CHEMICAL AND BIOLOGICAL STABILITY OF CRYSTALLINE VITAMINS Dz AND DB AND THEIR DERIVATIVES

BY W. HUBER AND 0. W. BARLOW

(From the Research Laboratories, Winthrop Chemical Company, Tnc., Rensselaer, New York)

(Received for publication, April 6, 1943)

Fritz, Halpin, Hooper, and Kramke (1) presented evidence that vitamin D is not stable when added to various dry carriers as it must be added of necessity to mixed feeds. The carriers used include dried whey, sucrose, oyster shell, mineral mixtures, sand, kaolin, and MnS04 in mixtures with such solvents as ether, corn oil, tuna oil, and the U.S.P. reference oil with a menstruum such as cereals and ground corn. Instability is due to num- erous substances or conditions that promote oxidation. Temperature coefficients and other data indicate that the destruction is due to a chemical reaction.

Thus, all forms of vitamin D, including activated ergosterol in a crystal- line form or when adsorbed on casein (ertron), activated T-dehydrocho- lesterol (crystalline vitamin D3), as well as vitamin D from natural sources, are susceptible to destruction. This change occurs more rapidly in a dry dispersed state than in emulsions.

Various methods for stabilizing vitamin D have been tested. The most effective protection is afforded by coatings which prevent contact with air. Thus, coating agents such as dried whey with hydrogenated fats and cal- cium stearate impart materially more stability than does whey alone. Mineral mixtures deteriorated rapidly, but were satisfactorily stabilized by hydrogenated fats or even molasses. Packing in inert gases additionally retards the rate of destruction.

Baird, Ringrose, and MacMillan (2) reported that the vitamin D con- tent of chick ration fortified with cod liver oil and stored at summer tem- peratures for 32 weeks showed no measurable destruction. Jung (3) noted that peanut oil solutions of vitamin DS were less stable than similar solu- tions of vitamin Dz, but the stability on addition of stabilizing substances such as yeast, alfalfa meal, and CaC03 was satisfactory.

The present report deals with the stability of crystalline D vitamins and their derivatives per se, as well as when incorporated in pharmaceutical preparations.

During the last 4 years large amounts of vitamins Dt and Da and their derivatives have gone through our hands and we have tried to establish their data in terms of melting point, specific rotation, and ultraviolet ab-

125

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Vita

- m

in D2

Da -

TABL

E I

Melt

ing

Poin

ts,

Spec

ijic

Rota

tion,

an

d Ul

travio

let

Abso

rptio

n of

Vi

tam

in

Dz

and

D3

Este

rs

Este

r

4-Ni

trobe

nxoa

te

(4)

t

3,5-

Dinit

robe

nzoa

te

(5)

4.Meth

yl-3n

itrobe

nzo-

at

ei

4-M

et.h

yl-3,

5-din

itro-

benz

oate

(6

) 4.

Nitro

benz

oate

(4

)

3,5-

Dinit

robe

nzoa

te

(4)

4-M

ethy

l-3,5

-dini

tro-

benz

oate

s

M.p

.

“C.

94.5

- 95

146

-147

119

-120

116

-117

125

-126

132

(Ben

zol-

met

hano

l) 14

1 (E

ther)

128-

129

-

Spec

ific

rota

tion

r-l?

1.6

per

cent

so

lution

degr

ees

t105

.2

Chlor

ofor

m

t107

.4

Acet

one

+91.

5 Ch

lorof

orm

+8

6.5

Acet

one

t106

.8

Chlor

ofor

m

j-105

.0

Acet

one

+95.

8 Ch

lorof

orm

j-9

2.2

Acet

one

j-114

.6

Chlor

ofor

m

j-116

.4

Acet

one

+97.

9 Ch

lorof

orm

+9

6.3

Acet

one

+106

.6

Chlor

ofor

m

flOl.2

Ac

eton

e

1 _

. I

_-

Max

i- m

um

E X

10-S

A 261

30.8

f

0.4

265$

22

.5

f 0.

4

261

22.2

f

0.4

2653

21

.0

3~

0.4

261

30.8

f

0.4

265$

22

.8

f 0.

4

265$

20

.8

f 0.

4

Appe

aran

ce

Stur

dy

prism

s

Prism

s,

from

ac

eton

e.

