PAST ert anes Bas Rates AS he POS tea
Sa
my
S, vee 2A
etree! x
A oy - % peal a i ste
ashing acdpabes Tee, aged ale
eter Cris ees
5 ee eeten vei gatigehyerrestmpemanubbbertels + 7 1 six
o Sigh tacte state Ageia age ie meee
oda tg Peay
moe aad a a
y Librar
€ Ny
—
Set Re eects
i
dG ee
.
an
CAMBRIDGE UNIVERSITY PRESS London: FETTER LANE, E.C. OC. F. CLAY, Manacer
Evinburgh: 100, PRINCES STREET Zonvon: H. K, LEWIS, 136, GOWER STREET, W.C. WILLIAM WESLEY & SON, 28, ESSEX STREET, STRAND
Yaris: LIBRAIRIE HACHETTE et CIE
Chicago: THE UNIVERSITY OF CHICAGO PRESS
Bombay and Calcutta: MACMILLAN AND CO., Lop. Toronto: J. M. DENT AND SONS, Lp.
Tokyo: THE MARUZEN-KABUSHIKI-KAISHA
All rights reserved
THE
BIOCHEMICAL JOURNAL
EDITED FOR THE BIOCHEMICAL SOCIETY
BY
W. M. BAYLISS, F.R.S. AND
ARTHUR HARDEN, F.R.S.
EDITORIAL COMMITTEE
Dr E. F. ARMSTRONG Dr F. G. HOPKINS Pror. V. H. BLACKMAN Pror. F. KEEBLE Pror. A. J. BROWN Pror. B. MOORE Mr J. A. GARDNER Dr W. RAMSDEN
Dr E. J. RUSSELL
Volume VIII, 1914
Cambridge :
\
at the University Press PA /
1914 |
CONTENTS | eee No. 1 (February)
I. Milk—Its Milk Sugar, Conductivity and Depression of Freezing Point. By Lizias CuarLorre JAcKson and ArTHuR CectiL Hamet RoTHERA 1
II. Osmotic Phenomena of Yolk of Egg. my Witt1amM ALEXANDER
OsporRNE and Hinpa Estetite Kincarp : . . ee 1 III. Caseinogen and Casein. By Arruur ee : or ee IV. A Contribution to the Sigs of a sae ea: Organism. BY Joun Matcotm DrumMMoND ; 38 V. oonglramaneet a New Active Princip of got By ARTHUR JAMES Ewins 44 VI. On the Pkavion of Trypsin in the Paci of a Spe Precis pitate. By Acnes Exten Porter. . ae = —sVII.:- The Viscosity of Some Protein Solutions. By auiness Chics and Eva Lusrzynska. (With two Figures) p ; ; 59
VIII. The Quantitative Estimation of Urea, and Indirectly of f Allantoin, in Urine by means of Urease. BY Rosert Henry Apers PLIMMER and
Rours Finpy SKeiron ; : 70 IX. The Cholesterol of the Brain. Il. The ee of * Oxycho lesterol” and its Esters. By Mary Curistine RosENHEIM F 74
-X. The Cholesterol of the Brain. III. Note on the Cholesterol Contents of Human and Animal Brain. By Mary Curistins RosENHEmM 82
XI. On the Resistance of al toa Solutions to Heat. By EDWARD
AFFORD Epir_. : . 84 XII. On the Aoiiea of Céaduintine Envmeas on Caseinogen. By ArtHuR HarpEN and ArcHriBALD Bruce MacaLtLum i 90
XIII. The Enzymes of Washed Zymin and Dried Yeast (Lebedefi) Ii. Reductase. By ArrHur Harpren and Roxtanp Vicror Norris . 100
No. 2 (April)
XIV. The Formation of a Peptone from Caseinogen by the Prolonged Action of Dilute Hydrochloric Acid in the Cold. By Casimir Funk and JAMES WatTeR McLEop , 107
XV. The Galactosides of the eal Sa The SEAT of Pliseacad and Kerasin by the Pyridine Method. By Orro RosenHEmm. (With Plate I) 110
XVI. The Galactosides of the Brain. III. Liquid Crystals and the Melting Point of Phrenosin. By Orro RoseNnHEm™M. (With Plate II) . 121
XVII. Remarks on Dr Symons’ “Note on a Modification of Teichmann’s Test for Blood.” By Witit1Am DuNBAR SUTHERLAND and GopaL CHANDRA Mitra . ; : i ; ; E : . 128
vi CONTENTS
XVII. An Improved Hydrogen Electrode. y GEORGE STANLEY Warote, (With 1 Text-figure) . ;
XIX. The Estimation of Lactose and Grek by the Soprigé Todide Method. By Sypney Wituiam Coxe. (With 4 Text-figures)
XX. A Volumetric Method for the Estimation of Ethereal and inouen Sulphates in Urine. By Orro RosenneEmm and Jack Ceci, DRuMMOND
XXI. A Note on the Production of Casein from speak? BY SamuEL Barnett ScHRYVER
XXII. Herzig and Meyer’s Reaction Applied to poacs fad poss Acids. By Josnua Harotp Burn . : : ;
XXIII. Studies in Protein Hydrolysis. By ee ‘Wynn Pratt Prrrom. (With 3 Text-figures)
XXIV. Diagrammatic Co-ordination of Photanel Relating to koa gation of Sols. By Grorcr Stantey WaLpote. (With 4 Text-figures)
XXV. Action of Pepsin and Trypsin on one another. gaa Paper.) By Epwarp Srarrorp EDIE
XXVI. The Flower Pigments of dntiorhintoh majus. TIL. The Red and Magenta visemes By Murret WHELDALE and Harotp LLEWELYN Bassett Sgt erg ea a ; ; } ‘ P
XXVII. The Constitution of Pseudo-Muscarine (“Synthetic Muscarine”). By Artnur James Ewins ; : ; : ; ; ; 2
No. 3 (June)
XXVIII. The Enzymes of Washed Zymin and Dried Yeast (Lebedeff). III. Peroxydase, Catalase, Invertase and Maltase. = ARTHUR HARDEN and Syivester SoLomon ZILvA
XXIX. An Investigation into the vive Clbcniieat: Moctlanial: of Haemolysis by Specific Haemolysins. No. II]. The Electrical Conductivity of Sensitised Corpuscles and the Action of Inorganic Ferments or Metal-Sols upon them. By Uprenpra Natu BRAHMACHARI .
XXX. On the Estimation of B-Hydroxybutyric Acid, By ERNEST Laurence Kennaway. (With 4 Text-figures) :
XXXI. Counter Diffusion in Aqueous Solution. By Wisma ALEX- ANDER OsBoRNE and Litias CHARLOTTE JACKSON
XXXII. The Curative Action of Autolysed Yeast against listed Poly neuritis. By Evetyn AsHiry CooPErR
XXXII. A Note on the Bases of Gasworks Coal- tar which are believed to be the Predisposing Cause of Pitch Cancer, with Special Reference to their Action on Lymphocytes, together with a Method for their Inactivation. Part I: Auxetic Action. By Dorotuy Norris .
XXXIV. The Viscosity of Protein Solutions. IT. Peeudoglobulin at Kuglobulin (Horse). By Harrierre Caicx. (With 1 Text-figure)
PAGE
131 134 143 152
154
.. aoe
170
193
=
=
217
261
CONTENTS vii
No. 4 (August) PAGE XXXV. Colouring Matter contained in the Seed-coats of Abrus ies catorius. _ By Sarast Lan Sarkar / : , . 281
XXXVI. The Action of Diazomethane on Caseinogen (Preliminary Communication). By ArTHur GEake and Maxiitian NIERENSTEIN . 287
XXXVII. The Mechanism of pe cstv of Bacteria by sates Sera. By Witui1am Joun TuLLoca ; . 293
XXXVIII. The Mode of Oxidation of Certain ed Acids with Branched Chains. By Henry SranLey Raper... 320
XXXIX. The Chemical Nature of a Bacterial paca ye By. Eric Epwin Arkin. (With 2 Text-figures) : : 328
XL. The Relations of Vitamine to —— By Evretyn AsHLEY CooPER : ; ‘ een 5. 2 ated
XLI. The Relative Amounts of B-Hydroxybutric Acid and Aceto- acetic Acid excreted in Acetonuria. By Ernest LaurENcE KENNAWAY.
(With 3 Text-figures) . ; ‘ : : 2 , ; f ; 1 BOO XLII. Some New Beene ae Active Derivatives of Choline. By ArTHur JAMES Ewins : . 366
XLII. On the Respiratory Exchange in Fresh Water Fish. Part I. On Brown Trout. By Jonn AppyMAN GARDNER and ConsTANCE LEETHAM.
(With 2 Text-figures) . ; ; ; ‘ A ; ; P . . 374
XLIV. A Method for the Estimation of Sugar in Blood with Observa- tions on some Modern Methods. By ArrHur Duncan GARDNER and Hue
MacLean 3 ; k : : : ; ; Z 381
XLV. The Apparent Formation of Euglobulin from Pseudo-Globulin
and a Suggestion as to the Relationship between these two Proteins in “\ Serum, By Harriette CHICK . ‘ : : : F : ; . 404
XLVI. The Hydrolysis of Glycogen by Diastatic Enzymes. Part III. Factors influencing the End-Point of the 0 By Rotanp Victor Norris. (With 2 Text-figures) . ; : 3 : . 421
XLVII. On the Purine Metabolism of Rats. By Harotp Ackroyp . 484
XLVIII. The Production of w-Hydroxy-s-methylfurfuraldehyde from Carbohydrates and its Influence on the Estimation of Pentosans and a Pentosans. By Mary Cunninenam and CuarLes Dorke . : . 438
viii CONTENTS No. 5 (October)
. ~ PAGE XLIX. The Urease Content of certain Indian sg By Haroup Epwarp ANNETT A POS : . - 449 L. On the so-called Paacipedable Phosphaiides” wd Huex Mac ; LEAN. ; 453 LI. A Note on ‘the Black ree in the Skin of an Adtabontion Black. By Wriu1am Joun Youne . 5 460 LIT. Quantitative Estimation of Aépaitio and Glutaminio Acids j in the Products of Protein Hydrolysis. By Freprerick WiLL1AM ForREMAN . . 463 LIII. The Transformation of Glutaminic Acid into [-Pyrrolidonecar- boxylic Acid in Aqueous Solution. By Freprerick WiLLiaM Foreman. 481 LIV. The Activation of Teypenoee By Horace MippLeton VERNON. (With 17 Text-figures) . : 4 wane he ; . 494 LV. The Influence of Sones Water Ingestion on Protein Metabolism. By Jonn Boyp OrR . ’ : 530
— LVI. The Determination of the Gangsonition of she Different Protald of Ox and Horse Serum by the Method of Van Slyke. By Perctvat Hartiey 541
LVII. The Gelatinisation of Pectin in Solutions of the Alkalies and the . Alkaline Earths. By Dororay Haynes. (With 4 Text-figures) : . 553
No. 6 (December)
LVII. The Action of Poisons on Reductase and Attempts to isolate this Enzyme. By Davip Fraser Harris and Henry JERMAIN MAuDE CREIGHTON .. 585
LIX. On the en ores aa in Fresh-' Water Fish, Part I: on Brown Trout. By Jonn Appyman GarpNner and Constance LEETHAM.
(With 1° Text-figure) : : 591 LX. The Chemical Investigation of the Phidpliotangstate Phocintets from Rice-Polishings. By Jack Cecrt DrumMonp and Casimir Funk . 598
LXI. A Comparison between the Molecular Weights of Protagon and of the Phosphatide and Cerebrosides obtainable from it. By ALEXANDER
LyaLL PEARSON... 616 LXII. Notes on Rectator Misco: Rae indenting etc. iL By . GrorGE STANLEY WaLPote. (With 1 Text- -figure) : ; 628
LXIII. The Estimation of Allantoin in Urine in the Proseiiie of Glucose. By Rosperr Henry Aprers Piimmer and Rura Fitpy Sxexron 641
LXIV. The Gravimetric Estimation of Minute es of ho By Henry Srantey Raper : 649
LXV. The Rate of Inactivation = Heat of Penossidas in Milk. L By Sytvester Sotomon Zitva. (With 4 Text- -figures) . : . 656°
LXVI. Liver Eee in ena By GEorRGE oats and Henry Hatterr Dae ; ; ; : ‘ . 670
®
I. MILK—ITS MILK SUGAR, CONDUCTIVITY AND DEPRESSION OF FREEZING POINT.
By LILIAS CHARLOTTE JACKSON and ARTHUR CECIL HAMEL ROTHERA.
From the Biochemical Laboratory of the University of Melbourne. (Received Nov. 18th, 1913.)
In this paper the results of the application of measurements of the electrical conductivity, percentage of milk sugar and depression of freezing point will be discussed chiefly from the standpoint of the physiology of milk.
In fact the measurements were obtained rather as the basis from which to draw certain deductions and interpretations, than as data for the identification and characterisation of different milks.
This latter work has already been accomplished and it is only necessary to give a brief survey of it in so far as it bears upon the subject matter of
this paper.
Depression of freezing point. .
The average depression of freezing point of cows’ milk as found by 10 independent workers [Koeppe, Sommerfeld, 1909, p. 147] lies between 0°552° and 0°572°. In a recent paper Pins [1912] gives results of a large number of determinations of the depression of freezing point of the milk of 40 different cows, in which he finds that the largest proportion of the values obtained lie between 0°556° and 0°560°.
In the milk of sick animals higher values are often obtained [Schnorf, 1905, Pins, 1912].
The constancy of the depression of freezing point in normal milk is insisted upon by Guiraud and Lasserre [1904]. Parmentier [1903] points to the constancy of A even in milk from inflamed quarters of the udder. Schnorf [1905] draws attention to the constancy of the conditions determining the osmotic pressure of the milks from different quarters of the udder, and notes that the two fore-quarters and two hind-quarters go together.
Bioch, vir 1
2 L. G. JACKSON AND A. C. H. ROTHERA
H. Dreser [1892], in discussing the osmotic condition of different body fluids, lays stress on the closeness of the depression of freezing point of milk (A=0'57) to that of the blood (A=0'58 to 0°59). This osmotic equili- brium between the blood and milk is again emphasised by J. Winter [1895].
Rennet coagulation of milk has no effect on the depression of freezing point according to Hotz [1902] but boiling the milk slightly diminishes it.
The depression of freezing point of human milk shows greater individual variations than that of cows’ milk, and Koeppe believes this to be due to variations in the salt content of the food. Strauss [1900] however denies that alterations in the food affect the depression of freezing point of the milk.
- The following values have been given for human milk : Koeppe A 0:495 to 0°630 Barthe [1904] A 0°590 to 0°610
Villejean [1905] A 0°535 to 0°615 Grassi [1906] A 0-540 to 0°740
Electrical conductivity.
In contrast with the relative constancy of the depression of freezing point of the milk of mammals, the specific conductivity shows greater variations, although under normal conditions the value for the conductivity in any one species does not vary to any large extent.
The following averages are given by different authors :
Cows’ milk. Koeppe [Sommerfeld, 1909, p. 149] K at 25° C.=0°0043 —0-0056
Lehnert [1897] and Koeppe [1898] fF », ~=0°00487 —0°00551 Binaghi [1910] =i ee = », =0°00494 -0:00517 Schnorf [1905] ws ee re 5, =0°00485
Goats’ milk. Binaghi [1910] is », =0°00470-—0-00499
Human milk. K at 18° C.=0°00149 — 0-00843
Friedrich Petersen [1904], examining cows’ milk, notes that the first drawn portions have a lower electrical resistance (higher conductivity) than the last drawn portions.
He notes that the resistance diminishes (i.e. conductivity increases) at the end of lactation. Also, in colostrum the resistance is less. He finds no proportionality between specific gravity and electrical resistance or total solids and resistance, but mentions a general proportionality to the ash though there is no direct relationship.
H. Hotz [1902] notes that milk whey has a higher conductivity than milk itself,.due to removal of the caseinogen. C. Schnorf in his book
L. C. JACKSON AND A. C. H. ROTHERA 3
mentions a rise of 10-17°/, in the conductivity when the caseinogen is removed by-rennet. The caseinogen mechanically obstructs the carriage of electricity by the moving ions.
Differences in the conductivity of milk from different quarters are assigned to different amounts of non-electrolytes.
In small milk yields the conductivity is higher, and he notes the high conductivity of colostrum except in the very early stages, where the large amount of protein depresses the conductivity.
He also mentions the uniformly high values obtained for the conductivity in udder inflammations. A paper by Bugarsky and Tangl [1898] will be referred to later on in connection with the effect of protein upon conductivity. They find 1 gram of protein per litre to depress the conductivity 0°25°/,. The experiments _ were conducted :
(a) By dialysing blood serum for two months till free from salts, and using this serum concentrated as a means of adding different amounts of protein to 0°8°/, salt solution.
(b) By dialysing 15 to 25c.c. blood serum against 150 to 250c.c. water until equilibrium was established and then noting the higher con- ductivity of the fluid outside the dialysing tube as compared with that inside the tube, and estimating the percentage of protein in the latter fluid.
Milk sugar.
The sugar from the milk of the cow, sheep, goat, mare, ass, dog and woman is lactose [Déniges, Bonmartini, Sommerfeld, 1909, p. 193]. The existence of a different sugar has from time to time been assumed owing to discrepancies between the reduction and polarimetric methods of estimation.
Scheibe [1901] has however shown that making due allowance for errors in both methods the results obtained do not differ. Consequently the con- clusions of Schméger, V. Raumer and Spath, and Landolph [Sommerfeld, 1909, p. 194] that another sugar is present in addition to lactose are unfounded. That the milk sugar is diminished in amount in milk from inflamed quarters of the udder is well known [J. Bongert, Sommerfeld, 1909, p. 555).
METHODS EMPLOYED.
Depression of freezing point. The Beckmann apparatus was used with an additional outer jacket and a mechanical stirrer. The determinations were made upon milk samples from which the greater part of the fat had been
1—2
4 L. G. JACKSON AND A. C. H. ROTHERA
removed by centrifugalisation. The experimental work was carried out as advised in a paper by Dekhuyzen. |
Electrical conductivity. Kohlrausch’s method was employed. It is very simple and can be very quickly carried out. For these reasons it is specially suited to the examination of milk.
The Wheatstone’s bridge consisted of a wire wound on a cylindrical drum, with a sliding contact, operated by revolving the drum. The electrodes of the conductivity cell were replatinised at intervals and the cell standardised from time to time with a specially prepared solution of potassium chloride. All determinations were made at exactly 25°C. For all the measurements to be quoted in the present paper the cream was separated from the milk before the determinations were made. |
Milk Sugar. The method selected was to determine the optical rotation of the filtrate after precipitating the milk proteins with Wiley’s acid mercuric nitrate solution. This solution is made by dissolving mercury in twice its weight of nitric acid of specific gravity 142 and adding an equal volume of water after solution is completed.
In the case of cows’ milk 3 cc. of the reagent are necessary per 100 c.c. of milk, but in the milk of smaller animals with 7°/, protein or more 6 or 7 cc. are necessary. It had to be assumed that the volume of precipitate formed was equivalent to the volume of Wiley’s reagent added. Undoubtedly this assumption was not always correct and therefore slight experimental errors are present in the estimations, and milks from different species of animals do not show strictly comparable sugar determinations. This criticism does not, however, apply to the milks of the same animal, and here the sugar estimations are relatively to one another highly accurate. 3 |
Milk samples always had the fat removed before being used, and in the case of small amounts of milk very rich in protein (cat and dog) the samples were diluted with an equal volume of water before being precipitated by the
Wiley’s reagent. A Schmidt and Haensch triple field polarimeter was the instrument employed for taking the rotations.
THE INVERSE PROPORTIONALITY OF MILK SUGAR AND SALTS,
The substanges in milk which are chiefly responsible for its osmotic pressure are the milk sugar and soluble salts. The osmotic pressure of milk is, however, dictated by the osmotic condition of the blood, or in other words the conditions governing the osmotic pressure of the milk (as measured by the depression of freezing point) are to be sought in a fluid of fairly uniform
L. C. JACKSON AND A. C. H. ROTHERA 5
composition, whose osmotic condition is kept constant by the action of the kidneys.
This explains the constancy of the depression of freezing point in milk, and it follows that the substances in milk chiefly responsible for this physical manifestation cannot vary independently, but must be inter-related.
If the milk sugar is high the salts must be low, otherwise the osmotic pressure would be unusually high, and conversely, if the milk sugar is low the salts must be high or the osmotic pressure would be lower than normal.
Tables giving collected analyses of the milk of different species of animals, such as the following table quoted from Droop Richmond’s Dairy Chemistry, do, as a rough generalisation, show an inverse proportionality between milk sugar and salts,
Animal Sugar Ash Cow 4°75 0°75 Goat 4:22 0°76 Ewe 4:28 0:97 Buffalo 4°7 0-90 Woman 6°8 0°20 Mare 6°89 0°30 Ass 6°50 0°46 Mule 4°80 0°38 Bitch 3°09 0°73 Cat 4°91 0°58 Rabbit 1°95 2°56 Llama 5°60 0-80
That the relationship is not exhibited in a closer degree than is shown by these tables is due to the fact that the analyses of ash give no information as to the relative amounts of soluble and insoluble salts. In cows’ milk, for example, much of the calcium phosphate (probably two-thirds of it) is in a state of colloidal suspension. This practically contributes nothing to the osmotic pressure and so should be deducted from the ash analyses to derive an estimate of the soluble salts. It is not, then, the total ash or salt content of the milk which is to vary inversely to the milk sugar, but the soluble ash, which alone appreciably contributes to the osmotic pressure.
The insoluble salts may be regarded as inert and it may be pointed out that their presence is probably a device of nature to get beyond the limita- tions imposed by the necessity of adjusting the osmotic pressure of the milk upon that of the blood. Colloidal calcium phosphate permits of a high calcium content of cows’ milk without unduly reducing other soluble con- stituents such as the sugar and salts of sodium and potassium.
Undoubtedly also the limitation of an adjusted osmotic pressure has led to the fixing of a disaccharide as the sugar of the milk, which weight for
6 L. ©. JACKSON AND A. OC. H. ROTHERA
weight has only half the osmotic effect of the monosaccharide, dextrose, present in the blood.
The sugar and soluble ash then should vary inversely in any two milks whose salts are the same in character and whose osmotic pressures are equal.
Now in the case of soluble salts their osmotic effects are dependent upon their ionisation, and as the electrical conductivity of a salt solution is dependent upon the same factor, we have in the conductivity of milk a very close indication of the osmotic value of the soluble milk salts.
Putting this reasoning to the test of experiment it ought to be shown that conductivity and milk sugar vary inversely.
There are of course disturbing factors, some of which will be discussed later, but whereas it is only a very rough generalisation that ash and sugar are reciprocating, it can be shown that the inverse relationship between sugar and electrical conductivity is practically a law.
It is seen best of all in milk from one animal taken at the same milking, but from different teats. In the case of a cow it is possible to obtain four different milks from the four distinct quarters of the udder. If obtained at the one milking they will have within the limits of experimental error the same depressions of freezing point.
Here are the figures for the milk sugar and conductivity as estimated in the milk after separation of the fat.
TABLE I. Connection between the conductivity of milk and the percentage of lactose.
Milk from the four different quarters of the cow’s udder.
Percentage Quarter K of lactose Series I Left anterior 0:00596 5:41 Right ,, 0-00613 - 5:30 Left posterior 0-00630 5°20 Right _,, 0:00641 5:13 Series II Left anterior 0-00500 5°70 Right = 0:00502 5°68 Left posterior 0:°00506 5°61 Right ,, 000512 558 Series III Left anterior ge } 0-00527 5-74 Left posterior bia } 0-00531 5°68 Series IV Left anterior 0°00544 5°73 Right _,, 0:00545 5°64 Left posterior 0°00557 5°55
Right ,, 0-00555 5°59
L. C. JACKSON AND A. C. H. ROTHERA 7
It is seen that milk sugar and conductivity vary inversely, that in every series the milk from the anterior quarters has more milk sugar than the milk from the posterior quarters but a lower conductivity.
The same relationship is also shown in the case of human milk.
TABLE II. Human milk. Individual K Percentage lactose Remarks A 0:00259 6°85 Left breast. 0-00263 6°24 Right ,, B 0:00290 5°18 Left —_,, 0-00262 5-41 Right ,,
Note. These samples of milk were obtained from women who for forty-eight hours previously had practised nursing their baby from both breasts. Two breast pumps were used and applied simultaneously one on each breast. We found that the practice at the Women’s Hospital, Melbourne (from which institution most of our milk samples were obtained), was to nurse from only one breast at a time, saving the other breast for the next meal. This quite prevented samples taken simultaneously from right and left breasts showing the reciprocity of the above table, for the milk in each breast is not then secreted under comparable conditions of the blood.
In subsequent tables (see pp. 11, 12, 13) showing the composition of the milk of various species of animal, the sow, ass, mare, llama and goat, the inverse proportionality between conductivity and sugar is generally in evidence.
It has been our universal experience (which is in agreement with that of Guiraud and Lasserre, Parmentier and others) that the depression of freezing point of abnormal milk is not different from that of normal milk. One mammary gland may be giving a pathological fluid without the slightest resemblance to true milk yet with an identical depression of freezing point to the normal milk from the other gland (goat) or other quarters of the udder (cow).
