855 18,400
0.840 18,200

Av. 0.848 18,500 17,500? —6.9%
p6846 .8 0.99 19,100
0.99 19,100

[2V0 0099 19,100 20,000 ’405%

(16)

PART II

Studies in the Chromatography of Irradiated Ergosterol
on Activated Superfiltrol

Chromatography is the process of resolution of mixtures
by adsorption on Tswett columns. The principle objectives
attainable through use of chromatography are (l) as listed:

(1) Resolution of mixtures

(2) Determination of homogeneity of chemical
substances

(5) Purification of substances

(4) Quantitative separation of one or more sub-
stances from complex mixtures

(5) Determination of molecular structure

The following work is an attempt to apply some of these
uses to irradiated ergosterol in vegetable oil.

Irradiation of ergosterol in solution with ultraviolet
light leads to the formation of the anti-rachitic Vitamin D2
(calciferol). The photo chemical reaction which takes place
has been the subject of extensive investigation and it is now
believed that a series of reactions take place according to the

following scheme (2):

Ergosterol-erLumisterol-e>.Tachysterol-+-Calcifer01 (D3)-s>
Toxisterol L-Suprasterol I
Suprasterol II

Investigation seems to indicate that the lumisterol is

(17)

formed which in turn is converted into tachysterol and this
is converted into calciferol (5). Therefore, upon irradi-
ating ergosterol a mixture results which contains all or
various amounts of the above substances. ‘

Apparently, no chromatographic work has been done on
irradiated ergosterol. Esters of lumisterol have been puri-
fied by filtration of their solutions through a column of
alumina (4). The most successful methods of separating the
constituents of irradiated ergosterol have so far been purely
chemical, and this, of course, has not been quantitative.

The application of the physico-chemical procedure of
Ewing and Kingsley (5) to irradiated ergosterol dissolved in
corn oil have not been very satisfactory. Using this method,

the results are very often low.

Equipment and Reagents.-The equipment and reagents used
were the same as used in the work by Ewing and Kingsley (5).
The ether, however,was dried over anhydrous sodium sulfate

instead of phosphorus pentoxide before adding sodium.
Experimental Procedures

In the procedure develOped for natural fish liver
oils (5), activated SUperfiltrol was used as the adsorbent
and a skellysolve-ether-ethyl alcohol (50-10-1) mixture was
used as solvent and developer. In order to correlate results
with the original procedure, it was decided to use the same

adsorbent and developer. The term "irradiated ergosterols“

(18)

refers to solutions of irradiated ergosterol in vegetable
oil.

Procedure A.-The sample of the oil is weighed out and
dissolved in 10 ml. of purified chloroform. A 1 m1. ali-
quot of this solution is taken and to it is added 10 m1.
of antimony trichloride reagent. The mixture is swirled
for thirty seconds. Exactly three minutes after adding re-
agent, the extinction is measured on the Bausch and Lomb
visual photometer, using a 1 cm. cell. This value of
log. Io/I is converted to E(l percent, 1 cm.). This is
multiplied by the factor 19500 to convert to D units per
gram of sample. The above factor was determined eXperi-
mentally from work on Vitamin D3. Hewever, since Vitamin D2
gives the same maxima at SOOZyz as D3 with antimony tri-
chloride reagent, no error should be introduced by using
this factor.

Procedure B.-The sample is weighed out and to this is
added 10 ml. of N/2 alcoholic potassium hydroxide and the
resulting mixture is kept at 70°C for one hour with frequent
swirling. After this saponification treatment, the non-sa-
ponified residue is extracted with anhydrous-peroxide free
ether, using the same procedure as outlined by Kingsley and
Ewing (5). The resulting ether solution was evaporated to
dryness under reduced pressure and the residue taken up in
10 ml. of purified chloroform and the extinction measured as

in Procedure A.

(19)

Procedure C.--The sample of oil is weighed out and
this is dissolved in 5 ml. of skellysolve-ether-ethyl alco-
hol (50-10-1). This is added very carefully to a chromato-
graph column containing superfiltrol which has been previ-
ously wet with 10 ml. of the above solvent. The flask is
then rinsed with 5 ml. more of the solvent and this is
added to the column just before the previous 5 ml. disap-
pears. A pressure differential of 6 cm. of Hg is maintained
up to this point. The column is then develOped with 55 ml.
of the above solvent at a pressure differential of 8 to 10 cm.
of Hg. As the column develops a faint band will be noticed
moving down the column. After develOping,the material down to
2 mm. below this band is removed and discarded. The remaining
portion is eluted with 25 ml. of purified ether. The result-
ing filtrate is evaporated to dryness and taken up in 10 m1.
purified chloroform. The extinction is measured as in Proce-
dure A.