Need

les,

from

ch

lo-

rofo

rm-e

than

ol Ne

edles

Fine

ne

edles

fro

m

hexa

ne

Prism

s

Fine

ne

edles

Very

fine

need

les

from

ac

eton

e

COlW

Brigh

t ye

llow

Cana

ry ”

Pale

ye

llow

Light

“

Very

light

ye

l- low

Cr

eam

Light

ye

llow

Ultra

violet

ab

sorp

tion*

* Th

e ab

sorp

tion

spec

tra

were

tak

en

by

Dr.

G.

W.

Ewing

in

ou

r lab

orat

ories

wi

th

a Be

ckm

an

quar

tz sp

ectro

phot

omet

er,

mod

el DU

. t Th

e fig

ures

in

pa

rent

hese

s re

pres

ent

bibl

iogr

aphi

c re

fere

nces

. $

This

is

not

a

true

max

imum

bu

t an

ar

bitra

ry id

entif

icatio

n va

lue

on

the

slope

of

th

e ab

sorp

tion

curv

e,

the

true

max

imum

be

ing

outs

ide

the

rang

e of

ou

r ins

trum

ent.

$j T

hese

ar

e ne

w co

mpo

unds

. Th

eir

prep

arat

ion

and

their

an

alyt

ical

data

wi

ll be

pu

blish

ed

shor

tly

in

the

sect

ion

“New

co

m-

poun

ds”

of

the

Jour

nal

of

the

Amer

ican

Chem

ical

Socie

ty.

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

W. HUBER AND 0. W. BARLOW 127

sorption with the greatest possible accuracy. The esters of vitamins Dt and D3 with various nitrobenzoic acids are not only characteristic deriva- tives, but are also optimally suited for storage purposes from the standpoint of stability. They remain unchanged over a period of years if a slight photodecomposition is avoided by storage in amber bottles.

It is our experience that pure crystalline vitamins D should be prepared by direct saponification of rigidly purified esters rather than by recrystal- lization. The latter procedure is always accompanied to some extent by more or less decomposition based on determinations of the physical con- stants.

The data for melting point, specific rotation, and ultraviolet absorption on the 4-nitrobenzoates, 3,5-dinitrobenzoates, and 4-methyl-3,5-dinitro- benzoates of vitamins D2 and D3 are incomplete and widely scattered over

TABLE II

Melting Points, Specijc Rotation, and Ultraviolet Absorption of Pure Cr@alline Vitamins D,, Da, and Do-Cholesterol

Substance M.P

Vitamin Dz

6‘ DO

‘L Da- cholesterol

“C.

116-117*

84- 85*

119-120*

degrees

+48.2 +s3.4 +51.9 +84.8 +26.4

1.6 per cent solution

ultraviolet absorption,

265 X)

Chloroform 19.50-19.70 Acetone Chloroform 18.90-19.20 Acetone

I‘ 9.45- 9.60

Appearance (color, white)

Fine clustered needles

Needles

Fine clustered needles

* The sample was dropped into a bath 15” below the melting point. The rate of heating was 6-7” per minute.

the literature. Table I shows the physicochemical data of these esters and of the 4-methyl-3-nitrobenzoates of vitamins Dz and Da, which in our ex- perience are characteristic for the pure compounds.

Anderson, Bacharach, and Smith (7) have investigated extensively the properties of crystalline vitamin D,. We are able to confirm their values, although in our experience the E ::m. values are not scattered over quite such a wide range. Table II shows data which in our experience are the con- stants in terms of melting point, specific rotation, and ultraviolet absorption of pure crystalline vitamins Dz and D3 (8) and the vitamin DB-choles- terol double compound. The latter (9) is an addition product or com- pound of vitamin DS and cholesterol in molecular proportions. It can be obtained from crude vitamin DS resins in over 90 per cent yield and thus represents the most economical way to obtain a crystalline vitamin Da

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

i

- i

-

-

-

-

-

-

7 : - : c2 F

.- .!

c -

-

-

-

-

-

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

W. HUBER AND 0. w. BARLOW 129

preparation of standard purity. Pure crystalline vitamin Da itself is difficult to obtain, the yields varying from 40 to 80 per cent.

It is a wide-spread assumption that pure crystalline vitamins D, and Ds are stable compounds. From our experience, we cannot concur entirel) in this opinion on the basis of observed physicochemical characteristics. Such stability data under various conditions are illustrated by Table III.