In comparing the milk from an abnormal and normal quarter in the same animal the inverse proportionality of conductivity and sugar should be very apparent.
The method employed to disturb the secretion of one of the quarters was to return normal milk, guarding carefully against outside bacterial infection, and taking precautions to maintain the re-injected milk at body temperature.
Table III gives the results of two such experiments, and the figures showing the gradual recovery of the experimental quarter bear out the inverse relationship of milk sugar content and conductivity value very forcibly.
8 L. CG. JACKSON AND A. C. H. ROTHERA
From the records below it will be seen that the pathological milk shows at first a high conductivity and a very low percentage of milk sugar, but as recovery of the quarter takes place, the conductivity gradually becomes lower, and the sugar percentage rises again towards its normal figure.
In Experiment II the ash in a sample of milk from the affected left anterior quarter was compared with the ash from the corresponding normal quarter—the right anterior quarter.
It will be seen that though the total ash is only slightly higher in the pathological milk, the difference between the proportions of soluble and insoluble ash is most marked as compared with the distribution in the ash of normal milk.
TABLE III. Experiment I. 23/2/12.
On Friday morning February 23, 1912, 300 ¢.c. of milk were milked from the left anterior quarter of the cow’s udder.
A silver teat cannula was then inserted and another 400 c.c. drawn off into a bulb condenser as receiving vessel. Water at 40° was flowing through the outer jacket of the condenser. The cannula was connected to the condenser by spiral flexible metal tubing wrapped in cotton wool. The whole apparatus had been carefully sterilised prior to use in an autoclave.
By raising the condenser approximately 350 c.c. of the 400 c.c. withdrawn were made to flow gently back into the same quarter of the udder, the cannula never having been withdrawn.
This operation is a sure method of obtaining a temporary upset in the quarter of the udder so handled, the other quarters remaining unaffected.
Percentage Quarter K of lactose 23/2/12 (morning) Left anterior 0°00557 5°52 Right posterior 0-00566 5°52 23/2/12 (afternoon) Left anterior 0°01259 0°65 Right anterior 0°00581 5°65 Left posterior 0:00589 5°57 Right posterior 000592 5°62 24/2/12 (afternoon) Left anterior 0°00718 25/2/12 (morning) Left anterior 0°00707 4°13 Right anterior 000567 5°37 26/2/12 (morning) - Left anterior 0°00635 4°40 Mixed sample from : the 3 other poareeat O00bs1 isi 27/2/12 (morning) Left anterior 0°00612 5°41 Right anterior 0°00525 5°90 Left posterior 0°00542 5°85 Right posterior 0°00539 5°87 28/2/12 (morning) Left anterior 0°00579 5-11 Mixed sample from : the 3 other arent 0100589 Rae
Note. The experimental quarter is indicated by heavy type.
L. C. JACKSON AND A. C. H. ROTHERA 9
Experiment IT. 5/3/12.
On Tuesday morning-March 5th, 1912, about 200 c.c. of milk were milked from the left anterior quarter and then 300 c.c. were withdrawn through a sterilised teat cannula into a bulb condenser as in Experiment I.
Of these 300 c.c. approximately 250 c.c. were returned. The whole proceeding was neatly and cleanly carried out. The disturbance which followed was profound.
"lo %lo °/, Insol. Quarter K Lactose A Sol. ash ash 5/3/12 (evening) Left anterior 00114 1°50 0580 0-615 0°44
Right anterior 0-00569 540 0°575 0-285 0-625
6/3/12 (morning) Not enough milk obtainable from left anterior quarter to take either conductivity or rotation.
7/3/12 (morning) Left anterior 0°00734 2°86 - _— — Right anterior 0°00535 5°72 — — 8/3/12 (morning) Left anterior 0°00699 418 0°545 0°352 0°585
Right anterior 0-00535 5°36 0-543 0°250 0°617 Left posterior 000536 5°32 0-540 _ —
Right posterior 0°00536 — 0-545 ae whe.
9/3/12 (morning) Left anterior 0°00616 4°81 poe wie ak Right anterior 000520 5-44 ~Eee ae Be
11/3/12 (morning) Left anterior 0°00551 5°28 — ae —
Right anterior 0°00521 5°48 — — —
In the first set of analyses (5/3/12 evening), when the conductivity of the milk from the left anterior quarter was very high, the amount of soluble ash was very much greater than normal and the amount of insoluble ash considerably less than normal.
The same differences between the pathological and normal milks are still apparent on the morning of 8/3/12 but in a less degree, indicating the return of the left anterior quarter of the udder to a normal condition. It is interesting to note that in a pathological milk there may be the paradox of a diminished percentage of sugar and ash with an unaltered depression of freezing point. In the ordinary determinations of ash however, no distinction is made between the soluble and insoluble portions and moreover the heat employed in getting rid of the carbon causes the volatilisation of some of the chlorides.
The inverse proportionality between milk sugar and electrical con- ductivity may be shown by comparing the milks of different animals, and the manner in which the results obtained conform to theoretical considera- tions is, in view of the many disturbing factors, exceedingly satisfactory- The experimental difficulties are to be placed first, for many of the animals being unused to handling about the udders, had to be educated to give milk. Generally it meant establishing a habit, whereby the young were separated from the parent, and only admitted at stated times for their feed. When milk samples were required the young were allowed to start the flow of
10 L. ©. JACKSON AND A. C. H. ROTHERA
milk, and then removed. In many cases this meant a wet teat, and with the sow the young pigs leave the teats covered with a very slimy mucous.
With the smaller animals it was particularly difficult to get samples large enough for the determinations of freezing point and milk sugar content, and the cat, bitch and kangaroo had to be kept apart from their young for longer periods than could be called natural.
As already mentioned actual methods of analysis were difficult to keep free from small errors in the determination of milk sugar, chiefly on account of the varying protein contents of the milk.
Also the large amounts of protein in the milks of the smaller animals would undoubtedly depress the conductivity values, as will be shown later in discussing the influence of colloids on the conductivity of milk.
But apart from experimental difficulties and errors of analysis, there is the added complication that the milks of different animals and even of the same animal have a varying depression of freezing point. Strictly to compare two different milks with a view to showing that the milk with the higher milk sugar percentage will have the lower conductivity, it is necessary that both should have the same osmotic pressure, i.e. the same depression of freezing point.
Again many milk samples we obtained were abnormal, for the most part because the milk was still in the colostrum period or because the lactation was practically at an end.
The analyses of these samples have been discarded and the averages of the other analyses we have made taken for each individual species.
In some cases, e.g. sow, llama, kangaroo, ass, only one animal was available, but we have only used the results of our own analyses because we know that the conductivity, milk sugar and depression of freezing point determinations were all carried through on the one sample.
Collecting together the averages representing milk sugar and conductivity the following table is obtained :—
TABLE IV.
~
a Milk sugar °/o i Kin ascending ~ in descending ‘
Animal values K corrected values Animal
Mare 0-00208 0-00231 7-52 Mare
Ass 0°00247 0:00276 7°37 Ass
Woman 0-00252 0-00281 6°40 Woman
Sow 0:00375 0°00477 6°11 Sow
Goat 000499 0:00595 6°05 Goat
Cow 0-00503 0:00587 5:72 Cow
Cat 0°00537 0:00712 5°24 Cat
Bitch 0°00538 0°00742 5°17 Bitch
L. C. JACKSON AND A. C. H. ROTHERA 11
The only two milks out of order and therefore omitted from this table are those of the llama and kangaroo. Their non-agreement can be accounted for by abnormal depressions of freezing ne :
Thus, for the A=0-600° which is exceptionally high ;
: for the kangaroo A=0-515° which is lower than with other animals.
The corresponding values for these animals are
K K corrected Milk sugar Llama 0°00395 000465 - * 6°95 Kangaroo 0°00494 —- 2-66
The corrections for K have been made by taking the average figures for the percentage of protein in the milks of the various species of animals concerned as published by Droop Richmond in his Dairy Chemistry, p. 323, and adding 2°76 °/, of the conductivity for each 1 °/, protein in the milk.
‘The corrections practically leave the table unaffected. They might justify actually reversing the positions of cow and goat in the left-hand list, just as they justify placing the llama ahead of the sow. With the corrected K, the llama is most definitely brought more into line with the generalisation established by the table and is far from being a glaring exception, such as the kangaroo remains’.
It is apparent from the figures that the reciprocal relation between milk sugar and conductivity holds closely in a survey of the milk of different animals in spite of the disturbing factors previously discussed.
The following tables give our analyses for milks of the different animals we have investigated and from these tables the averages used in Table IV were compiled.
Percentage Depression of
Date No. K lactose freezing point Remarks
Human Milk. 23/3/11 1 0-00327 _ — Colostrum. 19/3/12 2 0-00391 8°48 — oi 25/3/12 3 0-00959 0°57-0°76 — Pathological. 4/4/12 4 0-00321 ret — Colostrum.
Jen 5 0-00214 6°78 — Normal. 17/4/12 6 0-00234 6°23 Sone -
— 27/4/12 7 0-00230 6°81 = .:
re 8 0°00221 6°81 —
- 9 0:00229 7°20 0-560
= 10 0°00223 6°45 0-550 22/7/12 11 0-00216 6°66 = 26/7/12 12 0-00207 6-80 — 14/8/12 13 0-00243 6-50 a 21/10/12 14 0°00274 5°14 oe 25/10/12 15 0-00264 5°79 — 17/2/13 16 000300 5°70 0-545
ag 17 0-00313 5-18 0°530
eine 18 0:00256 5°63 0-550
_ 1 This animal was in the later months of lactation, its young one spending most of its time out of the pouch.
12 L. C. JACKSON AND A. C. H. ROTHERA
Percentage Date No. K lactose freezing point 17/2/13 19 000278 5°04 0°535 24/2/13 20 0:00269 — — te 21 0-00260 _— = sa 22 0:00298 — — Ss 23 0:00305 — = 6/3/13 24 0°00290 6:19 0-530 oP 25 000294 6°50 0°535 “4 26 0:00258 6°66 0-540 ies 27 000263 6°23 0°542 Milk of the Mare.
25/9/12 1 0:00982 2°33 0°589 15/10/12) same 2 0-00395 6°55 0°561 4/12/12\animal 3 0:00214 7°49 0°560 6/1/13 4 0:00203 7°56 —
Milk of the Goat. 15/5/12 1 0°00433 — 0°590 és 2 0°00431 6°28 0-580 16/5/12 3 0:00467 — 0°577 18/5/12 4 0°00459 6-00 0°575 16/7/12 5 0°00536 5°98 —_— 14/5/13 6 0°00567 5°39. 14/5/13 7 000541 6°15 — ne 8 0:00554 6°08 — 15/5/13 9 0:00493 6:19 0°590 16/5/13 10 0:00517 6-13 0°563 19/5/13 11 0°00497 6:21 0°564 Milk of the Ass. All samples obtained from the same animal. 11/1/13 1 0°00249 7°35 0-513 16/1/13 2 0-00246 7°38 0°535 20/2/13 3 0°00794 4-28 0°564 ” 4 0:00888 — ee! a 5 001136 2-09 0-564 e 6 001138 1:25 — Milk of the Sow. All samples obtained from the same animal. 9/1/13 1 0-00469 4:30 — 10/1/13 2 0:00366 6°24 0°573 14/1/13 3 0:00378 6°15 0°565 16/1/13 4 0:00382 5°98 0°580 Milk of the Llama. All samples obtained from the same animal, 4th-5th month of lactation. 21/1/13 1 0-00324 751 0-605 20/2/13 2 0°00445 6°70 0-602 x 3 0-00417 6°96 0°589
Milk of the Cat.
15/3/12
22/3/12
Both samples obtained from same animal.
0°00557
0:00518
5°33
5:14
Depression of
\
— ooo
Remarks
12 months after foaling.
2 days ditto.
2 months ditto.
2 months ditto.
Right side. Left side. Mixed sample.
Obviously patho-
logical.
Littered 1/1/13.
Both samples were small in amount & obtained with difficulty. Fat was separated be- fore any determi- nation was made.
“Oe
L. C. JACKSON AND A. C. H. ROTHERA 13
Percentage Depression of
Date ———“No. = K lactose freezing point Remarks Milk of Bitch. 3/4/12 1 0-00611 6-2 16/7/12 2 000425 not obtainable a se 20/7/12 3 0-00465 4°15 se eres sc ae 27/8/12 4 0-00421 a 0-573 ecptow 4 Milk of Kangaroo. Both samples from one animal. Late stage of lactation. 27/1/13 1 0-00409 2°28 0-520 Very small sample. 8/2/13 2 0-00494 2°66 0515 Good sample.
First AND LAST PORTIONS OF MILK.
It is generally known that differences exist between the first milk drawn off from the mammary gland and the last milk. The latter for instance is very rich in fat, the former poor in fat. One of the many suggestions put forward to account for this is that the first and last milks are secreted under different conditions. This explanation has probably a great deal of
- truth in it, for not only is there the marked variation in fat content, but the early and late portions may show differences in depression of freezing point, milk sugar, and conductivity. The fact that late samples from the mammary gland in cows’ milk generally have a greater depression of freezing point than the early samples was noted by Pins.
With regard to the electrical conductivity, it is sometimes higher in the first portions milked and sometimes in the last portions. In the case of full milk from the cow the conductivity of the last portions is invariably lower owing to the larger fat content of the end milk. But if the milk be separated before measuring the conductivity then definite exceptions to the above rule are found.
The following table gives the results obtained with a single cow and with first and ‘last portions of milk from different quarters of the udder.
TABLE V. Conductivity of first and last portions of milk, or strippings, of cow. Quarter K (first portion) K (last portion) Series I Left anterior 0°00548 0°00554 Left posterior 0°00543 000548 Series II Left anterior 0°00611 0°00582 Right ,, 0-00621 0-00606 Left posterior 000629 0°00631 Right ,, 0-00636 0:00647 Series III Right anterior 000508 0°00514 Left posterior 000514 0°00512 Right ,, 0°00522 0-00520 Series IV _ Left anterior 000496 0°00496 Right ,, 0°00493 0°00505
The samples taken were approximately 100 ¢.c. in volume. The anterior quarters were yielding about 900—1100 c.c. the posterior quarters 1200—1400 c.c. of milk.
14 L. GC. JACKSON AND A. C. H. ROTHERA
With ian milk the conductivity in the later samples was invariably
higher than in the earlier samples.
TABLE VI. Human milk. Individual Early sample Late sample
A 0:00300 0°00313 B 000256 0:00278 C 000298 0-00305 dD. 000269 0-00260 E 0-00290 000294 F 0-00258 000263
The milk sugar also varies between first and last samples as shown in - Table VII.
TABLE VII.
In the cow First portion Last portion Left anterior quarter 5°53 °/, sugar 5°29 °/, sugar Right ,, aa 568 on. 5:04 ,, Left posterior ,, 5355 Diet ys Right ,, sas 522. Cy, 5:04,
These differences in the depression of freezing point, conductivity and milk sugar content of first and last samples of milk taken from a single quarter, or gland, all bear out the assumption that the first and last portions are secreted under different conditions. In fact everything points to:
(1) A steady slight secretion between the intervals of milking or suckling, and
(2) a reflex further secretion produced as the result of ‘the stimuli applied to the teat and gland.
In the cow (1) is relatively important and the animal comes to the milking shed with a great deal of milk already secreted. In other animals (1) is almost absent and only after the efforts of the young is the milk flow established as a result of a nervous reflex. We may mention the cat, bitch, sow and llama in this connection.
In women an intermediate condition is usually met with, there being some milk present in the breast at the time of suckling. The greater portion however is usually secreted under the reflex, initiated by the child. A child may occasionally take the milk already secreted and then have to work for as long as 10 to 15 minutes to establish the reflex flow. In such cases there
is the danger that the mother may remove her child from the breast without the reflex flow of milk having been established.
L. C. JACKSON AND A. C. H. ROTHERA 15
To strengthen the view that milk may be secreted under different conditions, examinations of milk, obtained before and after the putting of the child to the breast, were made with different women. We have to express our very sincerest thanks to Dr A. W. Robertson, Honorary Physician of the Women’s Hospital, Melbourne, for his kindness in procuring us these samples. The samples were taken under the supervision of one of us, and in every case the strictest precautions were observed.
Invariably the putting of the child to the breast changed the character of the milk. Some of the figures for milk sugar and conductivity have already been quoted separately.
The results are collected together in the following table :
Human milk. First and last portions contrasted.
First portion (before child put to breast) Last portion (after child put to breast) e Percentage 2 Percentage ee Individual K lactose A K lactose A 0-00300 5°70 0°545° 0-00313 5°18 0°530° - 0-00256 5°63 0°550 0-00278 5°04 0535 B 0-00269 -- o 0-00260 - _ — Cc 0-00298 -— - 0-00305 —- a= D 0-00290 6-19 0-530 ~ 0°00294 6°50 0-535 E * 0:00258 6°66 0-540 0-00263 6°23 _ 0-542
There is no rule as to the effect produced by putting the child to the breast. The depression of freezing point, milk sugar percentage and conductivity may either be diminished or increased.
It is certainly not merely a question of first and last portions as such, but of two milks secreted at different times. Being secreted at different times they would undoubtedly be influenced by the variations occurring in the blood, due to the meals, drinking and varying activity of the kidneys and glands of the skin.
The first and last portions with the child placed at the breast between the taking of samples is simply the best manner of showing the difference between the milk already secreted and that poured forth as the response to the stimulus of suckling.
In the following experiment the influence of the child is apparent iol only the first samples were taken, April 26th, 1913—evening.
A sample of milk was taken from the left breast and the baby then applied to this breast. There resulted a reflex flow of milk in both breasts and milk came dripping away from the right nipple. A sample was now taken from the right breast.
Right (child being on left breast) K=0-00229 Sugar 7°20 %/)
Evening 1 ret (before child applied) K =0-00230 ys 682%,
16 L. G@ JACKSON AND A. C. H. ROTHERA
The next morning a sample was taken from the right breast, the child then put to this breast and a sample now taken from the left side.
. . othe “46 0 xoning 29% tactic) pomp Seow 6
In this experiment right morning, and left evening, are comparable as the samples were taken before bringing the child to its mother. The bringing of the child and subsequent taking of samples from the left breast in the morning, and the right breast in the evening in both cases led to a milk with slightly lowered conductivity and definitely increased milk sugar from 6:46 °/, to 6°82 °/, in the morning and 6°82 °/, to 7:20°/, in the evening.
In evaluating the figures in this experiment allowance must be made in comparing morning milk with evening milk, for the generally higher con- ductivity and milk sugar content of evening milk, as compared with morning milk. The following analyses of the milk from another woman are typical of these morning and evening differences.
Morning and evening samples of milk.
Series III. Human milk (one individual).
Morning Evening cf Percentage = Z Percentage & K lactose A K lactose A Right breast 0°00295 5°14 0°610° Right breast 0°00333 5°35 0°700° Left breast 0:00289 5°06 — Left breast 0:00323 5°25 —
MORNING AND EVENING MILK.
Cows’ milk. With regard to morning and evening samples of milk it seems impossible to state any definite rule, although a large number of determinations have been made, but as will be seen later, it is probable that external climatic conditions, as well as the food factor referred to previously, affect the conductivity.
In the first series taken, where the milk of a single cow was employed, 7 six morning samples taken at. varying intervals were contrasted with six evening samples taken on the same days. |
In this series four times was the morning conductivity greater than the evening and twice the evening was greater than the morning.
In the next series on 15 different days, evening and morning samples of milk were obtained from the Willsmere Co. These samples were taken from large mixed quantities (150 to 500 qts.) of milk from a special dairy herd (Holstein) and, in this series, of the 30 samples examined, in two instances
L. C. JACKSON AND A. C. H. ROTHERA 17
morning and evening conductivity were identical, ten times the evening conductivity was greater, while three times the morning conductivity was greater though only pronouncedly so in one case.
Conductivity of morning and evening samples of milk.
Date K (morning) K (evening) Series I. 1 cow, 16/10/11 0-00611 0-00584 é 18/10/11 0°00613 0-00641 19/10/11 0-00652 0-00644 24/11/11 0:00633 0-00628 19/ 2/12 0-00516 0-00504 21/ 2/12 0-00501 000527 23/ 2/12 0-00509 0°00531 Series II. Willsmere samples, 5/12/12 0-00554 0-00563 from 150 to 500 qts. 6/12/12 0°00555 0°00554 9/12/12 0°00562 0-00562 10/12/12 0°00556 0°00562 11/12/12 0°00554 0-00555 12/12/12 000549 0°00557 16/12/12 000568 0-00566 18/12/12 0-00573 0-00559 17/ 2/13 0:00587 0-00593 18/ 2/13 0-00562 0-00593 19/ 2/13 0-00568 0:00585 20/ 2/13 0-00562 0-00573 27/ 2/13 000555 0:00599 3/ 3/13 0-00573 0-00582 4/ 3/13 0°00575 0-00575
These irregularities in the electrical conductivities of morning and evening milk are to be sought in the conditions under which the animals live during the hot months of the year in Victoria.
They. feed by night, as well as by day, upon grass which is burnt dry. They receive their chief water supply when brought to the sheds to be milked, and lastly, with a very changeable climate, a given night may be hotter than the following day.
A regular diurnal periodicity in the concentration of the animals’ blood (and consequently in the concentration of their milk) is therefore hardly to be expected.
In contrast to this, women’s milk appears to be regularly more concen- trated in the evening. This may be assigned to the regular daily eating and activity which, following the night’s abstinence and rest, produce a rise in the osmotic concentration of the blood.
Bioch, vit 2
18 L. C. JACKSON AND A. C. HL. ROTHERA
THE DEPRESSION OF THE CONDUCTIVITY BY THE PROTEINS OF THE MILK.
The effect of colloids simultaneously present in solution in water with inorganic salts may be postulated as diminishing the conductivity which the salts alone in water would otherwise show. ;
In the case of milk, freed as far as possible from fat globules, the chief source of obstruction would undoubtedly be the colloidal particles of protein.
The richer a separated milk in’ protein the greater probably the depression of the conductivity, so that in comparing the milks of different species of animals, the conductivity of each could not necessarily be taken as a quanti- tative indication of soluble ionised salts without a correction being made.
For instance in mares’ milk the protein averages about 1:5 °/, and probably has only a small influence on the conductivity of what may be termed the “protein free whey.” |
On the other hand there is every reason to think that in the milk of smaller animals such as the dog and cat, with their high protein content (quoted as high as 11°15 °/,), the conductivity of the protein free whey would be distinctly greater than that of the fat free milk.
The presence of the protein will diminish the power of the ions to transport electricity by reducing their available free paths; also perhaps the conductivity may be affected by the protein diminishing the ionisation of the salts.
In order to determine the extent, to which the protein in cows’ milk depresses the electrical conductivity, the following experimental methods were employed.
The milk was obtained directly from a cow on the premises and the cream separated from it as quickly as possible. One portion of the separated milk was weighed and then boiled for one hour in a large flask under a reflux condenser. After cooling, the milk was reweighed, and the very small loss o water made good by adding a special distilled water. J
The boiled and raw milks were then placed in two large beakers and in each were suspended dialysing tubes of peritoneal membrane. Each tube contained a measured volume (not exceeding 50 c.c.) of specially distilled water (K = 0:00007).
Toluene was added to the milk in each beaker (in all the earlier experi- ments), and when the weather was warm the beakers were placed in an’ ice chest at a temperature of about 12°C. At intervals 5 cc. samples were measured from the dialysing tubes and from the surrounding milk.
In accordance with physico-chemical laws, after a sufficient length of time,
L. C. JACKSON AND A. C. H. ROTHERA 19
equilibrium should be established between the milk outside the membrane _and the water inside.
The membranes in all later experiments were the prepared peritoneal sacs from the appendices of sheep, known technically as “ Fischblusen-condom” and here referred to as “peritoneal membranes.” They were apparently perfect in action and much to be preferred to parchment for this purpose. ~ We quote however earlier results obtained with parchment as the effects observed are the same as with the peritoneal membranes. Ewperiment 1. 26/10/11-28/10/11.
A. 630 c.c. raw separated milk plus 15 c.c. toluene: 25 c.c. pure distilled water dialysed against it for 2 days. Less than 5 c.c. of a yellow-green faintly opalescent fluid recovered from the parchment dialysing tube.
B. 630 c.c. boiled separated milk (1 hour’s boiling) treated as in A. 5 c.c. clear yellow-green fluid recovered from the parchment.
A. Raw milk. © B. Boiled milk. Before dialysis K=0-00518 Before dialysis began K=0°00514 After 5 K=0-00507 After % K=0-00503 Water & salts K=0-00570 Water & salts K=0:00577 from parchment
Experiment 2. 3 p.m. 27/2/13-8 p.m. 1/3/13.
A. 1250 c.c. fresh separated milk dialysed against two tubes of peritoneal membrane each containing 50 c.c. conductivity water. 15c.c. toluene were added to milk as preservative. Vessel placed in ice chest at 12°C. Evapora- tion prevented by parchment cover placed on mouth of beaker.
B. 1300c.c. fresh separated milk boiled for one hour under reflux con- denser, then dialysed against two peritoneal tubes each containing 50 Cc. conductivity water.
Other details as in A.
27/2/13. Before dialysis.
