146 COATES, HARRISON AND KON: THE CHICK ASSAY OF [Vol. 76 The Chick Assay of Vitamin B12 and the Animal Protein Factor BY M. E. COATES, G. F. HARRISON AND S. K. KON (Presented at the meeting of the Biological Methods Group on Tuesday, May 23rd, 1950) A method of assay with chicks for vitamin B,, and the animal protein factor is described and its accuracy discussed. The method has been applied to several natural materials, and a comparison with microbiological results has been made. Evidence is presented for the existence of other factors besides vitamin B,, in the animal protein factor. Difficulty in obtaining satisfactory responses to crystalline vitamin R,, is reported. ALTHOUGH the method now described was originally designed to measure vitamin B,,, it has become obvious irom our results to date that the test is not specific for vitamin B,, alone and that other as yet unidentified vitamins produce a similar growth response in chicks.We prefer therefore to describe the test as an assay of the animal protein factor (APF), although the standard used so far has been a preparation of vitamin B,, (Examen, Glaxo) presumed to be free from other factors, and results have been calculated in terms of vitamin B12. Preliminary accounts of our method of assay have already been presented.lP2 The present paper describes the application of the method to the assay of a number of natural materials ; wherever possible, results have been compared with those obtained microbiologically and it is evident that the two methods are often not in close agreement. EXPERIMENTAL PRODUCTION OF THE CHICKS- As previously reported,l normal stock chicks proved unsuitable for this assay, as their reserves of animal protein factor a t hatching were high.Chicks were therefore hatched from hens given an all-vegetable diet.2 I t was of major importance in all this work that both hens and chicks should be kept on wire screens to prevent access to their droppings, which have been shown by Rubin, Bird and Rotlhchild3 to contain components of the animal protein factor, probably as a result of microbial synthesis after voiding. A deficiency of animal protein factor in hens results in lowered hatchability of their eggs,2s4 so that to obtain adequate numbers of chicks we found it necessary to include a small but sufficient quantity of vitamin B,, in the hen diet to maintain hatchability at a reasonable level without excessive storage in the chicks.About 3 pg of vitamin 13,, per 100 g of diet proved much too high for this purpose,, but 0.3 pg per 100 g was satisfactory. In 9 months, with the addition of this amount of vitamin B,,, hatchability of the fertile eggs from 30 hens never fell below 50 per cent., yet the chicks showed marked signs of deficiency of vitamin B,, a t an early age.March, 19511 VITAMIN BIZ AND THE ANIMAL PROTEIN- FACTOR 147 DESIGN OF THE ASSAY- The assay was designed on similar lines to those already in use here6 for other members of the vitamin B complex, except that chicks of both sexes were used. The basal diet, with a high content of vegetable protein, had the following percentage constitution.ground maize, 38.6 ; ground barley, 20 ; defatted soya grits, 35 ; dried grass, 3 ; bone meal, 1.5 ; limestone flour, 1 ; sodium chloride, 0.672; MnS0,.4H20, 0.028; arachis oil, 0.2. To each 100 g of diet were added 400i.u. of vitamin A, 50 B.S.I. units of vitamin D,, 0-15pg of ribofiavine and 1 pg of nicotinic acid; the doses of standard and test preparations were incorporated in the diet. With chicks produced as described above, a linear relationship was established between their body weights at 4 weeks of age and the logarithm of the dose of vitamin B12. This relationship held over a range of 0.5 to 3.0 pg of vitamin B,, per 100 g of diet. As far as possible assays were done with at least three dose levels of both standard and test material, with 5 to 10 chicks on each dose.Calculations were made as recommended in British Standards Specification 911 : 1940,6 and in most assays the graphs of response to standard and test preparation proved straight and parallel to each other. ERROR OF THE ASSAY- The error of estimation unfortunately proved to be high, owing chiefly to the great variation in response among the chicks at each dose level. As the number of. chicks available TABLE I REPRODUCIBILITY OF RESULTS FOR THE CHICK ASSAY OF THE ANIMAL PROTEIN FACTOR Pooled T.F.L. at P Sample Result P = 0.95 Xa Result Error b g of vitamin B,, per g pg. of vitamin BI2 per g Fish solubles.. .. . . (i) 1.27 0.36 to 2-28 (28 to 180%) (ii) 1.23 0.68 to 1-90 (55 to 155%) (iii) 1.01 0.42 to 1-80 (42 to 176%) Fuller’s earth adsorbates of ( i ) 