Procedure D.-This is Procedure B and C combined. The
oil is first saponified and then run through the chromato-
graph. The extinction is measured as in Procedure A.

Procedure E.-This is Procedure C except that the column
is cut down to 2 m1. above faint band. This material is re-
packed into column after removal of band. The column is then
eluted with.25 m1. of purified ether. The extinction is mea-
sured as in Procedure A.

Procedure F.--Ewing and Kingsley (5).

(20)

The following oils were used in this work:
5772 Vit. Dz standard; 1 gm . 200,000 Du/gm (52,000,000 Du/gm)
12722 a a w a a w n u a

66701 Irradiated ergosterol in corn oil, Parke, Davis & Co.

45120 ! a a w u a n a
78272 a 1 a a w n w a
15155 " a a m n u w a
13192 3 3 fl 1 a U n a
84742 a n n w m a m a
26595 " “ in sesame oil, 3 ' fl

G ’ 7-dehydro-cholesterol in corn 011, National

Oil Products Co.

F ' ergosterol in corn oil, National Oil Pro-
ducts Co.
B5494 ” ' in corn oil, Gelatin Products Co.

An examination of Table I reveals the following facts:

In the majority of cases irradiated ergosterols give re-
sults by Procedure A which are lower than the bioassay. In
some cases, however, the results are higher. There is no cor-
relation between deviations from the bioassay. If the bio-
assays are correct to within fifteen percent, the following
conclusions seem to be indicated:

(a) In the case of oils which give high results with

Procedure A, there must be additional substances present which

(21)

give a color reaction with antimony trichloride reagent.
Lumisterol, tachysterol, etc., are known to give a reaction
with the reagent.

(b) In the case of low results some of the calciferol
seems to be in a form which does not allow it to enter into
the color reaction. It is quite possible that this unavail-
able calciferol is tied up as an addition compound with lu-
misterol (6). Saponification with N/2 alcoholic potassium
hydroxide fails to break any such addition compound.

(0) Upon chromatographing an irradiated ergosterol
according to Procedure C, there is some holdback of calci-
ferol in the column. This is not true for the calciferol
standards which are made by dissolving pure calciferol in
corn 011. Hence, again, it seems to be indicated that cal-
ciferol is tied up in an unavailable form in irradiated er-
gosterols. The material held back can be washed out of the
column with ether as is shown in Procedure E. It is of in-
terest to notice that Procedures A, B, and C give the same
result with calciferol standard 5772, whereas Procedure D
causes a lowering of about twenty-five percent in this same
oil. Apparently, saponification causes calciferol to be
held back on the chromatograph column.

(d) It is also of interest to observe that over the
course of several months, the value of Oil F by Procedure A
decreased from 656,000 to 440,000 Du/gm. This could be due

to decomposition or conversion of calciferol into an

 

 

(22)

unavailable form such as an addition compound with lumisterol.
(e) From this work it follows that a successful method
for the determination of calciferol in irradiated ergosterols
must involve the following steps:
(1) Conversion of unavailable calciferol to
free calciferol
(2) Separation of free calciferol from in-

terfering substances such as lumisterol, etc.
Special Studies

The Nature of the Chromatogram.-A solution of irradi-
ated ergosterol in vegetable oil forms a very simple visible
chromatogram when taken up in skellysolve-ether-ethyl alco-
hol (50-10-1) and passed through a column of superfiltrol.

Shortly after adding the solution, a very faint orange
band forms about 1 cm. from the surface of the superfiltrol.
This band slowly progresses down the column upon developing
with the above solvent. Under a pressure differential of
8 cm. of mercury, this band moves 5 cm. with 55 m1. of de-
veloper. No other bands are visible and the same chromato-
gram is obtained when the oil is saponified and extracted
with ether before chromatographing. The band in the latter
case, however, seems to be somewhat sharper and more intense.
If the ether content of the develOper is increased, no band
is formed. If the ether content is decreased, the band moves
down the column much slower. With a decrease in ether content,

an increase in alcohol causes the band to move somewhat faster.

 

(23)

By examining aliquot portions of the filtrate as it comes
through the superfiltrol column, it appears that nothing
comes through until the last 10 ml. of the 55 ml. of de-
veloper passes through the superfiltrol. This seems to in-
dicate further banding which is, however, invisible. Ex-
amination under ultraviolet radiation shows no such banding.