A. Raw milk and toluene K=0-00500
Conductivity water K=0-00007
B. Boiled milk and toluene K=0-00498
Conductivity water K=0-00007
Dialysis commenced 3 p.m. 27/2/13. i Dialysate Raw Boiled Ps =p ~ Date Time No. milk milk Tube 1 Tube 2
28/2/13 12 noon A K=0-00492 os K=0°00494 K=0-00478 B a K=0-00496 K=0-00460 K=0°00440 28/2/13 6.30pm. A K=0-00486 oe K=0-00496 K=0-00490 3 B — K=0-00486 K=0°00484 K=0-00470 1/3/13 12 noon A K=0-00488 _ K=0°00502 K=0-00508 B _ K=0-00492 K=0-00498 K=0°00494 1/3/13 8 p.m. A K=0-00488 — K=0°00512 K=0-00514 i B — K=0-00486 K=0°00514 K=0-00512
2—2
20 L. C. JACKSON AND A. C. H. ROTHERA
In both experiments quoted it was found that after two days’ dialysis the conductivity of the liquid from the dialysing tubes was greater than that of the surrounding milk.
Boiling the milk has apparently no influence on the end equilibrium.
The second experiment suggested a slower course in the establishment of equilibrium in the case of boiled milk and it might be that some fixation of soluble dialysable salts had occurred as the result of the boiling, the change being slowly reversible. It was realised however that diffusion processes and the three occasions on which the milk was disturbed in taking samples, would be responsible for the rate of dialysis.
Consequently in the experiments which follow, performed in mid-winter, stirrers driven by an electrical motor kept the milk in continuous movement during the whole course of the dialysis. Further, the percentages of fat and proteins in the milk employed were directly estimated and gravimetric estimations of ash in the dialysate were made. As will be seen from the records which follow there is no reason to assign any special importance to the slower establishment of equilibrium to which attention was called in Experiment 2.
Experiment 3. 9/6/13.
1600 cc. fresh separated milk (from one cow) and 5c.c. toluene were dialysed against 50 c.c. conductivity water. The beaker containing the milk and the dialysing tube was placed in a water bath, which during the course of the experiment maintained an average temperature of 8°C. The milk was kept in constant movement by means of an electrical stirrer, and a filter paper, kept slightly moistened, was arranged over the top of the beaker to check evaporation.
An estimation of the fat by Babcock’s method gave the content 0:025 °/, and from a Kjeldahl determination of the nitrogen the protein was found to be 435 °/,, after correcting for dialysable nitrogen and dilution by 35 cc.
water. .
5 c.c. samples were removed at intervals from the milk and the dialysing tube.
Experiment set up 12.30 p.m. 9/6/13.
Date Time Milk
Dialysate 9/6/13 12.30 p.m. K=0-00747 — han 9.30 a.m. K=0:00736 K=0°00843 /6/13 4.30 p.m. K=0-00740 K=0:00843
11/6/13 10.15 a.m. K=0-00740 K=0-00839
L. C. JACKSON AND A. C. H. ROTHERA 21
In this experiment equilibrium appears to have been attained at 4.30 p.m. 10/6/13. Averaging the figures with those obtained at 10.15 a.m. 11/6/13, gives Milk K=0-00740 Dialysate K=0°00841 The differences between these figures, 0:00101, stated as percentage depression of the conductivity gives :—
Soa x 100 or 12-0] %Jp. This 12-01 °/, depression of the conductivity is produced by 4°35 protein in the milk, or 1°/, protein is equivalent to a depression of 2°77 °/, of the
conductivity.
Experiment 4a.
The milk used in this experiment was obtained from a cow in the third month of lactation. Immediately after separation a portion of it was taken and boiled for 45 minutes, the remaining part being kept in the meantime at a fairly low temperature. The two beakers were subsequently placed in a water bath at 9° C. and the milk was stirred as in the previous experiment, but no toluene was added.
A Babcock determination showed 0°1°/, fat in the separated milk and a nitrogen estimation by Kjeldahl’s method fixed the protein content at 3°07°/, after correcting for dialysable nitrogen and dilution with 10c.c. water.
A. 1100 cc. fresh separated milk dialysed against 25 c.c. pure distilled
_ water.
B. 1100cc. separated milk boiled for 45 minutes under reflux condenser,
then dialysed against 25.c. pure distilled water. Before dialysis.
13/6/13. 6 p.m. A. Raw milk K=0°00491 B. Boiled milk K=0-00487
Dialysis commenced 6 p.m. 13/6/13.
Raw Boiled Date Time No. Milk Dialysate - No. Milk Dialysate - 13/6/13 10.45p.m. A K=0-00489 K=0-00496 B K=0-00183 K=0-00508 14/6/13 10.20am. A K=0-00484 K=0-00529 B K=0-00485 K=0-00527 14/6/13 4pm. A K=0-00178 K=0-00526 B K=0-00180 K=0-00523
4c.c. of each dialysate were taken and the amount of ash estimated. The ash of the dialysate from the raw milk amounted to 0°0175 g. and that from boiled milk to 0°0180 g.
22 L. ©. JACKSON AND A. C. H. ROTHERA
In this experiment equilibrium is established at 10.20 a.m. 14/6/18.
Also there is no essential difference between the results for the raw and boiled milks. Consequently the average of the’ conductivities for the last four milks (two raw and two boiled) and for the last four dialysates may be used to calculate the depression due to the protein.
Average for milk K =0-00482 Average for dialysates K=0:00526 Hence conductivity is depressed 8°365 °/, for 3°07 °/, protein or 1 °/, protein depresses the conductivity 2°73 °/,. :
Experiment 4b. 20th and 21st July, 1913. | Eeey 1000c.c. separated milk (morning milk from several cows). Fat 0°25 °/o; K=0-00541. No toluene added. Two dialysing tubes each with 20c.c. conductivity water. Stirrer used throughout. Commenced 2 p.m. 20th July, 1913. Temperature 10'5°. in pil) ee 21/7/13. 11.30 an: Milk - K=0-00528
Dialysate 1. K=0-00577) 9) Il K=0-00579 —
2.30 p.m. Milk K=0-00528 Dialysate I K=0°00577 II K=0-00578 —
Results of Kjeldahl determinations. 5c. of milk at end of experiment = 18°9 c.c. N/10,
5 cc. dialysate I at end of experiment = 0°6 cc. V, /10,
5 cc. dialysate IT at end of experiment = 0°9 c.c. V/ 10, Blank determination = 0°00. . ;
- Hence 18 ¢.c. V/10 represents protein nitrogen of milk = 3:12°/, protein. K is depressed 8°65 °/, by this amount of protein or 1 °/, protein depresses
the conductivity 2°77 °/,. nal 7
Experiment 5. 21/6/13.
The milk used in this experiment was bulk milk obtained from a dairy. The same experimental detail was observed-as in the previous experiments, but in this case the separation of the cream from the milk was by no means complete. No toluene used. |
A. 2000 cc. fresh separated milk dialysed against 40 c.c. conductivity water.
B. 2000c.c. fresh separated milk, boiled for one hour under reflux condenser then dialysed against 40c.c. conductivity water
L. G. JACKSON AND A. @ H. ROTHERA 23
21/6/13. 12.30 p.m. A. Raw milk K=0:00527 _ B. Boiled milk K=0-00514 Dialysis commenced 12.30 p.m. ¥ Raw ? Boiled
Date Time = Milk Dialysate ~~ Milk Dialysate 21/6/18 = 2pm. A K=0-00522 K=0-00447 B K=0-00511 K=0-00451 21/6/13 4 p.m. A K=0°00519 K=0-00540 B K=0°00509 K=0-00564 21/6/13 5 p.m. ae — K=0°00563 B — K=0°00564 21/6/13 = 7.30p.m. A K=0-00522 K=0-00578 B K=0-00512 K=0-00570
At the conclusion of the dialysis, 10 c.c. of each dialysate were taken and evaporated to dryness and ashed. The amount of ash from 10c.c. of the dialysate from raw milk was 0°052 g. and exactly the same amount was given by the dialysate from the boiled milk.
_ The ash was then dissolved in dilute hydrochloric and 0°5 c.c. of a sodium acetate solution added; ammonia was added till the reaction was alkaline and then 1 cc. saturated solution of ammonium oxalate. The mixture was kept at 100° for ten minutes, then the oxalate precipitate transferred to a filter paper and well washed with dilute acetic acid and distilled water. Paper and precipitate were dried at 100° and ashed over a low flame. The ashes at this stage weighed.
(1) From boiled milk dialysate 0°0187 g. (2) From raw milk dialysate 00150 g.
In both cases the ash was moistened with 25°/, sulphuric acid, carefully dried and at first gently then strongly heated. (1) From boiled milk diabysate 0°0165 g. (2) From raw milk dialysate 0°0150 g.
Discussion of dialysis experiments.
The experiments all show the same qualitative results, but owing to absence of analyses for the milk as actually used in Experiments 1 and 2 and also to lack of evidence that equilibrium had definitely been reached, it is only the later experiments which can be used for obtaining an accurate quantitative evaluation of the depressant action of protein on the conductivity of milk.
It is most interesting to compare these values (average 2°76 °/,) with those of Bugarsky and Tang] obtained for the proteins of blood serum (2°5°/, depression of conductivity per 1°/, protein).
We feel that we may claim for Experiment 4a@ and b an almost perfect technique. Not even a preservative agent such as toluene was necessary.
24 L. G. JACKSON AND A. C. H. ROTHERA
Also Experiments 3 and 4a give uniform results with two totally different milks. In Experiment 3 milk very rich in protein was used from a cow at the end of lactation (8th month). In Experiment 4a the milk was from a cow in the third month of lactation. Experiment 5 was with market milk, which having previously creamed repeatedly clogged the separator. Owing to the large amount of fat present the experiment cannot be used to evaluate the depressant action of the protein upon the conductivity.
Analyses of ash show that the dialysates from raw and boiled milk not only contain the same amounts of ionised salts, as shown by their equal conductivities, but also the same amounts of total ash, and of calcium. Séldner’s statement [1888] that boiling milk renders a great portion of the calcium insoluble, a statement which has often been used to explain the absence of rennet coagulation in boiled milk, is incorrect in the strict sense. No calcium which is in true molecular solution can become insoluble or results, such as those just quoted, would not be obtained. The chief effect of boiling milk is a physical alteration of the colloids. Probably an insoluble irreversible calcium caseinogenate is formed which mechanically fixes colloidal calcium phosphate.
Such an explanation will account for Sdldner’s observation that with boiled milk less calcium will pass a porous clay filter than is the case with raw milk.
DAILY FLUCTUATIONS IN CONDUCTIVITY OF MILK ASSOCIATED WITH VARYING CLIMATIC CONDITIONS.
During the summer months in Victoria rapid changes from hot dry weather to a period of coldness and rain are experienced from time to time, and the following observations which extended over portions of the months of November, December, February and March include well marked extremes of summer temperature.
The milk used in this series of conductivity determinations was a special | supply from the Syme Model Farm, Gisborne, Victoria. This farm possesses modern refrigerating machinery. The milk is immediately put over the coolers and sent packed with ice by motor to Melbourne. The conductivity changes are therefore not due to bacterial changes in the milk but to climatic influences affecting the cows. As soon as the supply arrived at the Melbourne depot a composite sample from 3-10 cans, each containing 50 quarts, was carefully collected by the Dairy Expert of the Willsmere
Certified Milk Company and our thanks are due to him and to the for obtaining these samples for us.
Company
L. C. JACKSON AND A. C. H. ROTHERA
25
The weather reports quoted in the following table are abridged from the daily bulletins published in the Argus; the figures for humidity and the dry bulb temperature as recorded by the Melbourne Observatory were obtained from the same source.
Date 27/11/12 98/11/12 29/11/12
1/12/12 3/12/12 4/12/12
5/12/12 6/12/12
8/12/12 9/12/12
10/12/12 11/12/12 15/12/12
17/12/12 12/ 2/13 16/ 2/18 17/ 2/13 18/ 2/13 19/ 2/13 20/ 2/13 24/ 2/13
26) 2/13 27/ 2/13 2) 3/13 3/ 3/13 4] 3/13
6/ 3/13
Weather report
Fine, warm, pleasant
Hot, dry, north winds
Small fall of rain, followed. by $i: pleasant day, light cool winds
Light steady rain, sultry conditions
Tropical downpour, severe thunderstorm
Dull cloudy morning, followed by fine pleasant day, light southerly wind
Cloudy, sultry generally
Thunderstorms, followed by squalty showery weather
Unsettled, showery, ‘‘ unusual weather”
Cloudy but fine at first, gradually be- coming windy and threatening
Weather very disturbed, ot Cool, showery
Weather becoming finer, max. temp. higher than any recorded since 6th Dec.
Hot, sliglitly cloudy, light breezes
Fine, light variable winds and little rain
Close, sultry
Fine, warm, sultry ... ‘
Fine, unpleasant strong iaaesly waits
Fine, after a cloudy threatening morning
Cloudy & dull at first, afterwards bright
Northerly winds at first, followed by light sea breezes. Very hot
Cloudy, sultry, thundery
Bright, warm, sea breezes
Fine, clear, hot
Warm night followed by dull igltig dae: Thunderstorm
Oppressive weather, high humidity, small amount of rain
Cold with squally southerly wind, at times showery
Relative Dry bulb temperature, oF.
humidity -
9% Max. Min. Mean 57°3 77°4 43°2 60°3 37°6 = 83°6 53°7 68°6 59°6 67°3 57°9 62°6 74:3 73°9 61-1 67°5 74:3 75°0 49-0 62-0 63°0 = =70°3 53°0 61-6 49-0 84°7 51:7 68-2 56°3 68°8 56°4 62°6 63-0 67°3 47°3 57°3 60°6 67:7 54:2 60°9 61°3 62°3 50°7 56°5 61:0 57°6 44:2 50°9 56°6 70°2 475 ~=—s- 5588 52°3 83°7 59°3 71°5 61:0 82:8 55°2 69°0 74°6 74°8 64:5 69°6 57°3 850 =: 608 72°9 436 78°0 60-3 69-1 49°3 73°0 = 575 65°2 59°3 66°6 57°74 62°0 443 950 59°9 717-4 69-0 71°0 62:0 66°5 68°3 84-0 60-0 72°0 486 93-4 48°5 70°9 51°6 95°3 70°2 82°6 81°3 76°5 64°9 70°7 68°3 62°8 55°0 58°9
~
K 0:00578
0°00573 0:00562
0°00570 0°00557 0°00554
0°00555 0:00554
0°00562 0-00556
0°00554 0°00549 0°00568
0°00573 0-00585 0:00587 0°00562 0°00568 0-00562 000559 0°00573
0°00555 0°00577 0-00573 0°00575 0°00551
0-00562
By reference to the table it will be seen that on November 27, 28, 29 and December 1, warm weather was experienced and the average conductivity obtained for the milk during that time = 0:00570.
On December 3 there was a large fall of rain and a lowered conductivity recorded = 0:00557.
On December 5 a quick cold change followed a hot, sultry day and from that date until December 11 there was a period of cold, showery stormy
weather and the average conductivity during this time = 0:00555.
26 L. @. JACKSON AND A. C. H. ROTHERA
- On December 15 and 17 the weather was becoming finer with rising temperatures and the mean conductivity for these two days = 000576. On various days between February 15 and 19, both inclusive, fairly high temperatures were recorded and the average conductivity for 5 days= 000572.
On February 24 and 27 and March 2 and 3 the dry bulb temperatures
were high and the average conductivity = 000574.
February 26 was cloudy and thundery with moderate temperatures and K = 0:00555.
The weather of March 4 was very oppressive with a high humidity and K =0°00551.
By March 6 the weather was onlee and squally and the conductivity v was found = 0°00562. :
It will thus be seen that as a generalisation it may be stated that hot dry weather causes the conductivity to become higher, while cold, wet or very humid weather has the reverse effect. :
The value given for the relative humidity is the mean of three determina- tions made at 9 a:m., 3 p.m. and 9 p.m. respectively. .
The maximum and minimum dry bulb temperatures are those recorded during twenty-four hours. |
The conductivities quoted in the table are all for early morning samples of milk and placed against them are the climatic details of the previous day. The conductivities obtained ranged from 0:00549-0: ‘00587.
SUMMARY...
(1) Inmilks secreted from different quarters of the cow’s udder and from the right and left breasts in women, the electrical conductivity and percentage of milk sugar show a strict reciprocity provided that the secretion of the milk samples corresponds to the same period. In this case the milk samples are
secreted against the osmotic pressure of the blood with its variation over that
period, . They will all have the same osmotic pressure and if i in one sample the sugar is higher than in another, then the electrical conductivity will be lower.
(2) The reciprocity of milk sugar content and electrical conductivity is well seen in the milks from a pathological gland which i is slowly recovering and becoming normal.
(3) In a comparison of the milks of diffetont species of animals the reciprocity between milk sugar and electrical conductivity is evident.
— ate BN “aS — Cae hea 7 * f
L. C. JACKSON AND A. C. H. ROTHERA 27
(4) It is shown that the milk secreted under the stimulus of removal (milking or natural suckling) differs in character from that secreted pre- viously. The contention is that the condition of the blood has not kept absolutely constant, and that the reflex milk is secreted against a slightly different blood from that against which the previously formed milk was secreted.
(5) Morning and evening samples of cows’ milk have been compared. The evening milk generally has the higher conductivity but exceptions exist.
(6) The exact effect of the proteins of cows’ milk in diminishing the electrical conductivity has been estimated, the value found being a diminu- tion of 2°76 °/, of the conductivity for every 1°/, of protein in the milk.
(7) The dialysis experiments employed for determining this effect showed no difference between raw fresh milk, and the same boiled for one hour. Also there was no evidence that boiling has any effect on soluble calcium salts in a state of ionisation.
(8) The effect of climatic changes upon a Holstein herd of cows has been studied. The generalisation holds that hot dry weather increases the electrical conductivity of the milk, whilst wet or cold weather diminishes it. The climatic conditions affect the cows and so indirectly their milk.
REFERENCES.
Barthe (1904), J. Pharm. Chem. 20, 355. Binaghi (1910), Biochem. Zeitsch. 29, 78. Bugarsky and Tangl (1898), Pfliiger’s Archiv, 72, 531. Dreser (1892), Arch. expt. Path. Pharm. 29, 303. Grassi (1906), Ann. Ostetr. Ginecol. Guiraud and Lasserre (1904), Compt. Rend. 139, 452. Hotz (1902), Dissert. Ziirich. ; Phys. chem. Untersuch. Phys. Path. Kuhmilch. Koeppe (1898), Untersuch. iiber Salzgehalt Kuh- und Frauenmilchs. Habilit. Schrift, Giessen. Lehnert (1897), Dissert. Erlangen. Parmentier (1903), Miinch. Med. Woch. 758. Petersen (1904), Maly’s Jahresber. Thierchem. 34, 330. Pins (1912), Mileh. Zentr. 18. Scheibe (1901), Zeitsch. anal. Chem. 40, 1. Schnorf (1905), Dissert. Ziirich.; Phys. chem. Untersuch. Phys. Path. Kuhmilch. Sdldner (1888), Die Salze der Milch. Dissert. Nérdlingen. Sommerfeld (1909), Handbuch der Milchkunde. Strauss (1900), Kongress fiir innere Medizin. _ Villejean (1905), La crioskopie du lait. These de Paris. Winter (1895), Compt. Rend. 121, 696.
II. OSMOTIC PHENOMENA OF YOLK OF EGG.
By WILLIAM ALEXANDER OSBORNE anp HILDA ESTELLE KINCAID.
From the Physiological Laboratory, University of Melbourne. | (Received November 21st, 1913.)
The following short research arose from an attempt to employ yolks of eggs as models illustrating the osmotic behaviour of red corpuscles. An unbroken yolk immersed in distilled water slowly swells, the contents become cloudy and ultimately the membrane bursts. If another yolk be placed in 0°9°/, sodium chloride no change is observable. A third yolk floated on glycerol will shrink and display marked corrugation. So far there is close parallelism and these three experiments can be used for class demonstration. With strong salt solution however and with organic solvents the behaviour of the yolk is widely different from that of corpuscles, as might in part be inferred from the sclero-proteid nature of the vitellin membrane.
If the yolk be floated on to 10 °/, NaCl, or greater concentrations, there is
a marked swelling and not the shrinking which one might expect. As is well —
known the chief protein, or lecithoprotein, of yolk is soluble in strong saline. When dissolved it permeates either not at all or with great difficulty through the vitellin membrane and so conditions an osmotic effect. A series of 2M. solutions of NaCl, CaCl,, MgCl, and MgSO, all gave this swelling effect. With the sulphate solution the rate of distension was slightly greater. If a 50 */ CaCl, solution or a higher concentration be employed, shrinking can nevertheless be seen owing to the rapid extraction of water.
With the following solvents more or less interesting effects can be observed.
(1) Ether. The yolk sinks and slowly swells. About the end of the second day an accumulation of ether, deeply pigmented but transparent, may be observed in the upper part of the yolk. This ethereal solution increases in volume and produces a doming of the yolk. In some cases the latebra is beautifully displayed in the form of a tent attached to the vitellin membrane at the cicatricula. Eventually the membrane bursts liberating the contents,
AT
W. A. OSBORNE AND H. E. KINCAID 29
But as long as the membrane is intact the ether outside is unstained and indeed does not contain even a trace of solid matter. The dissolved substance is therefore imprisoned in the yolk and exerts its osmotic effect. If the yolk be placed in ether which has for some time been shaken up with broken yolk and separated from this, the swelling may be completely absent and no collection of ethereal solution may be observed.
(2) Chloroform. The yolk floats on the fluid but the sequence of events is very similar to that with ether except that the growing volume of coloured chloroform is found at the bottom of the yolk. So long as the membrane is intact the outside chloroform is unpigmented.
(3) Carbon disulphide. In every particular the action of this solvent is similar to that of chloroform.
(4) Alcohol. The yolk which sinks in this fluid does not swell and very soon the outside fluid is seen to be pigmented, though the membrane, as far as the eye can judge, is intact. Apparently the alcoholic solution can pass through the membrane just as alcoholic solutions of soap can diffuse through parchment paper; hence no osmotic effect is obtained.
(5) Petroleum ether. The yolk sinks in this but does not change. There is no extraction of pigment nor accumulation of solution inside. If broken yolk be shaken up with this solvent nothing but a mere trace of fatty matter is dissolved. The absence of osmotic effects is due therefore to an absence of solution. Boiling petroleum ether can however exert a solvent action on some yolk constituents and an unbroken yolk placed in this under a reflux condenser will show swelling and formation of a globule of solution under the membrane.
(6) Benzene. This acts in a manner almost identical with that of petroleum ether.
(7) Acetone. There is a gradual extraction of colour from the yolk; prolonged action may give a slight globule. If boiling acetone be employed the extraction of pigment proceeds more rapidly.
(8) Olive oil. No change is observable.
(9) Isotonic urea solution. The yolk swells fairly quickly and a globule is formed on the top as in ether. If a 4°/, solution of urea in 0°9 °/, NaCl be employed there is no effect. Urea solution can therefore be added to those mentioned at the outset of the paper as giving effects similar to those observable with red blood corpuscles.
a" ee
III. CASEINOGEN AND CASEIN. By ARTHUR GEAKE.
From the Bio-Chemical Laboratory, Chemical Department, A University of Bristol.
(Received November 29th, 1913.)
The question of the chemical identity of caseinogen and casein and with it, of the nature of rennet action, still remains open in spite of many attempts at a solution. Hammarsten suggested that by the action of rennet, caseinogen was hydrolysed to form two new proteins, casein and whey-protein. Koster [1881] analysed these proteins and found that casein contained somewhat more nitrogen (15°84°/,) than caseinogen and whey-protein much less (13:1-13°6 °/,). On the other hand it has been suggested that rennet action is either purely physical or is only concerned with the inorganic constituents of milk and that casein is chemically identical with caseinogen. During the progress of this work a new suggestion has been put. forward by van Slyke and Bosworth [1913], according to whom caseinogen is split by rennet into two molecules of casein.
If Hammarsten’s theory is correct, it should be possible to find some chemical differences between caseinogen and casein. Elementary analyses have yielded the following results :—
Caseinogen C H N $§ bs
Makris [1876] 53:02 7:42 14-20 = ~~ Hammarsten [1883-1885] 52°96 7°05 15°65 0°758 0°347 Chittenden & Painter [1887] ‘ 53-3 7:07 15-91 0-82 0-84-0°89 Lehmann & Hempel [1894] 540 7:04 15°6 0-771 0847 Ellenberger [1902] 53-07 713 15-64 0°76 08 . Laqueur & Sackur [1903] — — 15°45 0°757 0°772 Burow [1905] 52-825 7-095 15°64 0-725 0-808 Tangl [1908] 52°69 6°81 15°65 0°8382- = 0877 van Slyke & Bosworth [1913] 53°50 7:13 15°80 0°72 0-71 Mean 53°17 7:09 15°67 0°768 0-82 Casein Koster [1881] is ae 15°84 “a — Rose & Schulze [1885] 53°94 7:14 15°14 1:01 oe Raudnitz [1904] — — 15°5 0°7-0°88 Kikkoji [1909] = set = 0°85-0°87
van Slyke & Bosworth [1913] 53°50 7°26 15°80 0°72 0°71 Mean 53°72 7°20 15°57 0°87 0°79
A. GEAKE 31
The only reliable analysis of casein is that of van Slyke and Bosworth [1913] who obtained identical results for caseinogen and casein. They have also obtained somewhat more sulphur than phosphorus, which is in agreement with the supposition that there are equal numbers of atoms of these two elements in the molecules of these proteins.