134.5 68 t o 267 (50 to 198%) (61 to 144%) 1.17 0.99 to 1.37 (85 to 118%) 120.5 102 to 142 i 0*36 (72 to 118%) } 0.23 Streptomyces griseus liquor (ii) 113.9 70 to 164 is limited by the special conditions needed to produce them, we find it is impracticable to attempt to reduce the variance either by increasing the size of the test groups or by severely culling the chicks before the beginning of the test.It is possible that the accuracy could be improved if pedigree records of individual birds were kept ; the birds could then be distributed over the experimental treatments according to parentage, as is usual with the “litter mate” system adopted in rat assays. In spite of the error being so high, results were fairly reproducible, as can be seen in Table I.The results for three successive assays of fish solubles and for two of an adsorbate from Streptomyces griseus liquor agreed closely among themselves, although the error of each individual test was high. The value of x2 for each series of assays was very low indeed, and the pooled results had limits of error well within those usually accepted in a biological assay. COMPARISON WITH MICROBIOLOGICAL RESULTS Whenever possible microbiological assays were made of the samples of material tested with chicks. The results are shown in Table IT. We are very much indebted to our colleagues F. W. Wilby and J. E. Ford and to Dr. W. F. J. Cuthbertson, of Glaxo Laboratories Ltd., for these microbiological tests. It will be seen from the table that results by the two methods rarely agreed and that the chick assay gave on the whole considerably higher values.These discrepancies could be the effect of (a) destruction or incomplete extraction of vitamin B,, during preparation of the sample for microbiological assay, (b) the existence of vitamin B,, in a bound form available to chicks but not to micro-organisms or (e) the presence in the148 COATES, HARRISON AND KON: ‘THE CHICK ASSAY OF [Vol. 76 samples of other growth factors essential to chicks but not to EugZena gracilis or LactobaciZlus leichmannii. Two assays of fish solubles and two of whale solubles support the last as the most probable explanation for the following two reasons. (i) Samples 1 and 2 of the fish solubles were prepared from the same starting material, but of the two, sample 2 had been I20 TABLE I1 , 8 8’ -- - -----(y’ - COMPARISON OF THE CHICK ASSAY OF THE ANIMAL PROTEIN FACTOR WITH MICROBIOLOGICAL RESULTS Chick assay Microbiological assay r A - I A \ Sample Result T.F.L.at P = 0.95 Result, Organism pg of vitamin B,, per g pg of vitamin BIZ Per g 0.5 Euglena: gvacibis Fish solubles . . .. 1 1-17 0-99 to 1.37 2 0.30 0.16 to 0.46 0.4 97 3 1.63 1-10 to 2.60 0-7 17 Whale solubles . . . . 1 (a) 0-14 0.03 to 0.27 0.04 0 (b) 0.71 0.22 to 2.11 2 0.43 0-03 to 0.96 0.08 Lactobacillus leachmannii Whalemeat extract . . 0-38 0.110 to 0.61 0.1 Euglena gracilis Fuller’s earth adsorbates 1 120.5 102 to 142 about 60 Lactobacillus.. leichmannti 65 20 to 134 about 37 17 0 I Log Dose Interval Fig. 1. Assay of whale solubles.Doses of standard: 0.5, 1.0 and 2.0 pg per 100 g of diet. Response to standard - Response to whale solubles - - - - which agrees fairly well with the microbiological finding, was obtained with chicks from freshly-depleted hens ; the second, much higher, result was obtained in a simultaneous assay with chicks from hens that had received the all-vegetable diet for at least 12 months. The dose-response curves for these two assays are shown in Fig. 1. We suggest that the freshly- depleted hens, and consequently their chicks, were deficient only in vitamin B,,, but that the hens that had been longer on the diet had become d.epleted of other components of the animal protein factor as well, and that their chicks responded also to these other growth factors in the fish solubles.EXPERIENCE WITH CRYSTALLINE VITAMIN B,, Owing to the shortage of crystalline vitamin B,, it was impossible to use it regularly as a standard in chick assays and a microbiologically-standardised preparation of ExamenMarch, 19511 VITAMIN B12 AND THE ANIMAL PROTEIN FACTOR 149 was used instead. When, by courtesy of Glaxo Laboratories Ltd., a small quantity of the crystalline vitamin became available to us, we made several attempts to compare it with the potency of our standard preparation of Examen by chick assay. These attempts in- variably failed, as the response of the birds to graded doses of the crystalline material, and occasionally of the Examen, did not pass the test for linearity. The responses were haphazard and showed no particular trend, so it was unlikely that the non-linearity was due to instability of the vitamin in the