DISCUSSION

The various nitrobenzoic esters of vitamins D, and D3 can be stored for at least 5 years at room temperature without decomposition. They can be rigidly purified by recrystallization until their physicochemical data con- form to the characteristics in Table I. Pure crystalline vitamins D, and D3 with constants conforming to those in Table II can be obtained by direct saponification with methanolic KOH from these esters.

Pure crystalline vitamin DP, when sealed in amber evacuated ampules and stored in the refrigerator, does not change in appearance and in its physicochemical constants up to 9 months. With storage at refrigerator temperatures under air, decomposition occurs after 1 to 2 months, as indi- cated by the appearance of small oil deposits in the crystalline mass, to- gether with a decrease of the melting point and of the rotation. A similar detectable decomposition is noted after storage in sealed, air-filled ampules at room temperature for 2 to 3 days. Inert gases such as nitrogen or car- bon dioxide very materially diminish, but do not prevent, the decom- position.

Deterioration of pure crystalline vitamin Da is negligible after 12 months storage in amber evacuated ampules at refrigerator temperatures. In general it is more stable than vitamin DP when tested in the same manner. It seems well suited to serve as a standard reference preparation, provided it is stored under proper conditions. It also seems advisable to prepare fresh material every 9 to 12 months from rigidly purified esters with a standardized saponification procedure.

Vitamin Da-cholesterol, although suited for the economical production of a crystalline vit,amin D, preparation, shows a slightly inferior stability, as compared with vitamin Da.

The above results indicate strongly that crystalline vitamin D prepara- tions, unless handled with great care, are apt to lose potency in short order. Crystalline vitamin D preparations should be stored under optimal condi- tions and if the compounds are to be used for chemical studies their physi- cochemical constants should be rechecked if aged more than 12 months, although such changes on the basis of acceptable standards (40,000 inter- national antirachitic units per mg.) are not demonstrable by biological tests on rats or chickens.

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

130 STABILIlY OF VITIMINS Dr ASD D>,

&ability of Vitamin D on Basis of Biological Assay

The degree of utilization of vitamin D differs materially, depending on the medium in which it is administered. Lewis (10) demonstrated that crystalline vitamin D, is clinically more effective when administered in t.he daily ration of milk to the child than when given in oily solution. His findings in this regard have been confirmed by Eliot et al. (II) and by Shelling (12).

Reports on the stability of vitamin D in various menstrua, however, are thus far meager and have not definitely settled all aspects of the question. Bourdillon, Bruce, and Webster (13) studied the stability of olive oil solu- tions of irradiated ergosterol. They found that such solutions, when stop- pered and kept at O”, showed no deterioration, but that the vitamin D values of similar samples kept at room temperature deteriorated materially. The authors estimated, by extrapolation of their data, that the solutions kept at room temperature would lose half of their potency in 3 years.

Bruce, Kassner, and Phillips (14) studied this problem in a more detailed manner. Solutions of crystalline vitamin Dz in olive oil, cod liver oil, halibut liver oil, and paraffin oil were examined for stability; all solutions were sealed under nitrogen and were kept at 0”. These studies indicated that deterioration of the vitamin D occurred in all of the menstrua with the exception of the solutions in paraffin oil. However, the changes noted were not great, and the authors felt that the losses were within the limits of the experimental error inherent in biological assays for vitamin D. Such was certainly the case where one solution showed an apparent deterioration of 35 per cent after 11 months storage, but only 7 per cent after 19 months. Natural vitamin D kept at 0” under nitrogen for periods up to 3 years showed no change, except that the solutions in halibut liver oil seemed to be consistently less potent after various periods of storage.

On the other hand, Shelling (15) has shown that aqueous emulsions of oily solutions of vitamin D, although clinically very effective, deteriorate rapidly in the presence of air. Under these conditions, the antirachitic values of such emulsions deteriorated more than 90 per cent in 6 months. However, when the emulsions were stored for corresponding periods under nitrogen, only minor losses were observed. Studies on the stability of vitamin 1~2 in water and milk were also reported by Supplee et al. (16). These authors reported that propylene glycol solutions of vitamin D were unstable when diluted wit>h water, but entirely stable when diluted with milk during the period of assay.

According to Fuchs and van Niekirk (17), crystalline vitamin D, sealed in ampules under air loses from 35 to 55 per cent of its potency during 2 years storage in the dark. The literature, therefore, indicates that the

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

W. HUBER AND 0. W. BARLOW 131

stability of vitamin D, to aging, either as crystals or in solutions, is de- pendent largely on the protection of the vitamin from oxygen.