The object of the following investigation was to determine the difference, if any, between caseinogen and casein both in elementary composition and in Hausmann numbers.
Most of the nitrogen estimations quoted above have been carried out by Kjeldahl’s method and are thus inclined to be too low. The author has made estimations both by Kjeldahl’s and by Dumas’ method. Sulphur estimations are usually carried out by one of the various fusion methods described in the literature. These all lead to the precipitation of barium sulphate in the presence of comparatively excessive amounts of alkali salts, usually chlorides. It has however been shown by Allen and Johnston [1910] and Johnston and _ Adams[1911] that the presence of even relatively small amounts of such salts destroys the accuracy of sulphate estimations. This source of error has been avoided in the analyses described below by adopting a slight modification of Carius’ method.
The following mean results were obtained for caseinogen and casein :—
Caseinogen Casein “lo lo
Cc 53°20 53°05 H 7°09 7°03 N (Dumas) 15°63 15°81 N (Kjeldahl) 15°61 15°62 8 1-015 1-009 rE 0-731 0°809
The results for sulphur are higher than have been previously obtained for caseinogen. It will be seen above that Rose and Schulze [1885] obtained 1-01 °/, sulphur in casein in agreement with the author’s result. The sulphur contents of caseinogen and casein appear to be identical, but casein apparently contains more phosphorus than caseinogen. The difference is however not sufficient to warrant the supposition that the two proteins are chemically different.
As will be seen feom the following figures the Hausmann numbers for the two proteins are also too close to establish any definite difference.
Caseinogen Casein
°/) total N °/, total N Ammoniacal N 10°23 10°31 _Melanin N 1°53 1-66 Diamino N 22°94 24-03 Monamino N 65°31 63°90
100°01 99°90
32 A. GEAKE Osborne and Harris [1903] obtained from caseinogen °/, total N°
Ammoniacal N 10°31 Melanin N 1°34 Diamino N 22°34 Monamino N 66°01
100-00
with which the author’s results are in substantial agreement.
EXPERIMENTAL PART. ELEMENTARY ANALYSIS OF CASEINOGEN AND CASEIN.
The caseinogen used was Kahlbaum’s “Casein nach Hammarsten ” carefully freed from fat by prolonged extraction in a Soxhlet apparatus with ether.
The casein was prepared from milk by the action of rennet and purified by Hammarsten’s method. The last traces of fat were removed as above. Two specimens were separately prepared and analysed.
Before analysis the samples were allowed to stand at least 16 hours exposed to the air of the balance room, and when each portion was weighed out for an estimation a second portion was taken and dried to constant weight at 40° in vacuum over P.O, and the results corrected for the percentage of water thus found. It was found impossible to handle the very hygroscopic dry proteins with any certainty of accuracy. The ash was also estimated and allowed for. All the results are calculated for the ash-free dry proteins.
Carbon and hydrogen estimations were carried out in an ordinary combustion tube filled with alternate layers of copper oxide and lead chromate. The substance in the boat was covered with a mixture of lead chromate and potassium bichromate.
Nitrogen estimations were carried out both by Kjeldahl’s and by Dumas’ method.
Sulphur was estimated by the following modification of the Carius method. About 0°5 grm. of the protein was heated in a sealed tube with 7-8 ce. fuming nitric acid for two days at 300°. After opening the tube the contents were diluted and filtered and the filtrate treated with a slight excess of barium chloride over that required to combine with the sulphuric and phosphoric acids. No precipitate was obtained. The solution was evaporated to dryness, a few cc. of concentrated hydrochloric acid added and again
evaporated to dryness completely to remove the nitric acid. The residue
aa ee! foe aoe a ee
A. GEAKE 33
was taken up-with 50 cc. of water and 22 or 45 cc. of dilute (6°5 °/,) hydrochloric acid added. The barium sulphate was thus obtained in a granular condition. It was collected on a layer of “BaSO, asbestos ” in a Gooch crucible, washed and dried in an air oven at 110-120° to constant weight. From Allen and Johnston’s [1910] results the following corrections were made for the solubility of barium sulphate :—
2-2 ec. dilute HCl add 0-5 mgm. BaSO,. 4-5 ec. dilute HCl add 0°6 mgm. BaSQ,.
_ Phosphorus was estimated in the filtrate from BaSO, by Gregerson’s [1907] modification of Neumann’s [1900] method. Caseinogen. Ash : : reed =0-57 °/, except where otherwise stated.
C & H 0-2118 g. containing 10-30 °/, H,O; 0-3690 g. CO,; 0-1403 g. H,0.
0-2214 g. - 6-54 °/, H,O; 04010 g. CO,; 0-1492 g. HO. 0-2415 g. as 654 °/, H,O; 0°4377 g. CO,; 0°1549 g. H,O. N (Dumas) 0-2227 g. containing 6-54 °/, H,O; 27-2 cc. nitrogen over 50 °/, KOH at 15°5° C. and 760 mm.
N (Kjeldahl) Ash= 10-30} =0-26 ),; H,O=11°30 °/,. 0-2510 g. required 24°63 cc. N/10 H.SO,.
0-2494 g. be 24-72 ce. oe 0°2507 g. is 24-66 ce. ie 0°2493 g. ae 24-43 ce. 54 S 0°5008 g. containing 7°69 °/, H,0; 0°0332 g. BaSO,. ;
0-5055 g. a 7-69 9], H,O; 0-0354 g. BaSO,. P 0°4936 g. containing 7-69 °/, H,O required 29-61 cc. N/10 NaOH.
0°5055 g. os 7°69 °/, H,O a 30-71 ce. es
Ash-free dry caseinogen : Mean C=53°25, 53°15, 53°20 aS ae 53°20 Jy H=7-03, 7°32, 6°93 .:. re =p 7°09 N (Dumas)=15°63__... 15°63 N (Kjeldahl) = 15-54, 15-76, 15 61, 15- 54 15°61 S=0-992, 1-048 Bey “é pe 1-015 P=0°727, 0-734 ys ee ore 0-731
Casein (Prep. I). Ash=0°31 °/, of dry casein. C & H 0°2408 g. containing 11°75 %/) water; 0°4133 g. CO,; 0°1576 g. H,O. 0°2040 g. - 11-70 %o water; 0°3493 g. CO,; 0°1390 g. H,O. N (Dumas) 0-2792 g. containing 11°75 %/) water; 33°6 cc. nitrogen over water at 12° C. and 742 mm. _ 0°3342 g. containing 11-70 °/) water; 40°0 cc. nitrogen over water at 15°3° C, : and 749 mm. S 0°5063 g. containing 11-75 °/) water; 0°0323 g. BaSO,. * 0-4939 g. ie 10-09 %Jy water; 0°0295 g. BaSO,. 0-4250 g. dry casein; 0°0312 g. BaSOy,. P 0-5063 g. containing 11-75 °/) water required 33°33 cc, N/10 NaOH. 0°4939 g. 3 10°09 %_ Cs, - 32°30 cc. N/10 NaOH.
Bioch. vat 3
34
A. GEAKE Dry ash-free casein (Prep. I): Mean = nary eet AGS es Ee 53°14 9/o H=6:82, 7°17... ez es ee 7:00 N (Dumas) = 15°85, 15-68 a ag 15°77 S=0-999, (0-919), 1:008 oe bi 1-004 P=0°829, 0°809 ve ie ay 0°819
Casein (Prep. II). :
Ash=0°'45 °/) of dry casein. C & H 02268 g. containing 11-95 %/) water; 0°3867 g. CO); 0°1539 g. H,0. 0°2127 g. a 11°94 °/) water; 0°3616 g. CO,; 0°1413 g. H,0.
N (Dumas) 0°2487 g. containing 11°95 °/) water; 30-1 cc. nitrogen over water at 17-2° C. and 754 mm, 02886 g. containing 11-94 °/) water; 33°7 cc. nitrogen over 50°/) potash at 15°7° C. and 756 mm.
8 0°5318 g. containing 11-94 °/) water; 0°0340 g. BaSQ,.
- 0°5154 g. 5 11°94 9/9 water ; 0°0337 g. BaSO,. P 05318 g. containing 11-94 °/) water required 34-05 cc. N/10 NaOH. 0°5154 g. a 11:94 %J) ,, a 32°08 ce. ri
Dry ash-free casein (Prep. II): Mean C=53°02, 52°88 ae aie ae 52°95 %/p H=7'14, 6°95... os a oat 7°05 N (Dumas) =15°88, 15°81 “5 ni 15°85 S=1-002, 1-024 = iss oe 1-013 P=0:810, 0°787 4 "te m 0°799
Kjeldahl nitrogen estimations of casein were carried out with a third
specimen prepared by the same method.
Ash=1-03 %) in dry casein.
11-41 Water = nasi =11°44 %p.
0-2478 g. required 24-13 ec. N/10 H,SOj. 0°2517 g. . 24-61 ce. 3
N=15'59, 15°65. Mean=15-62 %J).
The mean compositions of caseinogen and casein are thus :—
Caseinogen Casein "lo "lo
C 53°20 53°05 H 7:09 7:03 N (Dumas) 15°63 15°81 N (Kjeldahl) 15°61 15°62 8 1015 1 009 ¥ 0°731 0-809
Hausmann Numbers.
The caseinogen used was Kahlbaum’s “Casein nach Hammarsten,.” The ash and water were estimated and allowed for as above.
A. GEAKE 35
Three samples of casein were used.
(1) Prepared by the action of rennet on a solution of caseinogen in disodium hydrogen phosphate. The casein was precipitated by calcium chloride and purified by Hammarsten’s method.
(2) and (3) Prepared by the action of rennet on a solution of caseinogen in the minimum amount of NaOH. The rennet was destroyed by heating momentarily to 90° C. with steam and the casein precipitated with glacial acetic acid. It was purified by Hammarsten’s method. At each purification after the first a portion was washed with alcohol and ether and not further purified. In this way four fractions were obtained.
From 110 g. air-dry (about 100 g. dry) caseinogen were obtained :—
Prep. (2) Prep. (3) g. g- Fraction a 12 13 b 11 16 c 13 12 d 14 11 50 52
The fractions (2)a, (2)d, (3)b and (8)d were used for Hausmann numbers. |
The Hausmann numbers were estimated by the method given by Réhmann [1908]. Usually the diamino nitrogen was estimated in the phosphotungstic acid precipitate and the monamino nitrogen in the filtrate from this, in some cases only one of these two was estimated and the other obtained by difference. Figures thus obtained by difference are given below in brackets. The results are given in percentages of the total nitrogen obtained by addition of the four separate nitrogen percentages.
Caseinogen. 1st Series. (1) (2) (3) (4) Mean Ammoniacal N 10°52 10°64 10-06 - 9-73 10°24 Melanin: N 1-79 0°58 2-07 1-57 E 1-50 Diamino N (23°07) (26-19) 20°00 20°92 22-55 Monamino N 64°62 62°59 67°87 67°78 65°72 100-01 2nd Series. (5) (6) (7) (8) (9) Mean Ammoniacal N 10-03 10°32 10°35 10°26 10°17 10-23 Melanin N 1-34 1°55 1°57 1°59 1-71 1°55 Diamino N 23-28 22°66 23°77 22°98 23-56 23-25 Monamino N 65°35 65-48 64°31 65°23 64°57 64°99
The second series is the more reliable,
36 A. GEAKE
in (Prep. (1)). Casein (Prep. (1)) 3 ay ‘aj
Ammoniacal N 9°92 10°38 10°26 Melanin N 1-50 1:57 1°63 Diamino N (23°89) (25°59) 24°05 Monamino N 64°69 62°46 64:07
Casein (Prep. 2a). (1) (2) (3)
Ammoniacal N 10°78 10°38 9°87 Melanin N 1°46 1°43 1°49 Diamino N 23°94 25°14 22-18 Monamino N 63°82 63°05 66°48
Casein (Prep. 2d).
(1) (2) (3) Ammoniacal N 9-91 10-31 10°17 Melanin N 1-66 —_ 2:28 Diamino N — — 25°30 Monamino N — 62°99 62°24 Casein (Prep. 3b). (1) (2) (3) Ammoniacal N 10°42 10°29 10°55 Melanin N 1°59 2-00 1°51 Diamino N (28-06) (20°67) 24-66 Monamino N 59°92 67:04 (63-28) Casein (Prep. 34d). (1) (2) Mean Ammoniacal N 10°65 10:00 10°33 Melanin N 1-65 1-70 1:68 Diamino N 22°47 22°11 22:29 Monamino N 65°23 66°19 65°71 100°01
The mean of all the results gives the following values for caseinogen and
casein :— Caseinogen Ammoniacal N 10-23 Melanin N 1°53 Diamino N 22-94 Monamino N 65°31 100°01
Casein 10°31 1°66 24-03 63°90 99-90
Mean 10°27
1°52 24°37 63°85
100°01
A. GEAKE 37
CONCLUSIONS.
The difference between caseinogen and casein is thus scarcely appreciable. If for caseinogen the second series alone is taken, the difference is still less than that given above. No difference could be established between the different fractions of casein.
In conclusion I wish to thank the University Colston Research Committee for a grant which defrayed most of the expenses of this work.
REFERENCES.
Allen and Johnston (1910), J. Amer. Chem. Soc. 32, 588.
Burow (1905), Inaug. Diss. Basel. .
Chittenden and Painter (1887), Studies from the Yale University, 2, 156. _ Ellenberger (1902), Arch. Anat. Physiol. Suppl. 313.
Gregerson (1907), Zeitsch. physiol. Chem. 53, 452.
Hammarsten (1883), Zeitsch. physiol. Chem. 7, 220.
—— (1885), Zeitsch. physiol. Chem. 9, 273.
Johnston and Adams (1911), J. Amer. Chem. Soc. 33, 829.
Kikkoji (1909), Zeitsch. physiol. Chem. 61, 139.
Késter (1881), Biol. Centralbl. 2, No. 2; Maly’s Jahresberichte, 11, 14.
Laqueur and Sackur (1903), Ergeb. d. Physiol. (Asher and Spiro), 2, Abt. I, 232. Lehmann and Hempel (1894), Arch. ges. Physiol. 56, 558.
Makris (1876), Inaug. Diss. Strassburg.
Neumann (1900), Archiv Anat. u. Physiol. See also Zeitsch. physiol. Chem. 37 and 43. Osborne and Harris (1903), J. Amer. Chem. Soc. 25, 323.
Raudnitz (1904), Monatsh. Kinderheilk. 2.
Roéhmann (1908), Biochemie, p. 299.
Rose and Schulze (1885), Landw. Versuchsstat. 31, 115.
van Slyke and Bosworth (1913), J. Biol. Chem. 14, 203.
Tangl (1908), Pfliiger’s Archiv, 121, 534.
ae
IV. A CONTRIBUTION TO THE STUDY OF A PROTEOLYTIC ORGANISM.
By JOHN MALCOLM DRUMMOND. From the Frankland Laboratory, Manchester University.
(Received December 22nd, 1913.)
During a recent investigation of the bacterial population of a number of sewage sludges, the author isolated an organism which possessed the property of causing the rapid liquefaction of gelatin.
It was thought that a more detailed examination of this organism and its mode of action might prove of interest. Accordingly a pure culture on gelatin peptone bouillon was obtained in the usual way by the method of repeated “plating” and the characteristics of the organism were investigated.
On gelatin plates slightly greyish colonies develop in from 24 to 36 hours. These colonies are round with whitish opaque centres and the plate is rapidly liquefied.
Gelatin peptone bowllon (G.P.B.). A cup of liquefied medium appears within 48 hours and a funnel-shaped growth can be observed along the track of the needle. Liquefaction continues downwards, finally yielding an amber coloured viscous liquid which in old cultures darkens in colour and becomes ~ more mobile. A dense white deposit forms at the bottom of the tube leaving the supernatant liquid transparent. A putrefactive odour is noticed which disappears as the age of the culture increases. .
Glucose gelatin peptone bouillon. Ina shake culture, gas production occurs and later liquefaction sets in on the surface of the gelatin.
Nitrate peptone bouillon. Nitrate is reduced to nitrite.
Ammonia is produced in peptone cultures.
Potato. A brownish growth with little tendency to spread.
Agar peptone bouillon. A white fern-shaped growth spreading along the track of the needle. :
Milk. Coagulation followed by alkalinity. The characteristic reduction of neutral red and gas evolution occur. The organism is, therefore, a member
- J. M. DRUMMOND 39
of the Proteus or intestinal group and the laboratory number D4 was assigned to it. Microscopically the organism is a short broad bacillus of dimensions 14x06. Its length, however, varies somewhat. It can exist singly but generally forms chains often of considerable length. It.is motile, does not form spores and in old cultures usually takes up an involution form.
To ascertain whether the proteolytic action of the organism was due to the presence of an extracellular enzyme the following experiments were carried out. A short stab culture was made in nutrient gelatin and the organism allowed to develop until about one-fourth of the gelatin was liquefied. To this was added in one case toluene and in another a few drops of water saturated with thymol, and the tubes incubated at 21°. Liquefaction continued, and since the action of most organisms is inhibited by thymol and toluene it follows that a proteolytic enzyme was probably secreted by the bacillus.
In other experiments a large quantity of nutrient gelatin was liquefied by inoculation with the bacillus and the products submitted to filtration ‘through a Chamberland filter candle. By this means an absolutely sterile filtrate was obtained. This was checked by plating: no colonies developed and microscopical examination of the filtrate failed to show the presence of any bacteria.
With this filtrate experiments according to the method described by Fuhrmann were made.
100 grams of gelatin were dissolved in 1 litre of water saturated with thymol. A little very finely powdered cinnabar was then added and the mixture thoroughly shaken.
15 to 20 cc. of this mixture were poured into each of a number of tubes, which were placed in a sloping position and a jet of cold water allowed to flow over them until the gelatin began to set. They were then kept in an upright position and the gelatin allowed to solidify completely ; the cinnabar becomes evenly distributed through the gelatin.
The sterile filtrate was then poured into these tubes in sufficient quantity to cover the top of the gelatin slope, a little thymol water being added to _ prevent the growth of moulds ete.
Liquefaction set in readily and its progress could be followed by the gradual disappearance of the red gelatin slope, the cinnabar as liberated collecting at the junction of the enzyme solution and the solid gelatin. Toa portion of the sterile filtrate absolute alcohol was added and the precipitate
obtained filtered off. ‘ After drying im vacuo at 40° to remove alcohol, this precipitate was
40 3 J. M. DRUMMOND
redissolved in sterile water containing thymol and the above experiments repeated with similar results.
Two tubes of gelatin peptone bouillon were next taken and the medium _ melted. ‘To one was added a little of the precipitate obtained as above and a few drops of thymol water, while the second tube was kept without additions, to act asacheck. Both tubes were incubated at 37° for 24 hours and then placed in ice when the check tube of gelatin at once solidified while that containing the precipitate remained liquid after standing in ice for over one hour.
Thus the precipitate contained the proteolytic enzyme, and while its action was rapid at 37° and less so at temperatures below this, no optimum could be obtained since the gelatin was liquid at 27° and consequently no rate of action could be determined above that temperature.
These experiments, therefore, prove that the power of causing the lique- faction of gelatin, exhibited by the organism, is due to the presence of an extracellular proteolytic enzyme.
It had been observed in a previous experiment that upon standing in the presence of the organism, the enzyme contained in the products of liquefaction of nutrient gelatin was destroyed, whereas the sterile enzyme solution retained its activity, at any rate for several days. It was proposed to endeavour to determine when the enzyme first makes its appearance and when it is destroyed and a number of experiments were made with this end in view.
Several tubes of the following description were made. A piece of narrow glass tubing about 2 inches long and 4; inch internal diameter was fused to a thin glass rod 4 inches in length. The tube portion was graduated in millimetres and filled completely with thymol gelatin containing a small quantity of cinnabar. Before each tube was filled it was gauged by inserting a wooden peg in the open end. By this means it was ensured that the same surface area of gelatin would be in contact with the solution in every case and thus comparable results be obtained. |
The rod attached to this tube was passed through a cork of such a size as to fit a number of culture tubes. Cultures were made in peptone water, inoculation being made by loop from an old culture of D4 in gelatin peptone bouillon, and after 24 hours’ incubation at 37° toluene was added to one tube, thoroughly shaken and one of the above described tubes introduced so that about 2 millimetres of the gelatin tube were below the surface of the peptone water. 72 hours after inoculation a second tube was treated in a similar
manner and also other tubes, 192 hours and 888 hours respectively after inoculation,
J. M. DRUMMOND 41
Any liquefaction of the gelatin that might occur would be due to the presence of the enzyme and the rate would be proportional to the amount of enzyme present.
The limit to which liquefaction had proceeded could easily be read on the graduated scale, a sharp line being discernible between the solid and liquid gelatin; the former being red and opaque when viewed by reflected light and the latter clear and colourless. The results of one set of experiments are given below.
(i) 24 hours culture, liquefaction after 20 days= 0-0 mm.
(ii) 72 2 > oe 2 = 88 2 (iii) 192 ,, ss 2 2 =22'5 ,, (iv) 888 ,, 2 os 2” =16°4 ”
The enzyme was not secreted at once but increased in amount as the age of the culture increased up to a certain point after which it was slowly destroyed.
This disappearance of the enzyme also occurred when the organism was cultured in peptone bouillon.
(v) 48 hours culture, liquefaction after 30 days=44°0 mm. (vi) 49 days ” ” 33 ” = 68 ,,
From the foregoing experiments it is seen that the enzyme was secreted when the organism was grown upon nutrient gelatin, peptone bouillon and _ peptone water. It is therefore conceivable that the enzyme is produced in the ordinary metabolism of the bacillus and regardless of the medium upon which it is cultivated. To test the accuracy of this hypothesis cultures were made in sterile egg albumin in water and also in gelatin containing a few inorganic salts. In the latter no liquefaction occurred and in the experiment with egg albumin no enzyme was found to be present (as measured by the thymol gelatin method), although when examined microscopically a number of the bacilli were found in a normal condition, and when cultured on nutrient gelatin rapidly liquefied the medium. It therefore appears that the presence of a peptone is essential to the production of the enzyme, which may be a combination of a substance secreted by the organism with a peptone. This enzyme thus formed then attacks the protein present, peptonising it and so converting it into a form suitable for the production of still more enzyme, thereby increasing the action.
If this is the case it might be expected that the higher the protein present the more enzyme would be produced provided the action could be started.
49 | J. M. DRUMMOND
The following results corroborate this, the figures given being the amount of liquefaction which occurred after 30 days.
(a) Peptone water 18 mm. (b) Peptone broth 44 ,, (c) Nutrient gelatin G.P.B. = 93°6 ,,
Bi verinont were made to determine whether the enzyme possessed a pepsin or trypsin character and also to ascertain whether coagulated egg albumin was attacked by the sterile filtrate obtained from a culture in egg albumin although it had no action upon gelatin.
The tubes described below were therefore made up, a strip of coagulated egg albumin being introduced into each.
1. 2cce. sterile filtrate from G.B.P. culture +3 ec. thymol water.
2. Ster= 35 % a - $206. = »» +1 ce. 1 J, sodium carbonate solution.
3. 2 ce. sterile filtrate from G.B.P. culture +2 ec. thymol water +1 ec. 0°2 °/, hydrochloric acid solution.
4. 4 cc. sterile filtrate from egg albumin culture +2 cc. thymol water.
5. 4c. ,, =e =, te RC oat »» +1 ce. 1°/, sodium carbo-
nate solution. 6. 4 cc. sterile filtrate from egg albumin culture +1 ce. themol water +1 cc. 0°2 °/, hydrochloric acid solution.
After several weeks considerable decomposition had occurred in 1, 2 and 3, and to a similar extent in each case, but no change in the albumin could be detected in 4, 5 or 6.
Therefore no enzyme is secreted when the organism is grown upon egg albumin and the enzyme from nutrient gelatin cultures can act in both acid and alkaline solutions. ;
A number of experiments were made in order to obtain data characteristic of the enzyme which could be used for purposes of identification. _
Thus the rate of liquefaction of gelatin by the enzyme (obtained from a culture in nutrient gelatin) was determined in presence of six typical acids, the amount of each present in an individual tube varying from 0:1 °/, to 1 °/, of the total contents.
Certain precautions were taken to ensure strictly comparable results and a number of curves were obtained.
It was found that the rate of liquefaction was inhibited by every acid under observation but that the retardation varied both with the acid present and the concentration. A rise in concentration of the foreign substance invariably produced a decrease in the rate at which liquefaction proceeded, but for a given rise in concentration the amount of retardation was not necessarily the same for every compound; e.g. an increase in concentration of.
J. M. DRUMMOND 43
acetic acid from 0:1°/, to 0°3°/, merely slowed down the rate whereas a similar increase in concentration of hydrochloric acid was sufficient to arrest liquefaction altogether.
That the constitution of the acid present produces an effect upon the rate of liquefaction is shown by the fact that acetic and lactic acids produced the least and an approximately equal decrease in the rate of action. Tartaric and citric occupy an intermediate position while oxalic and hydrochloric acids exercised the greatest action. In the observations under discussion 0°6 °/, of oxalic acid arrested all liquefaction while 0°2°/, of hydrochloric acid reduced the rate of liquefaction by 80 °/, and 0°3 °/, was entirely prohibitive.