Identification of Above Band.--Since pure solutions of
calciferol show only an extremely faint band of the above
type, it was suspected that the band formed was either resi-
dual ergosterol or one of the intermediate substances. Upon
passing a solution of pure ergosterol in skellysolve-ether-
ethyl alcohol (50-10-1) through a column of superfiltrol,
the band described above again formed. The faint band was
cut out in some instances of runs made on irradiated ergos-
terol and the material eluted with ether. Upon filtering
and evaporating off the ether, the residue gave a pink color
reaction which is characterie of ergosterol.

To further confirm this work varying amounts of pure
ergosterol was added to samples of the calciferol standard
5772. These were then chromatographed as in Procedure C.
The band formed in the sample containing no added ergosterol
was extremely faint. In the samples containing ergosterol,
the sharpness and density of the band noticeably increased
with increased concentration. of ergosterol. The resulting
filtrates were evaporated to dryness, taken up in 10 ml. of

chloroform and E(l percent, 1 cm.) determined as in Procedure A.

(24)

Table II shows that there is no increase in the
E(1 percent, 1 cm.) due to ergosterol coming through the
column. It is doubtful if a quantitative separation is
possible if the concentration of ergosterol is much higher
than in Table II. Attempts to remove ergosterol by use of

digitonin were not very successful.

Adsorption of Calciferol on Superfiltrol

from a Solution in Benzene Skellysolve (2:1)

For this work a calciferol solution in skellysolve was
used (General Mills). One milliliter of this solution upon
evaporating to dryness gave an extinction of E = log Io/I = .42
by Procedure A. One milliliter samples of this solution were
evaporated to dryness and taken up in 25 m1. of benzene-skelly-
solve (2:1). This was shaken with approximately five grams of
superfiltrol for various periods of time and then filtered,
evaporated to dryness and the extinction determined as in
Procedure A. Table III shows the variation of log Io/I with

time. The results are also shown in Fig. 1.

Effect of Added Ergosterol to Calciferol Standard

.07 g.
.07 g.
.07 3.

Oil No.
.07 g. #5772

(25)

Table II

+ .0002 g. ergosterol

+ .0004 g. '
+ .001 g. '
Table III

.60
.61
.60
.60

EU}? 1 cm.) Du/g.

181,400
185,400
181,400
181,400

Adsorption of Calciferol on Activated Superfiltrol

Time (min.)
Extinction (log Io/I) .42 .05

0 5

10
.02

20
.00

 

 

 

 

0.40

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

H 0.50
\
o
H
to
o
'4 i
0.20 \
0.10 \\
E :i—L
10 15 20

Time in.minutes

Fig.1.--Adsorption of calciferol from.a solution in
benzene-skellysolve (2:1) upon activated superfiltrol.

£41

5772 200,000

12722 200,000

66701

45120

78272

15155

(26)

Table I

Procedures A, B, C, D, E, F Applied to the Calciferol

Bioassgy

250,000

250,000

225,000

Standards and Irradiated Ergosterols

A
191,000
187,200
191,000

191,000

208,400
208,400
208,500

208,500

189,800

‘196,000

191,000

198,900
196,900

179,000

B
195,000
187,200
191,000

187,200

187,200

198,800

142,800

189,800

145,500

Procedure

0
191,000
191,000
191,000

198,800

118,500

158,900

81,800

151,800

40--..ll“

150,500
148,600

144,700

E
181,400

181,400

191,000

191,000

158,400

150,500

F

108,500

140,000
148,000

142,000

155,900

127,400

 

93.} 11.285881

 

15192 250,000

84742 275,000

28595 400,000

C 225,000

F 440,000

A
188,000
187,200
179,100

187,200

242,800
227,500

221,900

885,900
829,200
665,900
858,900

858,900

152,500
152,500

151,500

858,900
592,500
579,000

579,000

(27)

Table I (cont'd.)

B
185,900

185,980

205,200

507,800
480,800

528,800

154,400

154,400

509,500
528,500
528,500

525,000

Procedure

C

84,900

90,900

90,900

549,700
549,700

504,900

144,250

588,800
588,800
580,900

588,800

D

280,500

B
189,800
142,000

187,200

145,520

187,200

550,750

550,750

114,800

116,800

78,600

108,100

545,500

Oil Bioassay

B5494 450,000

A
528,800
525,000
525,000

440,000

700,800
887,800
755,500

745,000

(28)

Table I (cont‘d.)