We have been interested for some time in the stability of vitamins Dz and Da in various menstrua. This paper presents a comparison of the relative stability of suspensions of vitamin D in water and milk and of solu- tions in vegetable oils or in propylene glycol. In view of the reports in the literature, it has seemed desirable to report stability data on solutions kept under the most severe conditions of storage likely to be encountered in practice.

TABLE IV

RecalciJication Values

Product

U.S.P. reference cod liver oil, No. 2-36. Distribution date, Oct. 17, 1936; dilutions with peanut oil; dose in 0.1 cc...

Drisdol, Lot DP-3319, No. 917. Room temperature 3 yrs.; dilutions with propylene glycol; dose in 0.1 cc. (Sample A).

Lot DP-3319, No. 917. Room tempera ture 3 yrs.; stock solution in water, then added to milk; dose in 4 cc. by stomach tube (Sample B). .

Lot DP-3319, No. 917. Same stock (water) solution as used for milk, diluted with distilled water; dose in 4 cc. by stomach tube (Sample C)

-

-

-

Daily dosage level (10 days)

- 0.01875 Y

2.79+

3.0+

0.36+ -

7.5 mg.

2.85+ 3.17+

0.05 Y

3.33+

3.25+

O.SS+

0.1 Y

Not tested

3.44+

1.55+

Methods

The several preparations were tested for vitamin D according to t,he procedures described in the United States Pharmacopoeia XI, or Volume XI, Second supplement. Eight to twelve animals were employed at each dosage level tested. All tests were completed prior to March, 1941.

Stability of Crystalline Vitamin Dz in Various Solvents

Table IV presents a comparison of the antirachitic potency of the U.S.P. reference oil and of crystalline vitamin Dz in propylene glycol, when

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

132 STABILITY OF VITAMISS D2 AND Ds

diluted (1) with propylene glycol, (2) with water, and (3) with milk. These data were obtained in 1936, at which time the solution of vitamin D, in propylene glycol (drisdol, Lot. DP-3319) stored at room temperature under test, was more than 3 years old. It had not changed in potency during that time.

Water-Vitamin D, in propylene glycol, when diluted with water under the condiCons described by Supplee et al., deteriorated markedly; i.e., a loss of more than 75 per cent occurred during the period of the bioassay even though the samples of aqueous supplement were stored under re- frigerator conditions. When the above freshly prepared aqueous stock

TABLE V

Effect of Storage at LO” on Potency of Solutions of Vitamin Dz in Propdene Glycol

LeYh storage -.__

wks.

Fresh 3

13 17 26 30 52

114 169 229

No. of lots

tested f unit

R. O.* V

I--

itamil Dt

1681 1.16 1.60 42 1.28 1.84 36 1.25 1.52

6 - -

50 0.95 1.49 4 - -

50 1.15 1.57 25 - -

2 - -

2 - -

1

Average recalcification values (f)

: unit

R. 0. ‘itamin D

I

2.49 2.87 2.51 3.10 2.47 2.89 2.56 2.61 2.25 2.83 2.54 2.81 2.26 2.78

- - - -

- -

1 unit

R. 0.

2.76 2.87

-

-

-

-

2.79 2.70 2.70 2.70

‘itamir D

3.06 3.16

- - - -

3.12 3.10 3.07 3.05

-

-

L

14 units .-

‘itamir 1 D R. 0.

3.54 3.32 - 3.47

3.46 3.31 3.36 3.44 3.23 3.16 3.39 3.43 3.32 3.21

- -

- -

-

-

2 units -- iitamin

D

3.44 3.66 3.51

- 3.13

- 3.38 - - -

* u.s.P. reference oil. i Vitamin D preparation. The estimated potencies are based on the assumption

that 1 mg. contains 40,000 U.S.P. units. $ Most lots were tested at three dose levels. The figures given are averages of

all available data on these 168 lots.

solution of drisdol was further diluted so that the supplement samples were suspended in milk, vitamin D assay values of the reference oil, drisdol Lot DP-3319 in propylene glycol (Sample A), or dilutions of Sample A in milk differed in no essential respect. Our observations indicate that drisdol per se is stable, but confirm both those of Supplee et al. and those of Shelling as to the instability of vitamin D in water (watery dilutions of propylene glycol solutions) and those of the former authors when the vitamin D preparation was diluted with milk.