It is probable also that the degree of dissociation of the acids has a considerable effect on the proteolytic activity of the enzyme, an increase in the concentration of the hydrogen ion exercising a prohibitive action.
This view is borne out by the position of the acids when placed in the order of the effect they have upon the rate of liquefaction, viz. hydrochloric, oxalic, citric, tartaric, acetic and lactic.
It is hoped that further work may be done on this subject, more especially with regard to the action of the bacillus upon blood serum and other albumins.
The author desires to thank Dr G. J. Fowler for his invaluable help and suggestions throughout the course of the work.
Vv. ACETYLCHOLINE, A NEW ACTIVE PRINCIPLE OF ERGOT.
By ARTHUR JAMES EWINS. From the Wellcome Physiological Research Laboratories, Herne Hill, SE.
(Received January 7th, 1914.)
Investigations carried out during the past few years have resulted in the isolation from ergot of several active principles, the presence of which adequately accounts for those actions of the drug which have been regarded as specially related to its therapeutic effects, each type of action having been advocated by one observer or another as a basis for its physiological standardisation.
The principles in question are (1) the alkaloid ergotoxine C;;H,0.N; (the hydroergotinine of Kraft [1906]) which was isolated by Barger and Carr [1907] and further investigated by Barger and Ewins [1910]; (2) p. hydroxyphenylethylamine, OH.C,H,.CH,.CH,.NH, [Barger 1909 ; Barger and Dale 1909], and (3) 8-iminazolylethylamine
HC: N a : ae - CH, -CH,- NH, [Barger and Dale 1910], the last two being amines derived respectively from the amino-acids tyrosine and histidine by decarboxylation. In addition, however, to those of apparent therapeutic importance, certain
other effects are shown in a more or less marked degree by all samples of the —
drug and by some with marked intensity. These effects have been familiar to the pharmacological workers in these laboratories for some years, but apparently have not been the subject of exact description or investigation. Conspicuous among these is an inhibitor effect on the heart, suggesting an intense though curiously evanescent muscarine action. Finding that the prominence of this effect in the action of different specimens of ergot ran closely parallel with their stimulant action on intestinal muscle, and that both effects were abolished by atropine, Dr Dale was led to suspect the presence in ergot of a principle producing both these effects. He was also able to
A. J. EWINS © 45
devise a convenient method for its physiological estimation by the use of a loop of rabbit’s intestine isolated according to Magnus’s method. When an opportunity recently occurred of obtaining an adequate supply of a preparation which exhibited this type of action with marked intensity, he suggested to me that an attempt should be made to identify the supposed new principle, and has followed the successive steps of its isolation with the physiological control. This paper deals with the chemical procedure by which the principle in question was isolated and identified. The details of its action will be described elsewhere by Dale.
The physiological effects described above resemble somewhat closely those described for muscarine, especially as the action in both cases is completely abolished by small amounts of atropine. On other grounds also it appeared not improbable that ergot might contain muscarine, since choline has long been recognised as one of its constituents [Brieger 1887], while Bohm [1885] had shown many years ago that fungi other than Amanita muscaria contain muscarine. Preliminary experiments carried out with a view to the isolation of the base, lent further support to this idea, since it was always found associated with the choline fraction. Thus it was completely precipit- ated from alcoholic solution, and partially precipitated from concentrated aqueous solution by mercuric chloride, and when the extract was frac- tionally precipitated by means of silver and baryta (Kutscher’s method) only the last (choline) fraction contained any of the active base with which we were concerned.
The actual isolation of the active base was finally accomplished by the method which is set out in detail in the experimental portion of this paper. In the first instance there was obtained only a very small quantity of a crystalline platinichloride; 2°8 milligrams in all. The base from this was found to be extremely active in producing the physiological effects with which we were concerned. Such chemical comparison as was possible with this minute quantity of material tended to support the identity of the unknown base with muscarine. The melting point, for example, was practically the same as that of synthetic muscarine platinichloride. A physiological comparison made by Dale with both natural and synthetic muscarines' showed however that the new base could not be either of these.
The clue to its identity was furnished by the observation, made repeatedly during the process of its isolation, that it was very susceptible to the action of alkali. In fact a dilute solution of the active base, if made distinctly alkaline with caustic soda at the ordinary temperature and neutralised again
1 For the former we are indebted to the kindness of Dr O. Rosenheim.
46 : A. J. EWINS
almost immediately, loses nearly all its original activity. This fact, and its constant association with choline, suggested that it might be a choline ester. This was the more probable since Hunt and Taveau [1911] had already described a number of choline derivatives, many of which were physiologically considerably more active than choline itself. Of these derivatives acetyl- choline had been indicated as one of the most active bases, and was also on general grounds the most likely to occur naturally. A small quantity of acetylcholine was therefore prepared by the method originally described by Nothnagel [1894]. Comparison of the physiological action of this base with that isolated from the extract of ergot showed that it was qualitatively and quantitatively the same. Further, by working up a larger quantity of the ergot extract, there was obtained sufficient of the crystalline platinichloride of the active base, to establish beyond doubt, by melting point and analysis, its identity as acetylcholine. |
Acetylcholine therefore exists in ergot and is the base responsible for the physiological action described at the commencement of this paper. That it occurs as such in the original ergot grains, is shown by the fact that a fresh extract made by boiling the drug with dilute alcohol, produces the effects shown by the extract prepared according to the directions of the British Pharmacopoeia. The presence of acetylcholine is consequently not due to fermentative or other changes taking place during the preparation © of the extract.
EXPERIMENTAL.
With the help of the physiological control above mentioned the following method was ultimately adopted and led to the isolation of the active base. .
The preparation available for investigation was a liquid extract, prepared according to the directions of the British Pharmacopoeia. 1600 cc. of this extract were concentrated under reduced pressure on the boiling water-bath to remove the alcohol. The residual syrupy liquid was diluted with water to 480 ec. and aqueous mercuric chloride added until no further precipitation occurred. The amount required was 1100 cc. of a saturated aqueous solution. The precipitate was filtered off, washed with water and, since it was found to be almost physiologically inactive, discarded. From the filtrate and washings the slight excess of mercury was removed as sulphide, and the excess of sulphuretted hydrogen by means of a current of air. The solution was then neutralised with sodium carbonate and concentrated on the boiling water-bath
A. J. EWINS 47
under reduced pressure to a thin syrup. This was poured into strong alcohol (92-95°/,) and allowed to stand for a few hours. The yummy precipitate was filtered off, washed with alcohol, and, being found to be practically inactive, discarded. The alcoholic filtrate and washings were taken to dryness and the residue dissolved in a small quantity of pure methyl alcohol, in which it was almost completely soluble. The methyl alcoholic solution was again pre- cipitated by the addition of four or five volumes of absolute alcohol. The precipitate was filtered off, the filtrate evaporated to dryness, and the residue completely extracted with smal] quantities of absolute ethyl alcohol. If necessary, precipitation by means of excess of alcohol was repeated, until the alcoholic solution obtained gave no further precipitate on addition of a large volume of alcohol. '
To the alcoholic solution (240 cc.) so obtained alcoholic mercuric chloride (560 ce.) was added until no further precipitate was produced. The mixture was allowed to stand over night. The precipitate, which contained practically the whole of the active base originally present in the alcoholic solution, was filtered, .washed with alcohol, and then extracted four times with boiling water, 150 cc. of water being used for each extraction. The portion of the mercury precipitate insoluble in hot water was found to be inactive and was therefore discarded. The hot aqueous extract on cooling deposited a further small quantity of precipitate which was very slightly active and was neglected. The clear, light yellow filtrate was then concentrated in vacuo on the water- bath. A precipitate, which was for the most part crystalline, soon commenced to separate. Concentration was continued until the volume was about 50 cc. The solution was then cooled and the precipitated mercuric chloride compound filtered off, when there was obtained 17 grams of a mixture consisting for the most part of the mercuric chloride compounds of choline and the active base. The salt was finely ground in a mortar, suspended in about 200 ce. of water, and decomposed by sulphuretted hydrogen. The mercuric sulphide was filtered off, re-suspended in water, and again treated with sulphuretted hydrogen, the process being repeated until decomposition was complete. The combined filtrates and washings were freed from sulphuretted hydrogen by a current of air, and the strongly acid solution treated with freshly precipitated silver carbonate until free from chlorine ions. The slight excess of silver in solution as carbonate was removed as sulphide and excess of sulphuretted hydrogen again removed by air. The slightly alkaline solution -was then neutralised with tartaric acid, and a further amount of the latter, equal to that required for neutralisation, was added. The solution was then taken to dryness in vacuo on the water-bath at 60°-70°, and the residue
48 : A. J. EWINS
completely extracted with absolute alcohol. The alcoholic solution was concentrated to small volume (15 cc.) and allowed to stand. After about 18 hours the acid tartrate of choline which had separated’ was — filtered off, and, when dry, weighed 0°3 gram. It was identified as choline by :— (a) the mercuric chloride compound m.p, 249-250", (b) the aurichloride m.p. 261-262", (c) the platinichloride m.p. 245°. An analysis of the aurichloride gave the following result : 0-1625 gave 0°0722 Au. _ Au=44-43 per cent. Calculated for C;H;,ONCl . AuCl,. Au= 44°47 per cent.
The alcoholic filtrate from the choline tartrate was next treated with an alcoholic solution of platinum chloride until no further precipitate was produced. The precipitated platinichlorides were filtered off and dried. The weight obtained was 1:3 grams. A small quantity of this platinichloride was decomposed by evaporating its aqueous solution to dryness with an excess of potassium chloride, extracting with absolute alcohol, evaporating’ off the alcohol, and dissolving the residue in a little water. This solution when tested physiologically was extremely active and it was found that practically the whole of the active base had been precipitated, as platinichloride. In order to separate the platinichloride of the active base from that of choline, which was still present, the main portion of the platinichlorides was treated with 2 ce. of boiling water in which it completely dissolved. The solution was then cooled to about 35°, when a small quantity of imperfectly formed polyhedra separated, which were quite different in appearance from the platinichloride of choline, which at the temperature indicated remained com- pletely in solution. The crystals were filtered off, washed with a little cold water and dried at 100°. There was thus obtained 0:205 gram of a platinichloride melting at 253-254°. On decomposing a few milligrams by the method already described, the solution of the free base obtained — was qualitatively and quantitatively indistinguishable in physiological - action from a specimen of acetylcholine prepared by Nothnagel’s method.
The platinichloride was recrystallised from a little hot water. It is considerably less soluble in water than choline platinichloride. On recrystal- lising, however, a certain amount of hydrolysis occurs, as was pointed out by Hunt and Taveau [1911]. On cooling there was obtained 0-060 gram of platinichloride, which in solubility, crystalline form, and melting point
? This method was employed by Honda [1911] for the separation of choline for muscarine.
A. J. EWINS 49
(256-257°) was identical with that of acetylcholine prepared from choline by the action of acetylchloride. Analysis gave the following results :—
0°0383 g. gave 0°0109 g. Pt. Pt=28°4 per cent. Calculated for acetylcholine platinichloride (C7;H,g02NCl)2PtCl,. Pt=27-8 per cent.
SUMMARY.
An active principle of ergot, recognisable by its inhibitor action on the heart and its stimulant action on intestinal muscle, has been identified as acetylcholine.
REFERENCES.
Barger (1909), J. Chem. Soc. 95, 1123.
—— and Carr (1907), J. Chem. Soc. 91, 337. — and Dale (1909), J. Physiol. 38, Proc. 1xxvii. —— —— (1910), J. Chem. Soc. 97, 2592.
—— and Ewins (1910), J. Chem. Soe. 97,. 284. Bohm (1885), Arch. exp. Path. Pharm. 19, 60. Brieger (1887), Zeitsch. physiol. Chem. 11, 184. Honda (1911), Arch. exp. Path. Pharm. 65, 454. Hunt and Taveau (1911), Hygienic Bulletin, Washington. No. 73. Kraft (1906), Arch. Pharm. 244, 336.
Nothnagel (1894), Arch. Pharm. 232, 266.
Bioch. vit : 4
VI. ON THE BEHAVIOUR OF TRYPSIN IN THE PRESENCE OF A SPECIFIC PRECIPITATE.
By AGNES ELLEN PORTER, Carnegie Research Scholar. From the Royal College of Physiciuns Laboratory, Edinburgh. (Received Dec. 17th, 1913.)
In a former paper on amylase [1913], 1 discussed the possibility of finding a substitute for complement in the recognition of antibody as well as the advantages which would follow if such a substitute were found.
Certain animal ferments are rather similar to complement in their general behaviour, each being a hydrolyst of a particular substrate by which it is first absorbed. Each is also readily taken up by different substances such as collodion and gelatin membranes [Porter 1909, 1910, Pribram 1910], or even inorganic precipitates [Hailer 1908, p. 283]. Pepsin absorbed into egg-white [Dauwe 1905, p. 426], or pepsin, ptyalin ete. into collodion [Porter 1910, p: 382] can be recovered in a small degree by a solution of the particular substance which they digest. Pepsin absorbed into collodion was however unable to act on a solid protein suspended in water [Porter 1910, p. 379]. As it is only by producing digestion that these ferments are recognised,— through the introduction of their particular substrate,—this possibility of partial recovery presents the chief difficulty in the use of ferments as sub- stitutes for complement. ;
The first attempt to absorb a ferment by means of a specific precipitate was made by Hailer [1908, p. 280]. His choice of rennet was unfortunate, as it loses hardly at all on dilution, being active at a dilution of 1 in 340,000 [Porter 1911, p. 394]. Still, though unsuccessful at stronger concentrations, his results do show with diluted rennet that some absorption had taken place. Ptyalin [Porter 1913, p. 601] displays only a very slight affinity for a precipi- tate of this nature, while taka diastase shows itself quite indifferent.
Ferments may be divided into two classes in respect to their behaviour with serum. In the one class are the proteolytic ferments which combine with serum proteins in the dissolved condition so firmly that they are inhibited by normal serum. On the other hand non-proteolytic ferments, the action of many of which is accelerated by serum, form much less stable compounds
A. E. PORTER 51
with precipitated serum proteins. In search of a suitable ferment, I treated a number of ferments by adding diluted serum and saturating with carbon - dioxide. Next morning the precipitate was removed, after centrifuging, and the upper fiuid tested for ferment. Emulsin and lipase were unaffected by this treatment, except that insoluble lipase is removed. On turning to proteolytic ferments, pepsin was found less influenced by the globulin precipitate than might have been expected, possibly because the acidity of the carbonic acid enabled it to produce some hydrolysis. A rennet was accidentally examined in the form of a mere trace in a sample of trypsin for which I have to thank Dr Cramer. After precipitation with carbon dioxide the upper fluid differed from the control in not being able to curdle milk at 56°, although it increased its viscosity figure.
Trypsin alone, of all the ferments which I have studied, is absolutely absorbed by a diluted serum precipitated with carbon dioxide, while it is only _ partly absorbed by the same dilution of the same serum before precipitation. If the upper fluid is left in contact with the precipitate only for one night, and then removed from it, it will be found by every test applied to it devoid
of trypsin. METHOD.
A trypsin solution was made by dissolving Armour’s pancreatin in water. The powder did not perfectly dissolve in the proportion of 1 gram to 100 cc., so that after standing for a night in the cold room, the upper fluid was separated off. The solution was’ kept in the cold room. Several methods were employed of testing tryptic action:
(1) by dissolving coagulated serum contained in Mett’s tubes, as - recommended by Dauwe [1905] in the case of pepsin,
(2) by dissolving fibrin,
(3) by forming hollows on serum plates, as recommended by Mueller
and Jochmann [1906],
.. (4) by rendering dilute caseinogen incapable of being precipitated by c 0°25 °/, acetic acid,
(5). by lowering the viscosity of caseinogen according to Golla [quoted from Eve 1910].
The most delicate and rapid method is undoubtedly the digestion of caseinogen. ‘This cannot be tested in the presence of serum by means of the ordinary method of slight acidulation, as the globulins of the serum are as easily precipitated by acid as is the caseinogen itself. The viscosity method of Golla, as described by Eve [1910], is very simple and accurate, very small
4—2
ps | A. E. PORTER
differences indeed being estimated with the help of a stop-watch. This method presents a very severe test of trypsin absorption because, as I have pointed out, an absorbed ferment is much more easily reached by a solution of its corresponding “substrate” (the specific substance on which it acts) than where this substrate is in solid form, suspended in a medium which is in contact with the absorbed ferment. The form of caseinogen used was 2, 3, and 4°/, casumen, a soluble preparation of caseinogen (that is to say easily soluble with heat and shaking), which was made up fresh on each occasion, Digestion was carried out at 56° to prevent bacterial action. The method of using serum tubes was found rather slow, but fibrin was a good deal used, being either digested in the presence of the serum precipitate or else treated with it and afterwards washed and transferred to an alkali. Digestion in this case was carried on at 37° under toluene. The method is somewhat rough so that fine differences are not distinguishable, but it is quite serviceable.
‘Trypsin in serum precipitated with CO,.
A. Examination of upper fluid alone. The following solutions were made up. (1) To 3 cc. ox serum were added 15 ce. trypsin solution and 12 cc. water (i.e. 50 /, _ trypsin, and 10 °/, serum). (2) A control trypsin solution 50 °/, in water. (3) N/10 NaOH containing 1-6 °/, NaCl, called NaCl—NaOH. (4) 10 °/, ox serum in NaCl—NaOH. (5) N/5 Na,CO,. (6) 10 °/, ox serum in N/5 Na,CO, (containing no NaCl as this rendered the serum too cloudy). No. 1, the mixture containing trypsin and serum, was saturated with CO., and left over night in the cold room. Next morning the upper fluid was separated from the precipitate after centrifuging.
This upper fluid was then tested for trypsin by the following methods:
With Fibrin. 5hrs. Ilday 2days (1) 0°75 ec. control trypsin + 0°75 ec. NaCl—NaOH be oa! — _ gone ie (2). "5, ari SR aae are serum—NaCl—Na0H <e — 0°5 “- i iegaet A ae Sere serum—Na,CO, ig eis 0°5 _— 0°75 i) eee absorbed trypsin upper fluid +0°75 cc. NaClI—NaOH_... _ 0 (5) ” ” ” aN ote Na,CO, ade por 0 _— 0
Fibrin in NaOH and Na,CO, also negative, except of course for swelling. With Mett’s Tubes (Serum).
Absorbed serum with Na,CO, or NaCl and Mett’s tubes was negative after a week, while control trypsin with serum under similar conditions was partly digested in a day or two.
With Serum Plates. 24 hrs. at 56°
(1)- Equal parts control trypsin + 10 °/, serum in NaCl—NaOH Faint hollow (2) i absorbed trypsin upper fluid + NaCl—NaOH Negative
A. E. PORTER © 53
With the Viscosity Test.
(Each contained 5 ce. 2°/, casumen; 10 hours at 56°.) Mins. Secs. (1) 0°75 ce. control trypsin + 0°75 ec. 10 °/, seram—NaCl—Na0OH 1 30 ae pa iy igs serum—Na,CO,_... 1 25 | ae sheorbed trypsin upper fiuid+0-75 cc. NaCl—NaOH... 1 46 (4) %9 » + 5, Na,CO; » = (5) 1-5 ce. H,O e a ses oo
With Caseinogen (tested by acid).
Serum containing globulins cannot be used with this test, so very dilute trypsin was substituted for trypsin serum.
(Each contained 2 cc. 0-1 °/, pure caseinogen, 24 hours at room temperature.)
Precipitate with 0-25 °/, acetic (1) 0°75 ce. 1-25 °/, trypsin +0-75 ec. Na,CO, =e 0 (2) ;; 50 °/, absorbed trypsin upper fluid +0-75 cc. . Na,CO, +
By the above tests it is clear that trypsin is only partially absorbed by 1/10 normal serum, but is wholly absorbed by the same strength of serum which has been precipitated by saturation with carbon dioxide. That is to say, if the precipitate bearing the trypsin is removed out of the serum the remaining fluid is found to be without trypsin.
It is however necessary to discover whether the absorbed ferment is so fixed that it cannot be recovered by the addition of more protein, especially in solution and in an alkaline medium.
B. Examination of upper fluid plus precipitate. The following mixtures were made, in water, and all precipitated by saturation with CO,.
(1) Ox serum 10 °/,, trypsin solution 50 °/).
(2) 29 5 by 29 23 39 (3) 9 2°5 ye 29 2? 9 (4) 7 1-25 "lo, ” > ”
(5) No serum Next day each mixture was shaken up, and the whole, containing the precipitated globulin, was tested for trypsin by the following methods: With Fibrin. (In this case the alkali was added to the precipitate before the fibrin.) Time, over night at 37°.
Result
(1) 0°75 cc. control trypsin + 0°75 ce. 10 °/, serum in NagCO; 0°5 dissolved (2) 2? 29 29 + 2? 5 "lo 2? 9 all 29
(3) 2”? LE) 9 + ”? 2°5 "lo ” 9 2° ”
(4) ”? 29 ? a ”? 1°25 *lo 2s 2 ”? 29
(5) 23 ” + 29 Na,CO, 29 2
(6) - Sinorbud trypsin (10 °/, serum) + 0°75 ec. NagCO; Slightly ,,
(7) 2? ”? 2 (5 (5 Jo ”? }-+ 2? > Almost ”
(8) » (25% 5, )t+ » » Dissolved (9) 9 99 29 (1°25 "ly 3° )+ 29 2”? >?
(10) ” H,0 — ” 2 0
54 | 4. BE PORTER
If the fibrin be added to the precipitated serum mixture while it is neutral, and after an hour removed, washed and placed in an alkaline solution,
less of the ferment is recovered.
Next day (1) 0°75 ce. control trypsin + 0°75 cc. 10 °/, serum in NaCl 0°5 dissolved (2), absorbed ,,. + 4, 16% NaCl 0
With the Viscosity Test. (Each +5 ce. 3 °/, casumen, containing 20 °/) N/5 Na,CO,.) Time 4 hours at 56°, then room temperature for 24 hours.
Hours ee Ne ae 24 _ min, secs, min. secs,
(1) 0°75 cc. control trypsin+0°75 ce. 10 o/, serum in Na,CO, 1 42 1 434 (2) ” 9 shh ee ‘3 1 36 1 36) Be ee ae a ee ‘ Te ee (4) ” ” ” + ” 1°25 "lo ” oe) 1 32 1 32 = (5) 9 ” ” i ” Na,CO, 1 30 1 27 | (6) - absorbed trypsin (10 °/, serum) + 0°75 cc. Na,CO, gee tae | 8 (7) 99 ”? ” (5 "lo ” a+ ” ” 1 47 1 45 . ae SY, RE et oa 1 6.12 (9) ” yo) (1865 yy RS % 1 40 1 37/8 (10) ” H,0O Bay 5 1 56 )
The above experiments show clearly that some trypsin can free itself from the serum precipitate and enter the casumen or fibrin. Still this tendency to recovery is not sufficient to obliterate the difference of the effects of original and precipitated serum upon trypsin, which is still marked enough to permit an expectation that a similar difference may be found in the case of a specific precipitate.
Trypsin with a specific precipitate.
In this case more protein was present,—in the form of antigen,—than in the last, where the influence of serum alone was under investigation. The antigen used was egg-white. The extent to which egg-white absorbs trypsin — was next investigated, both with fibrin and by the viscosity test. To 0°15 ce. normal ox serum, or 0°15 ce. saline solution, 0°15 ce. egg-white, in dilutions of from 1/10 upward, was added, followed by 0°75 cc. trypsin solution and 0°75 ce. N/5 Na,CO,, and after an hour’s contact, fibrin, or 5 cc. 3 °/, casumen, At these dilutions egg-white exerted no visible influence upon the results, other than by its presence diluting the serum by one-half, and so lessening its inhibitive effect.
Serum was obtained from rabbits which had been treated as described on p. 55, and used in the following condition:
A. E. PORTER 55
(1) Thrice injected, precipitation at 1/10,000 (absorption of trypsin positive). (2) Thrice injected, precipitation negative (absorption negative). (3) Once injected, precipitation at 1/1000.
Twice injected, precipitation not tested higher.
Four times injected, precipitation at ais. 000 (absorption in all sinilsoi: (4) Normal (absorption negative). (5) Normal (absorption negative). (6) Normal (absorption negative). (7) Normal (absorption negative).
Twice injected, precipitation at 1/5000.
Thrice injected, precipitation at 1/5000, not tented higher Sikiseption in both positive). (8) Normal (absorption ws cra
Eaperiment I.
Rabbit serum 1, to each tube added 0°75 cc. trypsin solution + 0- 75 ce. NasCO, N/5, after 0-5 hour 5 ce. 3 °/, casumen. Time 2 hours at 56°.
'Viaoopity
min. secs. (1) 0: 15 ce. Immune serum 1+0: 15 ee, 1/10 egg-white te Oe ‘ (2) 29 2” 9 ae 99- 1/100 2? 1 46 (3) ” ” ” 53 ” : 1/250 ” 1 46 (4) ” ” ” = ” . 1/500 ” 1 46 (5) 9 ” oe 1 ” 1/1000 ” 1 40 (6) ” ” 4 a ” 1/2500 ” 1 40 (7) ” ” ” + ” 1/5000 ” 1 40 (8) 29 Sos ” B.A ) NaCl 1 40
(All later viscosity experiments from this point were tested with a stop-watch.)