diets.This lack of linearity has not been encountered with crude materials, although it sometimes happened with Examen. Similar difficulties were experienced with assays of crystalline vitamin B12c, also obtained through the courtesy of Glaxo Laboratories Ltd. I t may be that other factors of the animal protein factor complex, present in fish solubles and possibly to a slight extent in Examen, are necessary for the proper utilisation of vitamin B12. I t is also possible that vitamin B!, alone, without the rest of the animal protein factor complex, upsets the balance of the intestinal micro-organisms and interferes with their contribution to the nutrition of the host. DISCUSSION OF RESULTS This comparison of values found by chick and microbiological assay on crude materials containing vitamin B,, shows very clearly that the two techniques are not measuring the same thing.Although it is probable (but by no means certain) that the microbiological methods measure only vitamin BI2, it is evident that values found by the chick tests as used by us represent other growth factors as well. This marked difference in results is particularly interesting with assays using Euglena gracilis, which is believed to be more specific than Lactobacillus Zeichmannii in its response to vitamin BI2. Considerable evidence for the complex nature of the animal protein factor has already been given by Stokstad, Jukes, Pierce, Page and FranklinJ7 by Carlson, Miller, Peeler, Norris and Hewer8 and by Menge, Combs and S h ~ r b , ~ who demonstrated the necessity for other members of the animal protein factor complex in animal nutrition, but as yet little is known of the importance, if any, of these factors in human pernicious anaemia.In work concerning vitamin B,, and its allied vitamins, therefore, it will only be discovered by careful comparison with clinical results whether the microbiological test or the chick test gives the truer assessment of anti-pernicious- anaemia activity, although at present the chick assay is apparently the only means of measuring the whole animal protein factor complex. We are very much indebted to Glaxo Laboratories Ltd. for gifts of crystalline vitamin B,, and standardised Examen, to Dr.J. A. Lovern of the Torry Research Institute, Aberdeen, for the samples of fish and whale products, and to Messrs. D. J. G. Black and J. Getty of the Poultry Department of Reading University for the supply of specially bred chicks for this work. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. Coates, M. E., Harrison, G. F., and Kon, S. K., Biochem. J., 1950, 46, vii. Black, D. J. G., Getty, J., Coates, M. E., Harrison, G. F., and Kon, S. K., Ibid., 1950, 46, viii. Rubin, M., Bird, H. R., and Rothchild, I., Poult. Sci., 1946, 25, 526. Bird, H. R., Rubin, M., Whitson, D., and Haynes, S. K., Ibid., 1946, 25, 285. Coates, M. E., Kon, S. K., and Shepheard, E. E., Brit. J . Nutvit., 1950, 4, 203. British Standards Institution, 1940, Specijication No. 91 1. Stokstad, E.L. R., Jukes, T., Pierce, J., Page, A. C., and Franklin, A. L., J . Bzol. Chem., 1949, Carlson, C. W., Miller, R. F., Peeler, H. T., Norris, L. C.. and Heuser, G. F., Poult. Sci., 1949, Menge, H., Combs, G. F., and Shorb, M. S., Ibid., 1949, 28, 776. 180, 647. 28, 750. THE NATIONAL INSTITUTE FOR RESEARCH IN DAIRYING UNIVERSITY OF READING DISCUSSION DR. T. BARTON-MANN enquired whether Miss Coates had had any trouble, when depleting parent Did Miss Coates collect eggs for incubation over a period from these depleted His own experience showed that depletion of parents led to egg-eating and that collecting stock, with egg-eating. parents? eggs over a period resulted in very great unevenness in growth of chicks.150 LARKIN AND STUCKEY: SOME OBSERVATIONS ON THE p o l . 76 He also referred to the work of Miller and Groschke,’ who had abandoned the depletion of parents After incubation of the eggs the chicks were fed a depleted diet for 14 days MISS COATES replied that she had not herself encountered egg-eating b u t had occasionally had trouble In her laboratory eggs were not individually marked, so that Parent birds appeared to be thoroughly She agreed that culling the chicks after a preliminary depletion period might and fed them a normal ration. and then only those chicks with weights ranging from 75 t o 90 g were selected for assay purposes. with cannibalism in some groups of hens. growth of the chicks could not be related to the age of the egg. depleted after 6 t o 8 weeks. be advactageous, but had not had enough birds available to do so. REFERENCE TO DISCUSSION 1, Miller, D. C., and Groschke. A. C., Quad. BidI. Illichigan State Colf., 1950, 32, 279.