Propylene Glycol---Solutions of crystalline vitamins Dz and D, were pre- pared to contain 0.0257 per cent of the vitamin by weight, Samples were

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

TV. HUBER AND 0. W. BARLOW 133

stored in full amber bottles at 5”, at room temperature, and at 38-40”. The bott’les were well sealed, but no attempt was made to displace air with CO2 before they were sealed. The incubator samples were subsequent’ly retested at somewhat longer intervals. Refrigerator samples were assayed occasionally simply as a control procedure. The results of these studies are shown in Tables V and VI.

ITegetcxbZe Oils-Solutions of crystalline vitamins I>2 and Da, 0.0257 per cent in sesame and corn oils, were prepared and stored under the same conditions as described above. The incubated samples were kept in one

TABLE VI

Effect of Storage at 40” on Potency of Solutions of Vitamin Da* in Propylene Glycol

Length of storage

- I-

wks. ‘.

0 8

13 17 22 52 67

156

-__~--..-~-~ f unit

--- R.0.t Vitamin D$

1.10 1.20 0.75 1.57 0.94 1.63 1.55 1.50 1.22 1.60

- - - - - -

____

Average recalcification values (f)

i unit _____.-

R. 0. Vitamin D --__

2.25 2.38 1.86 2.71 1.94 2.75 2.10 2.55

- - - -

- - -

1 unit __-

R. 0. --

- - - -

2.80 2.64 2.71 2.53

k%unin D R. 0. Vitamin D ___- .-- -

-

-

-

3.00 2.70 2.88 2.83

3.12 3.39 3.00 3.28 3.00 3.30 2.55 3.15 3.50 3.50

- - - -

- I -

T 2 units

* Tests of material by physicochemical and biological methods indicated that it contained 85 to 90 per cent of pure crystalline l33.

t u.s.P. reference oil. $ Vitamin D preparation. The estimated potencies are based on the assumption

that I mg. contains 40,000 U.S.P. units.

case up to 31 months; in other cases only those at room temperature re- ceived extensive study. The results are shown in Table VII.

A comparison of assay data of Tables VI and VII indicates that oily solutions are antirachitically less potent in comparison with solutions in propylene glycol. It is suggested that either the rancidity of the oil or incomplete absorption may explain these differences.

M&-Solutions of vitamin Dz in propylene glycol were diluted with bulked samples of Grade A milk so that each quart bottle on refilling con- tained 400 to 2500 units (1.75 to 10 drops of drisdol). The fortified samples from a single delivery from a large distributor were stored in the refrigerator for periods of time over which the milk would be entirely satisfactory for human consumption; i.e., up to 8 days. An unopened 8 day-old bottle was

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

134 STABILITY OF VITAMINS D3 AND D3

used to prepare the oral supplements for each day of the assay. After careful correction of all assays, because of the antirachitic value of the milk

TABLE VII

E.ffect of Storage under Various Conditions on Stability of Oily Solutions of Vitamins DZ and Da

Average recalcc~ion values Length of

storage Solvent Vitamin Temperature

“C.

40 Room

“

40 40 40

R. 0: (1 unit)

vi(p$t

mm.

0 31 39 54

0 12 15 31

Corn oil I‘ I‘

Sesame oil “ “

D2 I‘

‘I

“

D3 ,‘

I‘

“

2.61 2.89 2.74 2.83 2.74 2.83 2.68 2.77

: unit

2.62 2.62 2.71 2.84

2 unit -.----

2.90 2.69 2.88 2.93

Corn oil ,‘ ‘I I‘ ‘I ‘I ‘I

* u.s.P. reference oil t Vitamin D preparation. The estimated potencies are based on the assumption

that 1 mg. contains 40,000 u25.P. units.

TABLE VIII

Effect of Storage under Various Conditions on Potency of Ertron -

Lot Duration Temper- of storage ature

Average recalcification values (f) -

I

. . -

+ unit 1 unit 1; units

“C.

5 Room

I‘

38 38

-__.

-

-

U.S.P. reference

oil Ertron

- 1.52 - - - - - - - - - - - -

U.S.P. U.S.P. ?ference Ertron reference

oil oil

-

-

3.00 - - - -

- -

-

woks.

A 0 B 0 A 4 B 8 ‘I 11 “ 6

A 13

2.59 2.67 2.50 2.65 2.55 2.15 2.75 2.38 2.70 2.22 2.75 2.25 2.75 2.00

per se, no evidence of deterioration of the vitamin I) potency of the fortified milk was demonstrable.