Enperiment fs I.
Rabbit 3 injected once, positive, Rabbit 2 injected ‘ie eieiativie and Rabbit 4 normal. To each tube added 0°75 cc. trypsin solution, after 3 hours added 5 ce. 3 °/) easumen and 0°75 cc. N/5 Na,CO;. Digestion at 56° 0°75 hour, then 0° over night.
; Viscosity ie ~ min. secs. (1) 0°15 ce. Immune serum 3+0°15 ec. 1/10 egg-white 1. 32°5-
(2), ” 16 BR) Sg SOD 4, 1 31 (3) 9 ” 9 8+ et 1/250 +> Lt. 381
oO 93 op Be dg 2 DN 1 305
(5) ” ” ” 2+ ” 1/10 oe) 1 31°5
i (6) 2° Foie 3 oe 2+ 23 1/100 ” 1 31°5
(7) 5 v » 2+ 4, 1/250 ,, 1 Bl..
(8) 3° 9 99 2+ 33 NaCl 1 31°5
(9) ey Normal 2? 4+ ” 1/10 ” 1 29°5 (10), 9 so) Ay ge FO: 1 29 (11) ,, “99 Sate BAS as EPO 1 29 1 29
(12) ” ” ” 4+ ” NaCl
Experiment #1.
Rabbit 3 twice injected, Rabbits 5, 6, 7 and 8, all scien’: Each tube received 0-75'cc. trypsin solution, after half an hour’s contact 5 cc. 4 °/, casumen and 0°5 cc. NagCO; were added. Digestion at 56° for 4 hour, then cold room till next day.
56
Rabbit 7, twice injected. Each tube received 0°75 cc. trypsin solution, after } hour 5 ce. 3 % casumen and 0°75 ec. N/5 NagCO,.
Comparison of some rabbits’ sera with fibrin. Immune sera 3 and 7, both twice injected, normal sera, 4, 5,6 and 7. Each tube received 0°75 cc. trypsin solution, after a night’s contact in the cold room, fibrin was added to each mixture next morning, and after half an hour removed, washed, and placed in N/5 NagCOz; at 37° over night.
A. E. PORTER
(1) 0°15 cc. Immune serum 3+0°15 ec. 1/10 egg-white
(2) (3) (4) (5) (6) (7) (8) (9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
(21)
_Q)
(2) (3) (4) (5) (6)
33
3+ 3+ 3+ 3+ 5+
”
1/100 1/500 1/1000 NaCl 1/10 1/100 1/500 NaCl 1/10 1/100 1/500 NaCl 1/10 1/100 1/500 NaCl 1/10 1/100 1/500 NaCl
Experiment IV.
>
7
7
99
Digestion at 56° for 3 hours.
0°15 cc. Immune serum 7+0-15 cc. 1/10 egg-white
”
”?
7+ 7+ T+ T+ T+
1/500 1/1000 1/5000 NaCl
39
Experiment V.
(1) 0°15 cc. Immune serum 340-15 ce. 1/10 egg-white
(2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
Normal
”? Immune
hs]
”
3+ 4+ 4+ 5+ 5+ 6+ 6+ T+ 7+ T+ T+
NaCl 1/10 NaCl 1/10 NaCl 1/10 NaCl 1/10 NaCl 1/10 NaCl
9?
>
>
Days Frayed Almost gone Gone Gone
Frayed ve
%9 9
e Almost gone
oe Gone
A Frayed
7? 3°
9 39
Untouched Untouched
Frayed Frayed -
Viscosity min. secs,
Kee ee ee PEP ee Pe eP eEP PRP PPP ee
40
Viscosity min. secs. 1 382 1 32 1 382 1 31% 1 = 28°5 1 28°5
A. E. PORTER a,
Experiment VI.
Immune sera 3 and 7, and normal serum 7, were tested with fibrin and casumen. Rabbit 3 had been injected four times, Rabbit 7 thrice.
With Fibrin.
Hitherto the trypsin had been added to a mixture of serum and egg-white. Here 0°75 ce. trypsin solution was added to serum, and the egg-white added immediately afterwards. After three hours of contact fibrin was immersed for 0°75 hour and then washed and placed in N/5 NasCO3, covered with toluene, at 37° over night.
Days se _ (1) 0:15 cc. Immune serum 3+0-15 cc. 1/10 egg-white Whole Whole (2) ” ” wat sg YI ,, Gone Gone (3) ” 7” ” 3+ 37 1/500 37 3? 33 (4) ” 7 ” 3+ ” NaCl 3? ” (5) » ” Be ees es BRO, Whole Whole (6) a me Sane eee Fs! Laer 0-5 gone Gone (7) ” ” ” 7+ ” 1/500 9 9 9 (8) ” ” ” T+ ” NaCl Gone ” (9) $8 Normal Pd eat 1/10 ©. Almost gone ai (10) ” ” RSE ce So ie eeamagess Si! | ae Gone ee (1) %s » 4: «> Nal Almost gone za
With Casumen.
To 0°15 cc. of serum were added 0-75 cc. trypsin solution and 0°15 cc. of egg-white. © After three hours 5 ec. 3 °/, casumen and 0°75 cc. N/5 NagCO3 were added. Digestion at 56° for 0°75 hour, then all were transferred to the cold room for the night, and their viscosity determined next morning.
Min. Sees. (1) 0°15 cc. Immune serum 3+0°15 ce. 1/10 egg-white 130 (2) ” ” 3 «6 O+ 7s 501 gees dca (3) 9 ee Bey OUT, 1 39 (4) ” ” ” 3+ ” NaCl 1 28 (5) ” ” ” T+ ” 1/10 7 1 32-2 (6) ” ” ” T+ ” 1/100 ” 1 3174 (7) ” ” 39 7 e ” 1/500 ” 1 29 @) ts Sao hy 2 a eeS T8 | 1 28 (9) 8 Normal Seder eats 1/10 tS ts 3g (10) ” ” ” 7 4 or) 1/100 ” 1 32-4 (il) 0 ee ae). ROL 1 322
The above results show clearly that trypsin can be inhibited by a specific precipitate in a greater degree than by immune serum alone or by normal serum with antigen. This result is, however, only relative. Trypsin is partly inhibited by untreated serum, and it is not wholly inactivated by precipitated serum. Judging from the analogy of the carbonic acid precipi- tate where trypsin is wholly absorbed, but is in part recoverable, a total inhibition in the presence of a specific precipitate could hardly have been expected. A similar recovery of trypsin—from the antitrypsin of Weinland— has been described by Dastre and Stassano [1903]. Time, which would favour fixation, or inactivation, of absorbed trypsin through aeration [Levy 1905],
58 A. E. PORTER
also favours the recovery of trypsin from antitrypsin (Dastre and Stassano), a tendency which can also be seen in the above results. The recovery of trypsin is therefore an unfavourable factor which cannot be entirely avoided. It can be diminished by using a solid protein as a test for the ferment, especially in a neutral or slightly acid medium, to avoid tryptic action, as any such action would produce a soluble protein which could combine with the absorbed trypsin. After contact with the serum-trypsin mixture, the solid protein would be removed from the sphere of action of the precipitate into an alkaline medium suitable for promoting the action of any trypsin which it might have taken up. If such precautions are taken—and even without these precautions—trypsin-absorption might well be used in the recognition of antibody. At least, as far as my experience goes, trypsin is the most suitable ferment for the purpose.
CONCLUSIONS.
Trypsin is only partially inhibited in the presence of 10 °/, ox serum.
It is however completely bound by the precipitate produced in 10°/, ox serum by saturation with carbon dioxide and if this precipitate be removed the trypsin is completely removed with it.
The absorbed ferment is partly recoverable from the precipitate by protein, especially in solution.
This recovery is not however sufficient to obliterate the difference between the inhibition exerted by 10°/, untreated ox serum and that by 10°/, precipi- tated ox serum.
In the presence of a specific precipitate (immune rabbit serum and egg- white) trypsin is more inhibited than by immune serum alone or by normal serum and egg-white. This is especially the case where free trypsin is tested for by a solid protein, but also occurs if the test is carried out with protein in solution.
I wish to thank Professor Ritchie for his kindly interest and advice.
REFERENCES.
Dastre and Stassano (1903), Compt. rend. Soc. Biol. 55, 633. Dauwe (1905), Beitriige, 6, 426.
Eve (1910), Brit. Med. Journ. 1, 1540.
Hailer (1908), Arb. kaiserl. Gesundh. 20, 27.
Levy (1905), J. Infect. Diseases, 2, 1.
Mueller and Jochmann (1906), Miinch. Med. Woch. 1393. Porter (1909), Thesis, Edinburgh University.
—— (1910), Quart. J. Exp. Physiol. 3, 375.
—— (1913), Biochem. J. 7, 599.
Pribram (1910), Wien. klin: Woch. 1181.
=) gee Se My a:
VIl. THE VISCOSITY OF SOME PROTEIN SOLUTIONS.
_ By HARRIETTE CHICK ann: EVA LUBRZYNSKA. From the Lister Institute. | " (2 figures.) (Received December 24th, 1913.)
In the following work the same general methods were employed as in that already published by Martin and Chick [1912] upon the viscosity of solutions of caseinogen.
In the present instance an investigation was made of the influence of (1) concentration of protein, and (2) temperature, upon the viscosity of solutions of pure crystallised egg- and (horse) serum-albumin. For purposes of comparison, a few experiments were also made with whole serum, in some cases concentrated in vacuo in order to yield material of high protein content. The viscosity, as in the previous work, was determined by measuring the time of flow in an Ostwald viscosimeter, the figures obtained being relative to the time taken under similar conditions by pure water, which is expressed by unity. The concentrations throughout are expressed as grams per 100 grams of solution.
EGG-ALBUMIN.
The crystals of egg-albumin were obtained from egg-white by the method of Hopkins and Pinkus [1898]. The material was twice recrystallised, and then dialysed for two or three weeks against distilled water, till it contained an insignificant trace of ammonium sulphate. As considerable dilution took place during dialysis, it was necessary to concentrate the material in order to obtain the high concentrations necessary for the experiments. This was done by allowing the solution to evaporate at room temperature in vacuo over sulphuric acid, material being thus obtained which contained 28°/, by wéight of egg-albumin. The protein-content in this and other cases was determined by boiling a weighed quantity of the solution, after dilution and acidification, and weighing the coagulum on a weighed filter, after drying at 110°.
60 HL CHICK AND E. LUBRZYNSKA
TABLE I.
Influence of concentration of the protein upon the viscosity of solutions of pure egg-albumin (crystallised).
Temperature 25-2°C. Time of flow in viscosimeter for water 47°8 seconds.
Concentration Mean time of Density Coefficient of protein, flow in viscosi- of the solution of viscosity
% meter, secs. (H20 at 25° C.=1) (H,0 =1) 28°15 441°8 1:0805 9°99 26°83 367°4 1°0775 8-30 24°33 257°2 1:0693 5°81 20°12 159°2 10566 3°60 14°53 97°9 1-0402 2°21 8°877 69°4 1-0242 1°57 3°016 53°8 _ 10083 1:22
Influence of concentration of Protein. In Table I are given the details of an experiment showing the influence of concentration upon the viscosity coefficient in the case of this protein. The same results are shown graphically in curve (a) of Fig. 1. Solutions containing different proportions by weight of protein were obtained by dilution of the concentrated material described above. In order to apply the necessary correction in calculating the coefficient of viscosity, the density of the. solutions was directly determined by weighing a known volume in a pyknometer.
In the case of the weaker solutions, the viscosity remained low, increasing but slightly with increasing concentration of protein and only reaching a value equal to twice that of distilled water at a concentration of protein equal to about 13°/,. This is well seen in the curve in Fig. 1, where, up to a concentration of about 9 °/,,the slope is very slight, in fact the relation between viscosity and concentration of protein approximates to that obtaining in solutions of non-colloidal material, and might satisfactorily be expressed by a straight line. At higher concentrations the curvature becomes increasingly greater until, at a concentration of 28°/, protein, the viscosity reaches the figure of nearly 10. :
The curve expressing the relation between protein concentration and viscosity in case of solutions of egg-albumin is of the same general type as that obtained for caseinogen [Chick and Martin 1912]; in the case of the latter, however, a comparatively low concentration (7 to 8°/,) produced a viscosity equal to the value obtained with 28°/, egg-albumin. According to the hypothesis of Hatschek [1910, 1911, 1912] the difference between these two proteins in regard to their viscosity should be attributed to the fact that in a solution of egg-albumin much less water is appropriated
H. CHICK AND E. LUBRZYNSKA 61
YD 28
26
3)
19 14 Protein-content, °/, (by weight)
10
(\)
4 1 4 rn 4
a 2) sd "oO 0 wv i] N ei
Viscosity (H,O=1)
Fig. 1. Influence of protein-concentration upon the viscosity of solutions of various proteins.
Curve (a) —O— = pure egg-albumin. >» (6) —@®— = pure serum-albumin (horse). >» (c) —t whole serum, at protein-concentrations above and below the normal ; original serum contained 7-3 °/, protein.
62 -H. CHICK AND E. LUBRZYNSKA
by the protein phase than is the case with caseinogen. Consequently, in the former case, much higher concentration of protein is necessary to yield the high viscosity ‘characteristic of the condition in which the aggregates of the disperse phase are approaching contact with one another. In a later communication it is proposed to discuss this theory as regards both egg-albumin and other proteins in greater detail.
Influence of Temperature. The variation of temperature was necessarily limited, the maximum range being from 0° to about 40°. Higher temperatures could not be employed without canger of the protein becoming “denaturated.”
In order to make the necessary comparison with the time of flow taken by water, experiments were made over the. same range of temperature with an equal total volume of distilled water in each of the two viscosimeters employed in Exp. 1 and Exps. 2, 3 and 4 respectively. Smoothed curves
TABLE II.
Influence of temperature upon the viscosity of solutions of pure egg-albumin (crystallised) of varying concentration.
Mean time of flow in viscosimeter, seconds Coefficient of A
Protein Density at cr ~ viscosity (H,0 Exp. content, 25°C. (H,O0 at Tempera- Albumin Distilled water at the same No. *, 25° C.=1) ture, °C. solution (from curve) temperature=1) 1 7°04 1-0192 2°8 136°8 101°8 1°37 8°3 115-0 86-0 1°36 15°2 95:0 70°3 1:38 25-0 74:2 555 1-36 32-3 63-0 47°9 1:34 42-1 _§2°1 40°5 1°31 2 14°6 1:0404 2°8 179-2 86°2 2°16 86 147°4 72:1 2°13 14°7 123°4 61°5 2°09 15°1 ; 123°0 61-0 2°10 25:0 94-0 48-0 2°04 33°1 79°5 41:4 2°00 42°9 64:2 345 1:94 3 2071 10566 0 368-3 96°4 4-04 8-0 269°7 73°3 3°89 17:0 202°3 58:0 3°68 25-4 16174 47°6 3°58 33°0 132-1 41:4 3°37 (4176 sy. 105 B54 3°27 4 28-15 1-0805 0°6 _ 1170-2 . 93°8 13°48. 78 S25'C =, ae re 12°13 15°6 614:1 -.. 60:1 soft) Toe 25°4 440°9 . 47°6 10°01 33°9 344°9_ 40°8 9°13
41-9 283°7 35°1 8°73
H. CHICK AND E. LUBRZYNSKA 63
were drawn through the experimental points and from these the time of flow corresponding to any intermediate temperature could be read off. The values given in the 5th column of Table II were obtained in this way and, with the help of these, the coefficients of viscosity in the 7th column were calculated.
In case of the weaker solutions (from 7°/, up to 20°/,) the influence of temperature upon the viscosity was comparatively trivial, that is to say, the phenomenon was of the same order as that obtaining in the case of distilled water. A 7°/, solution of egg-albumin, with viscosity of 1°3, behaved- on heating as pure water containing a crystalloid in solution. Even with a 20 °/, solution, the viscosity relative to water was only about 20°/, greater
155
Viscosity (H,O at the same temperature =1)
5 10 15 20 25 30 35 40 45 Temperature, degrees Centigrade Fig. 2. Influence of temperature upon the viscosity. of solutions of various proteins with differ- ing protein-content. —O— = egg-albumin: curve a, protein content= 7:0 °/,.
> b, > 2 =14°6 ,, ” c, oe oy =20°1 ” 7 d, . 2 =28°1 ” —e— = serum protein: curve e, ae or: ae Rr St —+t— = whole serum: curve f, ,, ne TO:
2? 9; 99 ” -=18-1 ”
64 -H. CHICK AND E. LUBRZYNSKA
at 0° than at 41°6°, and not until the concentration of albumin reached the high figure of 28°/,, was temperature found to have any specially marked influence.
These facts are well expressed in the four curves a, b, c and d in Fig. 2, where viscosity (relative to water at the same temperature) is plotted as ordinate against temperature as abscissa.
Influence of Ammonium Sulphate upon Viscosity of Crystallised Hgg- albumin. It has been shown [Chick and Martin 1913, 2; Spiro 1904] that “ precipitation ” of egg-albumin by ammonium sulphate is a phase-separation and that a definite proportion of both salt and water is associated with the egg-albumin in the “ precipitate.” Arguing from analogy, it is exceedingly probable that crystallisation of proteins from strong solutions of ammonium sulphate is of a similar character. —
In order to ascertain whether such hypothetical association with ammonium sulphate had any influence on the viscosity of solutions, strictly comparable experiments were made, both before and after dialysis. Dabrowski [1912], on the basis of a comparison between the different rates of diffusion obtained with the two kinds of material, came to the conclusion that aggregates formed in solution by undialysed crystals of egg-albumin were much smaller, nearly one-sixth the size of those of dialysed albumin. It therefore seemed possible that such a fundamental difference in the type of solution might also be expressed by some significant change in viscosity.
The results obtained were, however, entirely negative (see Table III). | Solutions were made of exactly similar protein content, 6°98°/,, in the one
: TABLE III.
Viscosity of crystallised egg-albumin.
(a2) Before, and (b) after dialysis.
Mean time of flow in
viscosimeter, seconds Coefficient - sig ~ of viscosit : : 3°51°/)(NH4)o80, ((NH4)2S0, solu- Protein Ammonium solution in tion in exp. (a) content, sulphate, Tempera- Albumin exp. (a) and and H,0 in Exp. lo 7, ture, °C. solution H,Oinexp.(b) exp. (b)=1) (a) 6-981 3°51 0-6 145-9 109°9 1°35 2 99 25°5 75°9 58°8 1°31 ” %9 45:1 51°4 39°8 1°32. (b) 6-981 a 1-0 145-4 109-0 1:36 ” — 25°4 74°5 57°0 1°33 > — 45:1 49°2 38°3 1:31
* Density of the solution taken as 1-019 for the purpose of calculating viscosity coefficient.
H. CHICK AND E. LUBRZYNSKA 65
case from undialysed crystals, with their ammonium sulphate (= 3°51 °/,), and in the other from dialysed material. In the former case viscosity was deter- mined relative to pure water and in the latter case in relation to a 3°51 °/, solution of ammonium sulphate. Comparison was instituted at three different temperatures. ‘Temperature was not found to have any unusual influence on viscosity, nor was any difference traced between the two types of solution. At the same time it must be admitted that the concentration of protein employed (7 °/,) was rather low, and it would be well to repeat the experiment with stronger solutions.
SERUM-ALBUMIN (HORSE).
A series of experiments similar to those with egg-albumin were also carried out with pure serum-albumin. The protein in this case was also crystallised in presence of ammonium sulphate according to the method of Hopkins and Pinkus [1898] and, after recrystallisation, dialysed against distilled water in presence of toluene for some weeks. A solution was obtained, containing 20°65 °/, protein and traces only of ammonium sulphate.
Influence of Concentration of Protein. The density of serum-albumin solutions has been shown to bear a linear relation to the concentration of protein [Chick and Martin 1913, 1]. In order, therefore, to apply the appropriate correction in calculating the coefficient of viscosity it was only necessary to make direct determinations in a few cases. The straight line could then be drawn expressing the relation of density to protein content and the other values of density required be obtained by interpolation; such values are marked with an asterisk in Table IV.
TABLE IV.
Influence of concentration of the protein upon the viscosity of pure serum-albumin (crystallised).
Temperature 25:-4° C. Time of flow in viscosimeter for water 55 seconds.
Concentration Mean time of Density Coefficient of protein, flow in viscosi- of the solution of viscosity, %ly meter, secs. (H,0 at 25° C.=1) H,O=1 20°65 391-4 10593 7°54 19°24 310-2 1-0549* 5°95 17°85 249°2 1-0509* 4°76 14°54 159°8 1-0412 3-02 10°45 104°3 1-0296 1°95 5°19 71:3 10153 1°32 2-59 61-6 1-0075 1:13
* Interpolated values. Bioch, vi 5
66 AL CHICK AND E. LUBRZYNSKA
The experiments showing the influence of protein content on viscosity are detailed in Table IV and graphically expressed in Fig. 1, curve (b), where viscosity as ordinate is plotted against concentration of protein as abscissa.
The results show great similarity with those obtained for egg-albumin. In Fig. 1, curves (a) and () run closely together at first. As the concentra- tion of protein increases, however, it is seen that serum-albumin has a much higher viscosity than egg-albumin. |
Influence of Temperature. The influence of temperature was investigated in one experiment only, of which details are given in Table V, and graphically expressed in curve (e) Fig. 2. This experiment was made with the most concentrated solution available, containing 20°1°/, protein by weight. The viscosity relative to water was not only much greater than that of egg-albumin of similar strength (see Exp. 3, Table II), but showed a much greater change with alteration of temperature, viz. from 7°3 at 43° to 103 at 1‘6° (compare Table V with Exp. 3, Table II).
TABLE V.
Influence of temperature upon the viscosity of pure serum- albumin (crystallised).
Protein-content= 20°65 °/,. - Density=1-0593 (at 25° C.).
Mean time of flow in viscosimeter, seconds Coefficient of r ste ~. viscosity Temperature, Albumin Distilled water (H,O at the same °C. solution (from curve) temperature = 1) 16 1026-0 105-6 . 10:29 — 3°95 906 °2 98-2 “ 9°77 8°8 713-2 84-7 8-92 14:8 553°8 71:0 8:26 25°4 391-4 54°8 7:56 33°4 325°8 47°0 : 7°34 43°0 : 2751 40-0 7°28
_ WHOLE SERUM (HORSE).
The sample of horse-serum selected for these experiments contained 7 of, total protein. In order to obtain solutions of protein-content comparable to those employed above, some concentration of the serum proteins was necessary. By placing in shallow dishes for 48 hours in vacuo at room temperature material was readily prepared, which contained 18°/, protein. The consti- tuents of the serum did not appear to have been affected by the process, for, after diluting the concentrated material to obtain a solution whose protein-content was equal to that of the original serum, the viscosity was also
found to be the same as that previously determined (see Table V1).
H. CHICK AND E. LUBRZYNSKA 67
Influence of Concentration of Protein and of Temperature. The proportion of salt contained in serum was found to have an insignificant influence upon both the density and the viscosity of the system; hence the values of the coefficient of viscosity were calculated in relation to pure water as before. In Table VI are given the results of a series of experiments in which the protein- content varied from 1°8°/, to 18 °/, by weight.
In obtaining values for the density of the various solutions, the same method was adopted as in the case of serum-albumin; after a direct determina- tion had been made in case of one or two solutions, a curve was constructed from which intermediate values could be obtained. Such interpolated values are marked with an asterisk in Table VI.
TABLE VI.
Viscosity of horse-serum at concentrations above and below the normal ; the normal serum contained 7:3 °/, protein.
Temperature 25°C. Time of flow in viscosimeter for water 55-5 seconds.
Concentration Time taken for Density Coefficient
of protein, flow in viscosi- of the solution of viscosity
PS meter, secs. (H20 at 25° C,=1) (H,0 =1) 1-836 62°4 1:0069* 1°13 3°665 70°3 1-0143 1-28 5°134 79°4 1°0193* 1°46 5°84 82°3 1-0220* 1-52 PY 7°31! 92°8 10276 1°72 * 7-33? 92°5 * 1:0276 1-71 9°13 110°2 1-0344* 2°05 10°89 131°3 1-0405 2°46 12°67 156°6 1-0477* 2°96 14°71 206-2 1:0553* 3°92 16°27 253°6 1:0617 4°85 18-10 331°5 1:0679* 6°38 1 Normal serum. 2 Prepared by dilution from concentrated serum.
* Interpolated and extrapolated values.
From Table VI, and more readily from curve (c) Fig. 1, it is seen that for low concentrations of total protein the viscosity of whole serum also remained low, varying from about 1°10 to 1:7 as the concentration of total protein was raised from 1°80 °/, to that of the normal serum (7°3°/,). As the serum was concentrated, however, the viscosity increased rapidly with increasing concentration of total protein, and when it possessed a little more than twice the normal protein-content the viscosity relative to water reached
the value of 5:0. 5—2
68 H. CHICK AND E. LUBRZYNSKA
In comparison with serum- or egg-albumin the total proteins of serum 9
yield much higher viscosity when in solution. For example, while 18 %/, solutions of serum- and egg-albumin gave coefficients of viscosity equal to 49 and 2:8 respectively (see curves (b) and (a), Fig. 1), concentrated whole serum containing the same proportion of total protein had a viscosity of 6°38.