854,500

585,500

Procedure

C

498,000
585,200
585,200

579,000

D

589,900

 

(29)

SUMMARY

Part I.-A modified procedure for determining vitamins D
in fish liver oils has been develOped and applied to a number
of oils. By the use of this modification, the original pro-
cedure is greatly simplified and shortened.

Various causes of errors in the use of the procedure are

discussed.

Part II.-In the majority of cases, the calciferol in
irradiated ergosterols is in an unavailable form for deter-
mination by the use of antimony trichloride reagent. Saponi-
fication and chromatographing fails to convert the unavailable
calciferol to an available form.

The nature of the irradiated ergosterol chromatogram from
a solution in skellysolve-ether-ethyl alcohol (50-10-1) upon
activated superfiltrol has been investigated and the one band
formed has been shown to be due to ergosterol.

Separation of ergosterol from samples of calciferol in
oil by chromatography has been accomplished.

The rate of adsorption of calciferol in benzene skelly-
solve (2:1) has been investigated and shown to be very rapid

and complete.

(1)

(2)

(4)

(5)

(6)

('7)

LITERATURE CITED

Harold H. Strain, “Chromatographic Adsorption Analysis",

Interscience Publishers, Inc., New York, N. Y.

C. Ellis and A. A. Wells, “The Chemical Action of Ultra-
violet Rays", Reinhold Publishing Corp., New Yerk, N. Y.

P. Setz, “Photochemische Umwandlungen der Bestrahlungs-
produkte des Ergosterins”, Zeit. Physiolog. Chem., 215,
185 (1955)

A. Windaus, M. Deppe, and W. Wunderlich, aSome Products
of Illumination of 7-dehydro-cholesterol", Ann. 555,
118 (1957)

D. T. Ewing, G. V. Kingsley, R. A. Brown, and A. D.
Emmett, aPhysical-Chemical Method for Determination of
Vitamins D in Fish Liver Oils", Ind. & Eng. Chem.,,lg,
501 (1945)

H. R. Rosenberg, 1'Chemistry and Physiology of the Vita-

mins", Interscience Publishers, Inc., New York, N. Y.

A. Zimmerli, C. H. Nield, W. C. Russell, ”Letters to
the Editor, Jour. Biol. Chem., 1, 148 (1945)

 

 

, . l c . t . . . ., o . ... . | . . ... .~_. - .1. I. . t A .
v . .5 I V .. K . 4. ..s i u _ w an) \I. ... .... ... a. . .. .t
.. x . 1.. . O ’ . . r b... o . . . ..h’u..-. x; ‘._ ..4 ..., . . .
I o. no“... . flosv. r. VPUWé J, v 41 1“. (....wd 1.1 I- . ... $0er .2 .....o....“§.. .s O‘LV. . . ... 4 .\ .
..ab u. u Is... .7. . ‘91.. . var- ol‘fl‘ . ....I. 3..., . .....f .....L...» s: ‘Xx . 3‘ ~\n.m.......4. .. . .n .
. . m . t .. rt. .... ..J. ... . . . . ...A. Tuvh‘...
. .. . . . . . .. .. ....
I u‘ I .1 ' ‘ V . .I .0 I!“ c O n . I n .
I ~
N
. v A
o . .n
. . n r 4 . .v»
I l‘
. . I
. ..~ .
I I .I.Q .
s . t.
- .. u
.
w v
\ ‘
o

 

T541

method for
=—vitam§nsuj
Pt.2: Stu‘

 

     

)

  

 

activated

 

!Pt.l: A modified

.es in the c

i
. I
H141
Rage

H141

Hage

O

superfiltrol

155998

155998

physical—chemical
the determination 01

82 aagsrwwm

romatogra—

stem—0L—

 

 

 

 

 

 

 

 

 

 

 

 

‘.
( I
l
I
-\
|
I
l
I
' I
I
o
.\ .
' I
I Iv]
‘ i
' l
‘ \ I
.‘i .
‘.l
«,4 - 3.,
“1"} I
' , I' I
t
‘ \
' I
I
.'l II
II
.,I‘ I
.'v \ A4.
..4' .‘

\
I
L V
’ I
.
I
. l
6'
I
‘ I
v .
. |.
I W ‘
l K
I
\
I
1
t"
\
‘ I
‘7‘
a). .
I I
‘ I
\
)
..., l
'. ‘.
Y . \
.~ ‘ I
K
. ' _‘
l )4,
\
~ I
I
q:
1 .
Q
I.
..

ilNIHIIHIHHHI‘NHIIIHIIIHHHI

31293 02446 7288