A test of the stability of vitamin II2 in condensed milk was also made.

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

w. HUBEH AND 0. w. BARLOW 135

Samples were fortified in the following manner. 14.5 ounce cans of a well known brand of condensed milk were purchased on the openmarket. An area on the surface of the cans was sterilized with an antiseptic and punc- tured. 0.1 cc. of a propylene glycol solution of vitamin D2 in suitable con- centration was then added to the contents of each can in order to bring the antirachitic value up to 500 U.S.P. units per can. All cans were then resealed with solder (air not replaced), and sterilized at 120” for 30 minutes. The samples were stored under various temperature conditions as previously outlined. The incubator samples (38.0-40.0”) were assayed after 6 months storage, and the samples at room temperature at intervals up to 15 months. None showed any loss in potency.

TABLE IX

Vitamin D Assays of Additional Samples of Ertron and Vitamins DZ and Da

U.S.P. reference oil

DO% Plus values

wlits

t 2.06 1 2.61 1; 2.97

y. of claim present.

-

Ertron, No. 1109, received June 14,194o; stored m ice

box .

Per cent PIUS of I.U.

VdUeS refer- ence oil

~ __.

2.57 95.0 3.03 __-.

99.15

Ertron, No. 997, received Feb. 21, 1940

) Vitamin Dz in oil

1 VitahiDs in

Incubated at 38” 11 mos.

Stored in ice 1 million units vitamin D per box 11 mos. cc.; prepared Dec. 17,193s; stored

in ice box

1.75 37.5 2.66 105.0 2.97

- ‘er cent of I.“. Plus refer- values

ence oil - .-

130.0 3.00

- -~

130.0

Per cent of I.“. refer-

ence oil

132.5

132.5

Xtability of Ertron

Ertron is a form of activated ergosterol prepared by the Whittier proc- ess.’ An ethereal solution of the activated material is dried on casein to obtain the commercial product. Stability tests by biological assays were conducted on the capsulated material as purchased. On the basis of earlier ultraviolet absorption experiments the activity of this preparation diminished 50 per cent or more after storage for 3 months at room tempera- ture. The observed antirachitic effects of ertron with time are illustrated in Tables VIII and IX.

The antirachitic potency of different lots of ertron purchased through usual channels varied over a range of 75 to 105 per cent of that claimed (50,000 units per capsule).

1 Manufactured by the Nutrit.ion Research Laboratories, Chicago.

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

136 STABILITY OF VITAMINS Dq AWD Dg

DISCUSSION

There is nothing to suggest that either vitamin Dz or DB in propylene glycol shows any change in potency when stored in amber bottles at incu- bator temperatures (38.0”) even though sealed in the presence of air. In the case of vitamin Dz, the tests cover samples aged 53 months; in the case of vitamin Ds, they cover samples aged 36 months. Since these conditions are as drastic as would ever conceivably be imposed on such pharmaceutical products, it may be stated that for all practical purposes the solutions are stable. The solutions of these vitamins in corn oil also show no loss in potency during 30 months in the incubator.

On the other hand, ertron, which was satisfactory as to potency relative to claims when fresh, frequently shows some deterioration after only 4 weeks in the refrigerator. During 3 months in the incubator it loses more than one-third of its potency.

Solutions of vitamin Dz in propylene glycol do not lose their potency when dissolved either in bottled whole milk kept up to 8 days in the re- frigerator, or in canned condensed milk stored for 6 months at 40” or for periods up to 15 months at room temperature.

It is alsoapparent from these data that both vitamins Dz and Da in the crystalline form are more potent than the 40,000 I.U. or U.S.P. units per rng. as generally recognized and reviewed by Remp and Marshall (18). In general, the reference oil uniformly produces a lesser degree of recalci- fication than does the corresponding dose of these vitamins calculated on the 40,000 units per mg. basis in the rat and in the case of vitamin D3 in both rats and chickens. As calculated from more than 400 separate satisfactory assays, it appears that these crystalline vitamins are undervalued in the literature by at least 10 and more probably by 15 per cent.

Every comparative test of the two vitamins shows that they behave exactly alike as to potency in the rat, and that whatever potency figure is set for the one must be accepted for both. However, preparations of vitamin Da are more invariable in their biological characteristics. This vitamin should serve admirably as a standard reference preparation, either as a crystalline product of uniform chemical and physical characteristics, or more convenient’ly as an oil or propylene glycol solution. Such a prep- aration would serve equally well for rat and chicken assays.