The influence of temperature upon viscosity was also greater with the proteins of whole serum. With the concentrated material, containing 18 °/, protein, the coefficient of viscosity fell 34°/, (from 8°63 to 5:7) with ‘increase in temperature of about 40° (see Table VII). A slightly greater rise of temperature in case of solutions containing 20°/, serum- and egg-albumin produced a decrease in viscosity coefficient of only about 29°/, and 19 °/, respectively (see Tables V and II).
TABLE VIL.
Influence of temperature wpon the viscosity of whole serum (horse), (a) normal serum, (b) concentrated serum.
Mean time of flow in
viscosimeter, seconds Coefficient of Protein Density at ” viscosity (H,O content, 25°C.(H,0O at Tempera- Distilled water at the same Exp. °%lo 25° C.=1) ture, °C. Serum (fromcurve) temperature=1)
(a) 7°68 1-0290 0 218°2 1116 2-01 ae aa 8-0 164°9 86°7 1°96 39 a4 17°0 125°5 67°1 1-92 Fa ae 254 99°6 54°9 1:87 ” “ 33°0 83°6 47°3 1:82 és ae 41-6 70°4 40°9 1:77 (b) 181 10679 0-2 897-2 111°0 8°63 » % 8-2 581°5 86-0 7°22 ce) ” 14:3 462°3 71:9 6°86 » 25-0 328-9 55-4 6°34 » ae 32-2 270°3 48-0 6°01 ” i 39°8 224-6 42:1 5°70
It is evident that one or more of the remaining proteins in horse serum must, when in solution, possess a viscosity much higher than that of the serum-albumin, and the results of some preliminary experiments suggest that the small proportion of “ euglobulin ” contained in the serum exercises a disproportionate influence upon the viscosity of the whole.
In a second communication it is proposed to publish the results of experiments, similar to the above, carried out with purified samples of
H. CHICK AND E. LUBRZYNSKA 69
4
“euglobulin” and-“pseudoglobulin,” prepared from horse serum, and to & . — ; ; ; y ; include a general discussion of the theoretical bearing of the whole series
of data.
REFERENCES.
Chick and Martin (1912), Zeitsch. Chem. Ind. Kolloide, 11, 102. — — (1913, 1), Biochem. J. 7, 92.
— — (1913, 2), Biochem. J. 7, 380.
Dabrowski (1912), Bull. Akad. Sci. Cracow, A, June, 485. Hatschek (1910), Zeitsch. Chem. Ind. Kolloide, 7, 301.
—— (1911), Zeitsch. Chem. Ind. Kolloide, 8, 34.
(1912), Zeitsch, Chem. Ind. Kolloide, 11, 284.
Hopkins and Pinkus (1898), J. Physiol. 23, 130.
Spiro (1904), Beitrége, 4, 300.
VIII THE QUANTITATIVE ESTIMATION OF UREA, AND INDIRECTLY OF ALLANTOIN, IN URINE BY MEANS OF UREASE.
By ROBERT HENRY ADERS PLIMMER anp RUTH FILBY SKELTON.
From the Ludwig Mond Research Laboratory for Biological Chemistry, Institute of Physiology, University College, London.
(Received January 2nd, 1914.)
For the rapid routine analysis of urea in urine Folin’s magnesium chloride method [1901; 1902, 1; 1903; 1905] is probably the one most usually adopted, but it is not an easy method to acquire and it requires a considerable amount of attention. A good substitute might soon displace it. Folin himself has devised other methods for the estimation of urea in small quantities. We believe we have found a substitute for the first method by using the urease of the soy-bean [Takeuchi, 1909]. Marshall [1913, 1] has already published a method for the rapid clinical estimation of urea in urine by this means, but by his procedure the results are not sufficiently accurate, being only within about two per cent.; an extract of the soy-bean is required and results are only obtained after at least three hours. Marshall [19138, 2] has since modified his method by removing the ammonia after the conversion of the urea into ammonium carbonate by means of an air current.
Our procedure is similar to this latter one of Marshall, but it is much simpler ; accurate results are obtained in a little over two hours, and it is not necessary to use an extract of soy-beans.
In its essential features the method is no more than Folin’s method [1902, 2] of estimating ammonia in urine. By fitting together three or four cylinders and Allihn bottles in series with a sulphuric acid bottle at the end, duplicate estimations of ammonia and urea in urine can be carried out _ simultaneously. In the cylinders for the urea estimations are put 50 to 60 ce. of water, 1 g. of finely ground soy-bean and 5 (or 10) cc. of urine. These
R. H. A. PLIMMER AND R. F. SKELTON 71
cylinders are kept-in a water bath at a temperature of 35—40° and an air current is drawn through the series. After about an hour the rubber connections between the cylinders and bottles are disjointed and 1 g. of anhydrous sodium carbonate is dropped into the cylinders; they are then connected together again and the air current drawn through for another hour. To prevent frothing liquid paraffin Bp. has been used; it is superior to petroleum or toluene as it does not evaporate and it obviates the necessity of using a tube containing cotton wool between the cylinder and Allihn bottle. It is not necessary to carry out a blank experiment with soy-bean alone, since no ammonia was evolved by two different samples of the bean which were tested several times. The Allihn bottles are charged with excess of 0'1 N _ sulphuric acid (25 or 50 cc.) which is titrated with 0°1 N alkali using alizarin red as indicator.
The ammonia estimations are carried out exactly as described by Folin, except that liquid paraffin is used to prevent frothing.
The following are some comparative data obtained with 5 cc. of human urine by the urease method and Folin’s magnesium chloride method :—
Urease Folin ce. 0-1 N acid © ec. 0-1 N acid ee } 18-25 Pig } 18°55 os } Te eae es | 16-15 si } 14-2 phe } 14-25 bh } 19-1 oe } 19°25
The urease method has been used daily for some weeks and the results are all of the same order, the two data always coming out together or differing at most by 0°4 cc.; in fact all our latest determinations of urea in urine have been made by the urease method as we have become convinced of its accuracy’. The values are very slightly lower than by the magnesium chloride method, in which a decomposition of other urinary constituents, e.g. allantoin, or possibly urochrome as suggested by Haskins [1906], may occur.
The method has been tested on solutions containing different amounts of pure urea and the results are correct against a nitrogen determination by Kjeldahl’s method; thus
1 It is advisable to cleanse the cylinders immediately after use and occasionally with formalin, as the residues of protein, if left, form a suitable medium for the growth of bacteria, which may produce ammonia from amino-acids and other compounds,
72 R. H. A. PLIMMER AND R. F. SKELTON
Urea solution
taken ; : 2 cc. 23 19-35 cc. 01 N acid =9-68 for 1 ce. im ls amare pap ec se 4 cc. 88} 88-65, oo 0B
6 ce. fed 57°75, oe eee
gee Rare 25, a, ee
The urease of the soy-bean was shown to be specific by Takeuchi [1909] and its specificity has been more fully emphasised by Armstrong and Horton [1912], who also showed that its action is inhibited by ammonia; the removal of ammonia, as it is formed, by means of the air current will thus favour the completion of the reaction. Our method is in this way again more advantageous than Marshall’s.
Not only urea, but also allantoin, is decomposed by the magnesium chloride method of Folin, as is stated by Cathcart [1906] and Haskins [1906}. Since urease has no action upon allantoin the two substances can therefore be readily estimated in urines which contain both compounds; the difference between the two data will give the amount of allantoin. The two compounds and mixtures of the two compounds have been tested with the following
results :— 5 cc, urea (from above) - 6 ce. allantoin 10 ce. allantoin Kjeldahl 48-75 ce. 0°1N acid — 9°8 9°7}9°7 ec. OLN acid 9°5 Folin =— 4°5 EY 3) ec. O-1N acid roe OTR oe: ae 4:6 Urease 48-5 ec. 0O°1N acid 0 : 0 Hence 5 ce. allantoin = 4-87 ec. 0:1 N acid. 5 ec. urea+5 ce. allantoin 5 ec. urea +10 cc. allantoin Folin 53°4 é *B) : : 33.6) 58° ce. O-1N acid erst 57-4 cc. O-1N acid Urease 48°5) 4c. 48°5 485} 48°5 ,, ” 18.5} 48°5 ,, ” .. 5 ce. allantoin =5:0 ,, Bs 4:45 ,, BS
Several analyses of dog’s urine (volume for 24 hours made up to 1000 ce.) have also been made, the figures being in all cases cc. of 0°1 N acid :— |
: Urea + allantoin Total N Ammonia (5 ec.) Urea (5 cc.) Allantoin (5 ec.) (25 ee.) less ammonia less ammonia in 5 ee, ae - 16°7) «0. 14-4 9 8-3 16-6} 16°65 15.1} 14:8 1:85 156 : 13-0)... 11:2 : 61 12-9} 12°95 116} 14 1:55 . : 6°6 . 74 2-0 6-4 6:5 54} O2 1°3 21:3 . 19-1 17-2 3-0 19.3; 192 174} 178 1-9
R. H. A. PLIMMER AND R. F. SKELTON 73
The difference between the figures in columns 3 and 4 gives the amount of allantoin.
The estimation of allantoin in urine by Wiechowski’s method does not lend itself to routine work and our procedure is undoubtedly more rapid than that adopted by Miss Lindsay [1909].
SUMMARY.
The estimation of urea in urine is quickly and accurately made by decomposing it with urease (1 g. powdered soy-bean) at 35 to 40° for one hour. During this time the ammonia evolved is removed by an air current as in Folin’s method for estimating ammonia. One gram of anhydrous sodium carbonate is then added and the air current is continued for another hour. Liquid paraffin B.P. is very convenient for lessening the
frothing.
Since urease does not decompose allantoin and since both allantoin and urea are quantitatively decomposed by the magnesium chloride method of Folin, the amount of allantoin in those urines, which contain both compounds, is readily estimated by difference.
REFERENCES.
Armstrong and Horton (1912), Proc. Roy. Soc. B, 85, 109. Catheart (1906), J. Physiol. 35, Proc. viii. Folin (1901), Zeitsch. physiol. Chem. 32, 504. —— (1902, 1), Zeitsch. physiol. Chem. 36, 333. —— (1902, 2), Zeitsch. physiol. Chem. 37, 161. —— (1903), Zeitsch. physiol. Chem. 37, 548. —— (1905), Amer. J. Physiol. 13, 45. Haskins (1906), J. Biol. Chem. 2, 243. Lindsay (1909), Biochem. J. 4, 448.
Marshall (1913, 1), J. Biol. Chem. 14, 283. —— (1913, 2), J. Biol. Chem. 15, 487, 495. Takeuchi (1909), J. Coll. Agric. Tokyo, 1, 1.
| DW) he nea Oa » hs = Fak ‘ - a a "a Ron,
IX. THE CHOLESTEROL OF THE BRAIN. II. THE PRESENCE OF “OXYCHOLESTEROL ” AND ITS ESTERS.
By MARY CHRISTINE ROSENHEIM. From the Physiological Laboratory, King’s College, London. (Received January Ist, 1914.)
In a previous communication [1906] I was able to show, by the
methods then available, that Baumstark’s statement with regard to the :
occurrence of cholesterol esters in brain must be considered erroneous. My observations were in accordance with those of R. Biinz [1905], whose work was published a few months before mine. The methods used by Biinz and by myself were of an indirect nature, depending on variations in melting points etc., and my failure to isolate cholesterol esters from the oily mother liquors of the brain-cholesterol might be ascribed to the well-known difficulty inherent in this procedure. Since this work has been carried out, Windaus
[1910] has described a quantitative method for the estimation of free and —
combined cholesterol, which depends on the insolubility of the digitonin- cholesterol compound. It seemed therefore of interest to reinvestigate the question of the presence of cholesterol esters in brain by the help of this new method. This was all the more desirable as indications of their presence may be taken from statements occurring.in the recent literature on the subject. (Lapworth [1910], Lorrain Smith and Mair [1911], Lifschiitz [1913, 1].)
Cholesterol esters are now usually estimated in an indirect way, i.e. by
calculating as combined cholesterol the difference between the free and the.
total cholesterol content by Windaus’ method, estimated in relatively small aliquot parts of suitably prepared extracts before and after saponification. As my previous work showed conclusively that cholesterol esters, if present at’ all in brain, would be found only in exceedingly small amounts, this indirect method of estimation does not offer much hope of success, as the limits of error of the method are greater than the possible differences to be looked for. It was therefore decided to work up the whole brain and to
M. C. ROSENHEIM 75
remove the cholesterol completely by precipitation with digitonin, leaving the esters, if present, in the filtrate. One half of the filtrate was to be used for the actual preparation of the esters, whilst the other half was to be saponified, and the cholesterol, originally present as ester, precipitated with digitonin.
Although a whole adult human brain was examined in this way for cholesterol esters, no trace of them could be found. The same result was obtained with a child’s brain. In both cases, however, digitonin produced a distinct weighable precipitate in the final solution. The examination of this precipitate brought out the fact that the substance in question certainly did not consist of cholesterol. The substance recovered from its digitonin com- pound, in contradistinction from cholesterol, gave with glacial acetic and sulphuric acid. a violet colour, which on the addition of ferric chloride turned to a bright emerald green. This green solution, on spectroscopic examination, showed a very characteristic band in the red (between 2 630 and 2» 650), which has been described by Lifschiitz [1913, 1] as typical for the compound which he termed “ oxycholesterol.” According to him, “ oxy- cholesterol” is present in blood and in most organs with the exception of the liver and bile? (see also Unna and Golodetz [1909], Lifschiitz [1913, 2], Schreiber and Lénard [1913, 1 and 2).
Lifschiitz ascribes to this chemically ill-defined substance an important role in the metabolism of cholesterol, which seems to deserve more attention than it has hitherto received, the discussion of which, however, is outside the scope of this communication. The results of the present investigation in any case seem to justify the conclusion that in human brain a substance of this nature exists mainly in the form of esters.
EXPERIMENTAL.
For the first experiment the brain of a child of three months was used. Cause of death, pneumonia; weight of brain, 512 g. After removal of the membranes, the brain was finely minced. A water estimation was made in a sample of the mixed brain and showed 85°8 per cent. In order to obtain an idea as to the amount of digitonin necessary for the complete removal of cholesterol from the brain extracts, the percentage of cholesterol in the brain was estimated quantitatively. As the application of Windaus’ method to the brain is complicated by the presence of the large amounts of lipoids, the
1 In this connection it may be mentioned that, according to my observations, gallstones contain ‘‘oxycholesterol.” If this result should be confirmed by a more extensive investigation, it might have a considerable bearing on the problem of their formation.
76 M. C. ROSENHEIM
following method was used, which represents a combination of O. Rosenheim’s method for the preparation of cholesterol [1906] with that of Windaus for its quantitative estimation. As the same method was used in the subsequent experiments it may be described here in detail.
Five grams of minced brain were mixed with 15 g. of plaster of Paris}. After the mass had set hard, it was broken up and the drying completed in a desiccator. The brain plaster mixture was then extracted with acetone in a Soxhlet apparatus for several days, the extraction flask being changed repeatedly. The extraction was considered complete when the last extract after evaporation of the solvent left only a trace of residue, which did not give any precipitate when digitonin was added to its alcoholic solution. Eat
The combined acetone extracts were evaporated to dryness, the residue being taken up in 70 ec. of 95 per cent. alcohol, filtered and precipitated hot with a large excess (35 cc.) of a 1 °/, solution of digitonin (Merck) in 90 °/, alcohol. The contents of the flask were kept boiling for a few minutes and the precipitate allowed to settle at room temperature over-night. The digitonin- cholesterol was filtered on a weighed Gooch crucible, washed with alcohol and ether and dried to constant weight in a toluene oven at 105°. From 5 g. of fresh brain was obtained 071435 g. digitonin-cholesterol (mean of two analyses). According to Windaus the amount of cholesterol contained in digitonin-cholesterol is found by multiplication with the factor 0:2431 ; therefore the fresh brain contained 0-70 per cent. cholesterol.
The bulk of the minced brain was put into 1 litre of acetone, allowed to stand overnight and then filtered through a Buchner funnel under pressure. Five subsequent extracts each with 500 cc. cold acetone were made, and the extraction then continued three times with 800 cc. boiling acetone. Previous experience with other organs had shown me that by this procedure cholesterol and cholesterol esters were completely removed.
The dry residue of the combined extracts was treated repeatedly with boiling 95°/, alcohol and filtered hot. An alcoholic extract, amounting to 1 litre, was thus obtained, which contained the whole of the cholesterol, free and combined, present in the brain.
The total quantity of free cholesterol contained in the whole brain, as calculated from the result of the quantitative estimation (see above), amounts to 3°5 g., which would require about 10°5 g. of digitonin. In order to ensure complete precipitation, 13 g. of digitonin dissolved in 1300 cc. of 90°/, alcohol were added to the hot alcoholic brain extract. A copious precipitate formed and was allowed to settle over night. The complete removal of
cholesterol was tested for by the formation of a precipitate on adding a few
drops of an alcoholic solution of cholesterol to a sample (a few cc.) of the clear supernatant fluid. The precipitate was well washed with 90°/, alcohol and with ether. A clear filtrate was thus obtained which should contain the whole of the cholesterol esters present in brain.
* It is essential that the best quality of plaster of Paris should be used. The “superfine” quality as sold by makers of plaster-casts was found to be suitable.
M. C. ROSENHEIM 77
The filtrate was-heated on the water-bath until most of the alcohol had evaporated and then transferred to an automatic extraction apparatus (Maassen’s), in which it was extracted for many days with a mixture of ether and petroleum ether until the extraction was complete.
The ethereal extract was evaporated, the residue taken up with boiling absolute alcohol and saponified by boiling with sodium ethylate. Water was added and the saponified extract transferred again to the extraction apparatus. The final ethereal extract, which should contain the cholesterol of the whole of the cholesterol esters, was again evaporated, taken up in 70 cc. of hot 95 °/, alcohol and 70 cc. of 1°/, digitonin solution were added. There was no im- mediate precipitate, but after standing for some time at room temperature a small flocculent precipitate settled, which appeared to be quite different from the crystallised digitonin-cholesterol as usually obtained. The precipitate was treated in the usual way and when dry weighed 0°2131 g.
It will be shown in connection with the next experiment that this precipitate did not consist of digitonin-cholesterol, but it is of interest to note that on the assumption that one quarter of it represents cholesterol, the percentage of combined cholesterol in the whole brain would only amount to 0:01 per cent.
In the absence of any evidence to the contrary, it may be argued that cholesterol esters, even if absent in a child’s brain, may be present in the adult hyman brain.
In the next experiment, therefore, the whole of an adult human brain was treated in a similar manner. ‘The brain of a man (cause of death, heart disease) weighing 1270 g. contained 78-9 per cent. of water and 1-95 per cent. of free cholesterol (mean of two determinations) estimated as in the previous experiment.
The extraction was carried out as in the first experiment, ten cold and - three hot extractions with acetone having been made. In order to economise the valuable digitonin, of which at least 80 g. would have been necessary, a large bulk of the free cholesterol was allowed to crystallise out from the combined extracts. As the first crystallisation might possibly carry down any cholesterol ester present, the product was recrystallised twice from a mixture of alcohol and acetone, the mother liquors being added to the main extract. In this way, 21 g. of pure cholesterol were separated.
. To the final extract, dissolved in 95°/, alcohol, 850 cc. of 1 per cent. digitonin solution were added and the precipitated digitonin cholesterol removed as previously described. The whole of the free cholesterol was thus removed. The resulting filtrate was extracted with ether in the Maassen’s
78 . M. G ROSENHEIM
extraction apparatus as above described. One half (200 cc.) of the ether extract was reserved for the isolation of the esters as such, and the other half was saponified as before, and again extracted with ether. The final residue dissolved in 50 ce. of 95 °/, alcohol was precipitated with 70 ec. of 1°/, digitonin solution, The formation and appearance of the precipitate resembled closely that observed in the previous case. The dry precipitate weighed 02635 g- The whole brain would therefore yield 0°5270 g. of this substance. Assuming again that one quarter of it consists of cholesterol, the percentage of combined cholesterol in the whole brain would amount to 0°01 g. per cent., Le. exactly* . the same figure as calculated for the child’s brain.
Examination of the digitonin compound.
Having obtained a weighable amount of a digitonin compound in the two cases, it became necessary to investigate it further and to establish its identity or otherwise with digitonin-cholesterol. Windaus has already shown that, by means of boiling xylene, cholesterol can be recovered from its combination with digitonin. Before subjecting the substances obtained from brain to this treatment, it was thought advisable to test the reliability of this method when dealing with such small quantities as are obtained from brain. For this purpose 0°26 g. of digitonin-cholesterol, as prepared by direct precipita- tion of a brain extract, was treated in a suitable extraction apparatus with boiling xylene for 75 hours. From the xylene extract, cholesterol was isolated in the typical crystals, without any difficulty.
The digitonin precipitates, obtained from brain after the removal of free cholesterol, were then treated separately with xylene in the extraction apparatus. In both cases I was unable to detect any cholesterol crystals. The xylene extract left on evaporation an oily residue (approximately 35 °/, of the digitonin compound), in which under the polarising microscope some anisotropic sphero-crystals could be detected. All attempts to bring this 1esidue to crystallisation were unsuccessful. ee
The behaviour of the substance suggested a similarity to the product described by Lifschiitz [1913, 1] as “ oxycholesterol,” which according to him also occurs as ester in ox brain. With this name Lifschiitz designates for convenience sake the primary neutral oxidation products of cholesterol, which are as yet insufficiently defined chemically, but which under certain conditions give rise to a colour reaction possessing a highly characteristic absorption spectrum. By means of this spectroscopic behaviour “oxycholesterol” can not only be detected in the presence of cholesterol, but its quantity may be
M. C. ROSENHEIM 79
estimated spectrometrically by een with a standard “ oxycholesterol ” solution.
In order to compare the substances obtained from brain with “oxy- cholesterol,” I prepared some of this substance according to Lifschiitz by oxidising pure cholesterol with benzoyl peroxide in glacial acetic acid solution. Following scrupulously the detailed directions of Lifschiitz, there is no difficulty in obtaining the sticky amorphous product as described by him. ‘This was boiled with alcoholic potash, extracted with ether and the residue from the ether extract purified by means of methyl alcohol. After careful evaporation of the methyl alcohol and drying im vacuo, the product obtained was a yellowish brittle resin, insoluble in water, but soluble in all the usual organic solvents.
This substance gave the typical reaction as described by Lifschiitz. One milligram dissolved in chloroform gave with glacial acetic acid and sulphuric acid the first phase of the Lifschiitz reaction (violet), which changed almost instantaneously on the addition of ferric chloride to a bright emerald green, showing a well defined band in the red between » 630—650. A _ cholesterol solution remained perfectly colourless under the same conditions’. When these tests were carried out with the substances isolated as above described by means of their digitonin compound from brain, it was found that they gave the typical “oxycholesterol ” reactions, even in minute quantities. One milligram of the substances gave a strong Lifschiitz colour reaction, and the green solution showed the well marked typical absorption band in the red.
From some preliminary experiments, in which the coloured solutions were compared with a standard “oxycholesterol” solution in a Zeiss comparison spectroscope, it seems that their spectroscopic value is at least as high as that of the standard.
It is evident therefore that “ oxycholesterol” exists in the brain in the form of esters and that free “ oxycholesterol ” is precipitated from its alcoholic solution by digitonin. This latter fact has also been demonstrated by Lifschiitz [1913, 3].
1 In conjunction with Dr O. Rosenheim I have found another characteristic reaction of ‘*oxycholesterol” which may be carried out with the dry substance. If a minute quantity of dry ‘‘ oxycholesterol” (or the residue of a drop of a solution to be tested) is put on a slide or watch-glass and a drop of methyl sulphate (techn.) is added, a brilliant purple colour is at once produced. Pure cholesterol remains unchanged under these conditions, but on warming or on prolonged standing it develops a bright cherry red colour. The reaction is not given with pure methyl sulphate. The reason for this as well as the characteristic behaviour of cholesterol,
oxycholesterol and other substances towards methyl sulphate in solution and in the presence of ferric chloride ete, will be described in a subsequent communication.
80 M. ©. ROSENHEIM
With regard to the quantity of the esters present we may form an approximate estimate from the amount of the digitonin compound weighed as such. If we assume that in analogy to digitonin-cholesterol it contains a quarter of its weight of “ oxycholesterol,” the amount of the latter as ester would be 0:01 °/, in the fresh brain. Its amount, however, may be larger as we have at present no experimental data as to the completeness of its precipitation by digitonin.
An unsuccessful attempt was made to isolate the esters as such from the
solution reserved for this purpose (see above, p. 78). Just as is the case with free “oxycholesterol,” so also its esters are evidently not readily crystallisable [see also Lifschiitz, 1913, 3]. Various solvents were tried, but the substance did not show any tendency to crystallise, although under the polarising microscope some anisotropic needles could be detected in it. It gave no reaction for free “oxycholesterol,” but did so readily after saponi- fication.
In this connection an experiment may be mentioned which I carried out in order to satisfy myself that “oxycholesterol” is not an artifact, Le. a product formed from cholesterol esters during the lengthy process of saponi- fication, etc. For this purpose 0°25 g. pure cholesterol palmitate was carried through the process in exactly the same way as the solution of esters described above. The product finally obtained did not give any trace of an oxy- cholesterol reaction.
The occurrence of free “oxycholesterol” in brain.
As the above results have clearly demonstrated the existence of “oxy- cholesterol” esters in brain, it seemed of interest to search for the presence of free “oxycholesterol.” This substance forms, just like cholesterol, a compound with digitonin (Merck)', and it might therefore be expected that
any free “oxycholesterol” present would be found in the primary digitonin |
precipitate. Its presence in this can be easily shown if a small quantity be dissolved in warm glacial acetic acid. After cooling, the solution gives
1 This fact makes it unfortunately impossible to accept unconditionally the results of cholesterol determinations according to Windaus’ method, unless the digitonin compound after weighing is tested qualitatively, and if necessary, quantitatively, for ‘‘ oxycholesterol.” The question becomes still more complicated since Windaus and Schneckenburger [1913] have recently discovered that ‘crystallised digitonin” contains, besides digitonin, in variable quantities, an amorphous glucoside called gitonin. It is possible that ‘‘oxycholesterol” is precipitated by ‘‘crystallised digitonin” only owing to the presence of gitonin. This would explain the contradictory statements of Lifschiitz [1913, 3] and of Schreiber [1913], according to whom “ oxycholesterol” is not precipitated by purified digitonin. In any case, the whole method of cholesterol estimation in tissues by means of digitonin seems to require further investigation.
a
M. C. ROSENHEIM 81
Lifschiitz’s reaction. The dry digitonin-oxycholesterol compound also gives the methyl sulphate reaction mentioned in the foot-note on p. 79.
By means of these reactions I was able to show the presence of “ oxy- cholesterol ” in the adult brain referred to above, but not in the brain of the child.
' This result seemed suggestive if we remember that in the brain of new- born children the cholesterol and galactoside percentage is known to be low, whilst the percentage of unsaturated phosphatides is high as compared with an adult’s brain. One might assume that a similar relationship holds good for “ oxycholesterol.” In order to test this suggestion still further, I examined the extracts of another adult human brain, of a child’s brain (5 days old) and that of a human foetus (36 weeks old). Here again the adult brain gave a strong “oxycholesterol ” reaction, whilst that of the child gave a negative result. The brain of the foetus gave a positive reaction.
Before coming to a final conclusion with regard to this question, however, more experiments seem to be necessary.
CONCLUSIONS.
(1) The examination of the whole human brain by means of Windaus’ method has shown the complete absence of cholesterol esters.
(2) Human brain contains “oxycholesterol” esters to the extent of at least 0°01 per cent.
(3) Free “oxycholesterol” seems to be present in adult human brain, but not in the brain of young children.
The expenses of this research have been defrayed by a grant from the Government Grant Committee of the Royal Society.
REFERENCES.
Biinz, R. (1905), Zeitsch. physiol. Chem. 46, 47.
Lapworth, A. (1910), J. Path. Bact. 15, 254.
Lifschiitz, I. (1913, 1), Biochem. Zeitsch. 48, 373.
—— (1913, 2), Biochem. Zeitsch. 52, 206.
—— (1913, 3), Biochem. Zeitsch. 54, 212.
Rosenheim, M. C. (Tebb, M. C.), (1906), J. Physiol. 34, 106. ° Rosenheim, O. (1906), J. Physiol. 34, 104.
Schreiber, E. (1913), Miinch. med. Woch. 60, 2001.
—— and Lénard (1913, 1), Biochem. Zeitsch. 49, 458.
—— —— (1913, 2), Biochem. Zeitsch. 54, 291.
Smith, J. Lorrain, and Mair, W. (1911), J. Path. Bact. 16, 131. Unna, P. G. and Golodetz, L. (1909), Biochem. Zeitsch. 20, 469. Windaus, A. (1910), Zeitsch. physiol. Chem. 65, 110.
—— and Schneckenburger, A. (1913), Ber. 46, 2628.
Bioch. vir 6
X. THE CHOLESTEROL OF THE BRAIN. III. NOTE ON THE CHOLESTEROL CONTENTS OF HUMAN AND ANIMAL BRAIN.
By MARY CHRISTINE ROSENHEIM. From the Physiological Laboratory, King’s College, London.
(Received January 12th, 1914.)
The data as to the percentage of cholesterol in brain available in the literature are still rather scanty, and most of the estimations carried out before the introduction of a trustworthy quantitative method can only be considered as approximate. Since Windaus’ work has made it possible, by the use of digitonin, to obtain more accurate figures, some estimations of the cholesterol percentage in brain have been published. These were, however, mostly made incidentally with other work, and, moreover, they have been usually expressed in percentages of the moist organ or its chloroform extract. As the amount of water in brain is variable it is impossible, in the absence of any water estimations, to reduce the available figures accurately to a uniform standard.
The following estimations were made by a combination of O. Rosenheim’s [1906] method for the preparation of cholesterol with Windaus’ [1910] method for its quantitative estimation. The method, which obviates the difficulties due to contamination with other brain lipoids, has been published in full in the preceding paper [1914]. In some cases search was made for the presence of cholesterol esters with negative results.
Unfortunately this somewhat tedious work had been finished before the results of my previous communication.[1914] on the occurrence of “oxychol- esterol ” in brain were obtained. In the light of the latter work, the figures must be taken to express the sum of the quantities of cholesterol and
“oxycholesterol” present. Except in the case of human brain, “ oxycholesterol ” was not tested for.
Description of brain
Man (i)
Man (ii)
Child (aged 3 months)... Child (¢ aged 5 days) ... Foetus ( ? aged 36 weeks) Dog...
Cat ...
Ox (i)
Ox (ii) Sheep Rabbit (i) Rabbit (ii) Fowl Codfish (i) Codfish (ii)
M. C. ROSENHEIM
Water percentage 78°86 78°90 85°30 89-99 90°29 76°18 76°53 78°83 78°32 79°50 77°86 79°15 80°34 84°03 84-94
83
Cholesterol percentage in moist brain in dry brain
1°93 ) 4. oa 1-95 9-22 1°91) ,. oes 1-91 9-01 0°66 ) . et 0-69 4-89 0°53). : sles 0°53 5°29 0°39), ir 0-39 4-07 2°73 } 9. . 2-79 ; 276 11°59 2-29) o. ‘ ho 2°35 9-99 2-39 11-28 2°57 |. , oa 2°61 12°04 2-13 10°37 2°12 9°57 1-90 9-11 1:43 ),. Ree 1-45 7-40 1-92 12-02 1°79 11-89
The expenses of this research were defrayed by a grant from the Government Grant Committee of the Royal Society.
REFERENCES.
Rosenheim, M. C: (1914), Biochem. J. 8, 74. Rosenheim, 0. (1906), J. Physiol. 34, 104.
Windaus, A. (1910), Zeitsch. physiol. Chem. 65, 110.
6—2
XI. ON THE RESISTANCE OF TRYPSIN SOLUTIONS TO HEAT.
By EDWARD STAFFORD EDIE. From the Physiological Department, University of Aberdeen.
(Received January 12th, 1914.)
The acticn of heat on aqueous solutions of many enzymes has been studied more or less carefully for many years, and the general conclusion arrived at has been that all enzymes in aqueous solution are destroyed when heated for a short time to about 70° or 75°. So much is this the case that frequently the activity of a substance after its solution has been heated has been taken as proving that the substance in question is not an enzyme (e.g. secretin).
Since trypsin is practically without any digestive action in acid or neutral solution, the effect of heat on this enzyme was at first tested principally in alkaline solution, and it was found that such a solution rapidly became inactive at as low a temperature as 50° or sometimes even at 45° [Biernacki, 1891]. Vernon [1901] also found that fresh preparations_of trypsin lost more than half their activity when kept in 0°4°/, sodium. carbonate at 37° for an hour. Similar results have been found by Vernon in later experiments, and by other observers.
On the other hand Vernon [1904] found that the presence of protein protected trypsin solutions from the effect of heat to a considerable extent, and the same protection was afforded by proteoses or peptone. Bayliss and Starling [1903] had previously noticed the protective effect of proteins or their hydrolytic products on solutions of trypsin.
The action of acids on trypsin, however, has not been so fully studied. Langley [1881] found that trypsin was considerably weakened by warming its solution for 2} hours with 0°05 °/, hydrochloric acid. Wrdblewski, Bednarski and Wojezynski [1901] found that trypsin when kept at 37° for a few hours in hydrochloric acid of over 0'14°/, was considerably affected, and if 0°56°/, acid were used the enzyme was sometimes completely destroyed.
The question of the effect. of heat on ‘teypels was investigated later by
KE. §. EDIE 85
Schmidt [1910], who stated that trypsin in a slightly alkaline solution containing 5°/, of peptone could be boiled without being destroyed. The same protection against heat was afforded by a 2°/, solution of agar or a 10°/, solution of gelatin. Schmidt also stated that if trypsin powder were suspended in water-free glycerol it could be heated to 292° without being destroyed or much affected.
Schmidt’s work was repeated by de Souza [1911], who however found that 5 °/, peptone had hardly any effect in protecting trypsin solutions from destruction by heat. It appears from these experiments that an appreciable protection is afforded by 20°/, peptone under certain conditions, but if the heating is sufficient to cause complete destruction of the trypsin in pure aqueous solution, then the presence of peptone has only a slight protective effect. De Souza also tried the effect of heat on trypsin in presence of 20°/, peptone in acid, neutral and alkaline solutions. The solutions were heated to 80° for five minutes. No difference was observed between the acid and neutral solutions, the activity of these after heating being however slightly greater than that of the alkaline solution. Even in the case of the acid or neutral solution, however, over 85 °/, of the trypsin was destroyed, and about 90 °/, in the case of the alkaline solution.
Ohta [1912] also repeated the experiments of Schmidt, but failed to confirm his results.
In a paper just published, Mellanby and Woolley [1913] find that while trypsin is readily destroyed by heat in neutral or alkaline solution, if a solution of trypsin be made slightly acid, say with hydrochloric acid, it can be boiled for five minutes and yet retain considerable digestive power.
According to these observers, trypsin is destroyed in acid, more rapidly than in alkaline or neutral solution up to about 40°, but at higher temperatures the reverse is the case. At 40° there appears to be an optimum protective concentration of acid, above or below which the rate of trypsin destruction is accelerated.
Before the publication of the work of Mellanby and Woolley, and unaware of their experiments, I had tested the effect of heat on trypsin in connection with another research, and obtained results of a similar character. The method of testing the digestive power of the trypsin solutions was that of Hedin [1903]. The trypsin was allowed to act on a solution of caseinogen in presence of toluene and after a definite interval excess of tannic acid was added, to precipitate unaltered protein, meta-protein and proteoses. After standing 12 hours or more the precipitate was filtered off and the nitrogen
86 E. S. EDIE
determined in a portion of the filtrate by Kjeldahl’s method. Controls were also carried out to show the effect, if any, of the alkali used on caseinogen, and the amount of nitrogen not precipitated by tannic acid was also determined in each solution of trypsin used. |
The following are the principal results obtained :—
1. Benger’s Liquor Pancreaticus used as trypsin solution, 10 cc. of this —
requiring 1°6 cc. of N/10 sodium hydrate for neutralisation. 2 °/, caseinogen -n normal sodium carbonate was the substrate. A portion of the trypsin was
boiled for three minutes and cooled before adding the caseinogen. Digestion
was continued at 37° for three hours.
Digestion in ce. of N/10 nitrogen not ppted by tannic acid
(a) lee. trypsin, 20 cc. water, 40 ce. caseinogen 49°8 () 1 9 (boiled) ” oe) 29°7
In this experiment the effect of boiling trypsin in slightly acid solution was to leave 60°/, of the original digestive power unimpaired. |
2. 10 cc. of the above trypsin solution were neutralised with sodium carbonate and made up to 25 cc. with water. Three portions (a, b and ¢) were boiled for three minutes in neutral, alkaline and acid solution respectively, cooled, and kept at 37° with 20 ce, of 2°/, caseinogen in 2N sodium carbonate for three hours. .
Digestion in ec. of
N/10 nitrogen not ppted
by tannic acid (a) 2-5 ec. trypsin, 20 cc. water... ies nee 0:2 (b) 2°5 ce. trypsin, 19 cc. water, 1 cc. N Na,CO, as 0-1 (c) 2-5 cc. trypsin, 19 ce. water, 1 ce. N HCl an 20°8 (d) 2°5 unboiled trypsin, 20 cc. water _—.... a 20°9
It may here be mentioned that in all the experiments carried out, any differences in reaction due to the trypsin being boiled in acid etc. were adjusted before the caseinogen was added. Special care was taken also to ensure that none of the trypsin escaped being heated to 100°.
In the above experiment it will be seen that after being boiled in acid solution for three minutes, the trypsin still retained all its power of digesting caseinogen, while boiling in alkaline or neutral solution had completely destroyed the enzyme. .
The digestive power of this trypsin before and after being boiled as above was also tested on boiled ox fibrin, the amount of nitrogen in the filtrate from the undissolved fibrin at the end of the digestion being taken as the measure ef the action of the enzyme. It was found that on such fibrin trypsin acts
E. S. EDIE 87
only slowly, producing much less effect in a given time than when acting on caseinogen. Nevertheless the trypsin boiled in acid dissolved as much fibrin as the unboiled trypsin, while that boiled in neutral or alkaline solution again had no digestive power.
3. Merck’s trypsin used. A weak solution of this trypsin was dialysed - against running water for 18 hours and filtered. The solution was neutral. The trypsin contained 0:02°/, nitrogen. Three portions were boiled for three minutes (a, b and c) and then allowed to act on 20 cc. of 2°/, caseinogen in 0°4 N Na.CO, at 37° for three hours.
Digestion in ec. of
N/10 nitrogen not ppted by tannic acid (a) 25 ce. trypsin, 1 ec. N Na,CO, 13 pe 0 (b) 25 cc. trypsin, 1 ec. water ... pe oe 0-2 (c) 25 ce. trypsin, 1 ec. N HCl... ‘eh wn 21°4 _ (ad) 25 ee. unboiled trypsin ase ous Eph 28°9
In this experiment 75°/, of the original digestive power remains after boiling the trypsin in acid solution, but the trypsin is destroyed in neutral or alkaline solution.
For the rest of the experiments the trypsin used was prepared in the method described by Hedin [1905], An ox pancreas was minced and allowed to undergo autolysis at 37° in presence of water and toluene for a day and filtered. The filtrate was again kept at 37° for two days, dialysed against running water for two days, filtered, and kept with a little toluene.
This trypsin solution was neutral, contained less than 0°01°/, nitrogen, and gave practically no biuret reaction.
4, Three portions of this trypsin were boiled for three minutes, and cooled before adding 20 cc. of caseinogen (same as in last experiment). Digestion lasted three hours.
Digestion in ce. of N/10 nitrogen
(a) 25 cc. trypsin, 0°5 ec. N/10 Na,CO, _... na 0°2 (b) 25 cc. trypsin, 0-5 cc. water... aes wu 0°2 (c) 25 ee. trypsin, 0°5 cc. N HCl Sas a 65 (d) 25 ce. unboiled trypsin - ... on eal 10°5 (e) Control (water + caseinogen) a wee 0
In this experiment we see that over 60°/, of the original digestive power of the trypsin survives after the acid solution has been boiled, but none in the case of the neutral or alkaline solutions, the 0°2 cc. being within the limits of experimental error.
5. In order to see to what extent this trypsin would survive more prolonged heating, 25 ec. of the solution together with 5 ce. of N/10 HCl
88 | E.'S. EDIE
were brought to boiling in a flask and then put in a steriliser for 20 minutes. During the whole of this time the temperature throughout the interior of the steriliser was 100°. The contents of the flask were then cooled and neutralised.
To a fresh portion of 25 ce. trypsin 5 cc. of N/10 HCl were added, and immediately neutralised. Then to both flasks were added 20 ce. of the usual caseinogen solution.
25 cc. of water were treated in exactly the same way as the fresh portion of trypsin, and the flasks were kept at 37° for 4°25 hours. |
Digestion in cc. of
N/10 nitrogen (a) Boiled trypsin bes pa 4°2 (b) Fresh trypsin oa sie 16°7 (c) Control ve ‘an tee 0
The amount of nitrogen not precipitated by tannic acid, which was contained in the 25 cc. trypsin used, was also estimated, and allowed for in the above results. It corresponded only to 1 cc. of N/10 nitrogen.
From this experiment it appears that under suitable conditions a solution of trypsin can be heated to 100° for 20 minutes and yet retain 25°/, of its original digestive power.
6. I have repeated one of the experiments described by Mellanby and Woolley to test the effect of varying concentrations of acid on trypsin. My experiment was carried out at 45°, at which temperature the acid solutions were kept for 15 minutes. 20cc. of the usual caseinogen solution were added and digestion continued at 37° for three hours.
Digestion in cc. of
N/10 nitrogen (a) 25 cc. trypsin, 0-2 cc. N HCl By is 8°0 (b) 25 ce. fe 0°4 . sy es : 8°9 Bt eee OR a ss 8°9 (d) 25 ce. rages, jc tae ts ey 93 (e) 25 ce. 9 10 ...,, Gs oss 8°9 (f) 25 ec. unboiled trypsin a by ie: 9°8
In the above series the trypsin seems to be the least protected by the weakest acid (0:008 N), each of the other concentrations of acid having much
the same effect. In all cases at least 80°/, of the original digestive power remains after heating.
E. 8S. EDIE 89
ene SUMMARY.
Solutions of trypsin when neutral or alkaline are rendered completely inactive by boiling.
Acid solutions of trypsin, on the other hand, after being boiled retain a considerable power to digest caseinogen. In some cases there is no destruction of this digestive power at all.
The power to digest caseinogen appears to be less affected by heat than the power to coagulate calcified milk, this being taken as the measure of the activity of trypsin by Mellanby and Woolley.
It may be that these two evidences of the action of trypsin are due to different sets of groupings of the trypsin molecule, and that the groupings to which the digestion of caseinogen are due are more thermostable than the others.
REFERENCES.
Bayliss and Starling (1903), J. Physiol. 30, 61. Biernacki (1891), Zeitsch. Biol. 28, 62.
Hedin (1903), J. Physiol. 30, 155.
—— (1905), J. Physiol. 32, 468.
Langley (1881), J. Physiol. 3, 246.
Mellanby and Woolley (1913), J. Physiol. 47, 339. Ohta (1912), Biochem. Zeitsch. 44, 472.
Schmidt (1910), Zeitsch. physiol. Chem. 67, 314. de Souza (1911), J. Physiol. 43, 374.
Vernon (1901), J. Physiol. 27, 269.
—— (1904), J. Physiol. 31, 346,
Wroblewski, Bednarski and Wojezynski (1901), Beitriige, 1, 289.
XII ON THE ACTION OF COAGULATING ENZYMES ON CASEINOGEN.
By ARTHUR HARDEN Anp ARCHIBALD BRUCE MACALLUM (Beit Fellow).
From the Biochemical Department, Lister Institute. |
(Received January 12th, 1914.)
Hammarsten [1872, 1874, 1877] first demonstrated that the rennin action on caseinogen was specific and-independent of the action of the calcium salts.
His explanation was that the caseinogen molecule was split up into a large.
molecule (Kise) and a smaller one (Molkeneiweiss). The “Kiise” was rendered insoluble by the presence of soluble calcium salts and formed the clot. Since then little has been done to determine the chemistry of the clotting process, and our knowledge of this branch of the subject has up till recently been untouched by investigators. | }
The recent literature contains views which contradict the theory advanced by Hammarsten. Schryver [1913, 1 and 2] and Mellanby [1913] both consider that the rennin clot is probably a combination of enzyme and protein, and Schryver states definitely that rennin alone causes no ‘proteoclastic change. Bosworth [1913] has found that the rennin does not split off any nitrogen from the caseinogen which remains in solution when the casein
is precipitated by dilute acetic acid. The protein molecule has therefore
undergone no cleavage into its components.
This, considered in connection with the results of his earlier work with van
Slyke [1913], leads him to believe that the ferment breaks up the caseinogen molecule into two molecules of casein each half the size of the original molecule. In the case of basic calcium caseinogenate (containing 4 equivalents of calcium) the casein produced is soluble in water but is rendered insoluble by the presence of small quantities of calcium chloride. The caseinogenate containing two equivalents of calcium gives a casein insoluble in water’.
* The English nomenclature—caseinogen for the unfermented protein and casein for the product of fermentation—is used throughout.
A = ot
A. HARDEN AND A. B. MACALLUM 91
ENZYMES EMPLOYED.
The enzymes used were Witte’s rennin powder 1:300,000, Griibler’s trypsin, and a preparation made from the seeds of the Withania coagulans [Lea, 1884].
The rennin preparation was used in neutral aqueous suspension, was exceedingly active and fermented in a short time when the concentration was as low as 1:500,000. In the experimental work the concentration employed was 1 of rennin to 300,000 of caseinogen solution. The trypsin was used in aqueous suspension, in the proportion of 1 part of the solid preparation to 12,000 of caseinogen solution. The Withania coagulans enzyme was prepared by grinding up the seeds to a fine powder and extracting with 5 per cent. sodium chloride solution. Three volumes of absolute alcohol were added and the ferment precipitated. It was filtered, washed with absolute alcohol and dried at 37°. One gram of the powder thus obtained was made up to 100 ce. with 5 per cent. sodium chloride solution ; 5 cc. of this solution were used to ferment 100 cc. of caseinogen solution. The trypsin and rennin contained no calcium or phosphorus estimable in three grams of the dried preparation, while the vegetable rennin contained a trace of phosphorus and no calcium.
PREPARATION AND PROPERTIES OF THE CASEINOGEN.
At the commencement of the investigation it was found that the commercial preparations of caseinogen, even when obtained from reliable sources, did not give satisfactory results. In fact the rennin had scarcely any appreciable effect on them. A modification of Hammarsten’s method was finally adopted which gave satisfactory caseinogen preparations, Schryver’s plan of drying in different grades of alcohol being employed.
The method was as follows :—skimmed milk containing 0:1 per cent. of fat was taken, diluted with five times its volume of distilled water and 0:1 per cent. of acetic acid added slowly, the whole being constantly agitated until the protein separated out. The caseinogen was then allowed to settle and the supernatant fluid syphoned off. The mixture of caseinogen and diluted milk serum was poured on a cheese cloth filter and the remaining fluid drained off. The caseinogen was then washed by mixing it with a volume of distilled water equal to the quantity of milk from which it was obtained and shaking thoroughly. When settled the water was again syphoned off and the caseinogen washed a second and third time with the same volume of distilled water. It was next redissolved by shaking with 0-2 per cent. sodium carbonate
92 A. HARDEN AND A. B. MACALLUM
solution and the precipitation and washing carried out as before, the solution, precipitation and washing being repeated at least twice after the original precipitation from milk. Finally it was ground up successively with 33 per cent., 66 per cent. and absolute alcohol and given two half-hour extractions in an agitator with ether, filtered and dried for two hours at 37°.
These preparations upon analysis were found to be completely free from fat and had a phosphorus content of 0°87—0-90 per cent. The ash from 10g. contained no estimable calcium. Solutions were prepared by grinding the dried powder with moist calcium carbonate in a mortar, centrifuging and filtering to remove the excess of calcium carbonate, the caseinogen when treated in this way having a solubility of 19-20 ce. (5 ce. gave by Kjeldahl’s method enough ammonia to neutralise this volume of N/10 acid). The solutions were opaque and milky in appearance, neutral to litmus and contained the amount of calcium necessary to form the basic caseinogenate. After treatment by rennin these solutions readily gave precipitates when about 5 per cent. by volume of N calcium chloride was added. ‘This amount of calcium chloride alone with the original caseinogen solutions produced no precipitate from 0°-40°, but when the temperature was raised to 50°-60° a precipitate was formed. The precipitates produced by the action of enzyme and calcium chloride rapidly contracted to a small fraction of the volume of the solution, while those produced by calcium chloride and heat settled very slowly and did not shrink in volunie.
COMPOSITION OF CASEINOGEN AND CASEIN.
The experiments were carried out as follows: 200 cc. portions of casein- ogen solution were put into beakers and the enzymes added in the proportions already mentioned. Control experiments were made, using equal amounts of caseinogen solution and enzyme solution inactivated by boiling, and submitting these to the same conditions as those containing the active enzyme, The mixtures were placed in an incubator at 37° for one hour, after which 10 ce. N calcium chloride (5°5 g. per 100 cc.) were added both to controls and fermented preparations. In the case of the fermented solutions the precipitate at once came down, Precipitates from the controls were obtained by heating them to 55°. When the precipitates had settled the supernatant liquid was decanted and the precipitates thoroughly broken up and shaken vigorously with 500 cc. of 10 per cent. alcohol in a tall cylinder. When they had again settled the clear fluid was syphoned off and this process repeated three times.
A. HARDEN AND A. B. MACALLUM 93
The last washing was found to be free from chlorides. To inactivate the enzyme the precipitates were mixed with water and boiled for five minutes, the controls being similarly treated. All the precipitates were finally ground up with absolute alcohol and ether and dried at 37°.
The nitrogen, calcium, phosphorus ratios were determined as an index of composition, nitrogen being taken as the fixed unit of comparison. Phosphorus was estimated by Neumann’s method and nitrogen by Kjeldahl’s. Calcium was estimated by ashing the dried caseinogen or casein, dissolving the ash in hydrochloric acid and neutralising with ammonia until the phosphates were precipitated. These were redissolved by adding strong acetic acid, and the calcium precipitated as the oxalate and