Chemical tests of activated ergosterol or 7-dehydrocholesterol indicated that the characteristics of these products when impure, i.e. mixed with antirachitically inactive products, were materially less stable than the crystalline products wit,h a constant melting point and optical rotation.

SUMMARY

Physicochemical studies of the various nitrobenzoic esters of vitamins Dz and D3 indicate that no decomposition is demonstrable after storage at

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

W. HUBER AND 0. W. BARLOW 135

room temperatures for at least 5 years. The stability of crystalline vitamin Dz or Dz in the presence of air, nitrogen, or COz is materially less than when st,ored in an evacuated ampule. Vitamin Dz, when stored in evacuated amber ampules under refrigerator temperatures, undergoes no change in its physicochemical characteristics over a storage period up to 9 months. Vitamin D3 in general is somewhat more stable than vitamin Dz and no definite change is demonstrable after storage under optimal conditions for a period of 12 months.

Solutions of vitamins DS or Da in propylene glycol or vegetable oils retain their potency over long periods of time at a temperature of 40”. When such solutions are dissolved in milk, the vitamin D potency of the fortified milk remains unchanged over periods of time corresponding to the useful- ness of the menstruum as a food.

The data presented indicate that the usually accepted antirachitic value of 40,000 U.S.P. units per mg. for calciferol or crystalline vitamin Dz and crystalline vitamin DS is 10 to 15 per cent low.

BIBLIOGRAPHY

1. Fritz, J. C., Halpin, J. L., Hooper, J. I-I., and Kramke, E. H., Ind. and Eng. Chem., 34, 979 (1942).

2. Baird, F. D., Ringrose, A. T., and MacMillan, M. J., Poultry SC., l&35 (1939). 3. Jung, A., Cong. internat. chim. tech. ind. agric., Compt. rend. VI tong., Budapest,

2, 449 (1939). 4. Windaus, A., Deppe, M., and Wunderlich, W., Ann. Chem., 633, 118 (1937). 5. Askew, F. A., Bourdillon, R. B., Bruce, H. M., Callow, R. K., Philpot, St. L. F.,

and Webster, T. A., Proc. Roy. Sot. London, Series B, 109, 488 (1932). 6. Windaus, A., von Werder, F., and Gschaider, B., Ber. them. Ges., 66 B, 1006 (1932). 7. Anderson, F. W., Bacharach, A. L., and Smith, E. L., Analyst, 62, 430 (1937). 8. Schenk, F., Naturwissenschaften, 26, 159 (1937). 9. U. S. patent 2,264,320 (1941).

10. Lewis, J. M., J. Pediat., 6, 362 (1935); 9, 308 (1936). 11. Eliot., M. M., Nelson, E. M., Barnes, D. J., Browne, F. A., and Jenss, R. M.,

J. Pediat., 9, 355 (1936). 12. Shelling, D. H., J. Pediat., 10,748 (1937). 13. Bourdillon, B. T., Bruce, H. M., and Webster, T. A., Biochem. J., 26, 522 (1932). 14. Bruce, H. M., Kassner, E. W., and Phillips, G. E., Quart. J. Pharm. and Phar-

macol., 11, 46 (1938). 15. Shelling, D. H., Proc. Sot. Exp. Biol. and Med., 36, 660 (1937). 16. Supplee, G. C., Ansbacher, S., Bender, R. C., and Flanigan, G. E., J. Biol. Chem.,

114, 95 (1936). 17. Fuchs, L., and van Niekirk, J., Biochem. Z., 277, 32 (1935). 18. Remp, D. G., and Marshall, I. H., J. Nutrition, 16,527 (1938).

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

W. Huber and O. W. BarlowDERIVATIVES

AND THEIR3 AND D2VITAMINS DSTABILITY OF CRYSTALLINE CHEMICAL AND BIOLOGICAL

1943, 149:125-137.J. Biol. Chem. 

  http://www.jbc.org/content/149/1/125.citation

Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

alerts to choose from all of JBC's e-mailClick here

  tml#ref-list-1

http://www.jbc.org/content/149/1/125.citation.full.haccessed free atThis article cites 0 references, 0 of which can be by guest on M

ay 21, 2018http://w

ww

.jbc.org/D

ownloaded from