474 CLAYTON AND PRUNTY: ASCORBIC ACID DEPLETION METHOD [Vol. 76 The Ascorbic Acid Depletion Method for the Bio-Assay of Adrenocorticotrophic Hormone, and Preliminary Observations on the Use of Inhibition of Tissue Repair BY BARBARA E. CLAYTON AND F. T. G. PRUNTY (Presented at the meeting of the Biological Methods Group 0s Tuesday, October 24th, 1950) The adrenal ascorbic acid depletion method for the assay of adreno- corticotrophic hormone has been examined. Certain modifications -are given and some resuIts obtained by the authors. For the colony of rats used, A, the index of precision, had the value 0.2911 -4 0.044. ACTH inhibits the formation of granu'lation tissue in response to trauma. This fact can be used as the basis of its assay on mice, in which a quanta1 response is measured.SAYERS, Sayers and Woodburyl in 1948 published a method for the assay of adrenocortico- trophic hormone (ACTH) that depended on depletion of ascorbic acid in the adrenals of hypophysectomised rats. During the last six months this method has been used in our laboratory with certain modifications ; some results thereby obtained are described below. ADREN-4L ASCORBIC ACID DEPLETION TECHNIQUE- Wistar male rats weighing 75 to 130g were brought into the laboratory at least three days before use; they were fed ad lib. on Rowett diet 46* before operation and afterwards on bread and milk. Attempts to maintain their environment a t 70" to 75" C were not always successful owing to difficulties with the heating system. When sudden marked fluctuations occurred the technique of hypophysectomy and adrenalectomy were made more difficult, as blood clotting was sometimes impaired.Hypophysectomy was performed under ether anaesthesia. The freshness of the ether' was found to be most important if respiratory difficulties were to be avoided. The parapharyngeal approach was used, and, largely by feel, the initial hole in the skull was made with a fine pair of forceps and enlarged with a thick blunt probe, the gland being then removed by suction. A tracheotomy tube was not necessary after some experience had been obtained and is now no longer used. At autopsy the sella was examined for completeness of hypophysectomy. The solution being assayed was injected into the exposed left external iliac vein and one hour later the right adrenal was removed. Excised adrenals were cleaned, weighed to the nearest 0.1 mg on an analytical balance and transferred to trichloro-acetic acid.Ascorbic acid was determined by the dinitrophenylhydrazine method of Roe and Kuethner.2 New bottles of trichloro-acetic acid should be tested before use, as some sample:; give a precipitate during the final colour reaction. Soluble preparations of ACTH were injected in normal saline, less soluble ones were taken up in glacial acetic acid and diluted to 0.02 to 0.01 N , as suggested by Morris? The solutions were allowed to stand during the tests in a beaker of water over a tray of ice. The left adrenal was removed under ether anaesthesia 18 to 21 hours later. RESULTS The Sprague - Dawley rats in our colony showed great variation in the sizes of adrenal glands, the difference in an individual rat occasionally amounting to as much as 20 mg.In our Wistar rats the adrenal glands usually weighed 7 to 12 mg. In 90 per cent. of the rats the difference in weight between the two adrenals was less than 2 mg and in 23 per cent, the right gland was larger The strain of rat used for assay appears to be most important. * Obtained from Heygate and Sons.August, 19511 FOR THE BIO-ASSAY OF ACTH 4754 than the left. The ascorbic acid concentrations in the adrenals of hypophysectomised but otherwise untreated rats are given in Table I. Some of the percentage differences are much larger than those observed by Sayers et aZ.l Where this occurs it is due to the considerably higher concentration of ascorbic acid in the second gland (right) ; one example of this is given in Table I.TABLE I A RANDOM SELECTION OF THE ASCORBIC ACID CONCENTRATION OF THE ADRENALS OF UNTREATED HYPOPHYSECTOMISED WISTAR RATS Adrenal ascorbic acid, mg per 100 g Left Right A I 3 438.2 404.7 366.8 494.8 367.0 440.0 666.0 600.0 680.8 644.2 715-0 650-0 420.4 402.9 426.6 500.0 346.0 454.6 568.4 583.0 642.6 615.2 721-8 623.0 Difference in mg (left minus right) + 17.8 + 1.8 - 60.8 - 5.2 + 21.0 - 14.6 - 3.4 + 17.0 + 38-2 + 29.0 + 27.0 - 6.8 Difference per cent. of left adrenal 4-0 0.4 16-7 1-0 5.7 3.3 0.6 2-8 6.6 4.6 0.9 4.1 Average: 4-1 Standard Error: & 1-62 An analysis of variance of body weight and initial concentration of ascorbic acid showed no significant effect on the potency of ACTH determined.The precision of the method is given in Table 11. The standard error of the slope is approximately 60, and the standard deviation of 10 slopes is approximately 56, so no evidence of heterogeneity of slopes has yet occurred. The index of precision, A, is 0.291 2 0.044, compared with 0.176 & 0.016 obtained by Sayers et aZ.l The method is thus less precise with this strain of rats, and about 24 to 3 times as many animals are needed by us to give the same degree of precision as these workers got. TABLE I1 PRECISION OF THE ASCORBIC ACID DEPLETION METHOD Month June June June September September October July Preparation 23 200 : 365 203 : 935 J 10602 British Organon La-1-A J20507 Number of rats used per preparation 11 13 11 12 10 12 12 s 62.7 38.0 40.7 71.9 61.3 57.2 83.6 b 197-8 160-3 128-7 263.2 272.2 161.4 262.3 h 0.317 0.237 0.316 0.273 0.225 0.354 0.318 Mean = 0.291 Standard error = +0.044 Very occasionally it has been found in an otherwise satisfactory assay that one animal failed to respond.Several times, however, when assays have been carried out, it has been found that the responses of the individual animals have shown great variations independent of the dosage of ACTH given; indeed, nearly half may have failed to respond at all. This has occurred both with the old.er, poorly soluble, preparations and with the newer soluble ones. We have had correspondence with Dr. Hayes and his colleagues of the Armour Company about the results obtained by us on one of these poorly soluble preparations. They noted that the discrepancies we have found were within the limits of variation obtained by them, by both the Sayers methodl and the Munson m0dification.l They occasionally appear to have similar occurrences in large groups (40 to 50) of animals on any one day, and they ascribe this to biological variation.Nevertheless, for most determinations they reported h = 0.25. We believe that such aberrant results occur when we periodically use the offspring of certain parents in our colony, and this is being further investigated.476 CLAYTON AND PRUNTY: ASCORBIC ACID DEPLETION METHOD [Vol. 76 INHIBITION OF TISSUE REPAIR During the treatment of patients with ACTH: and cortisone it was noticed by one of us that healing of biopsy wounds was inhibited. This happening independently prompted Howes, Plotz, Meyer and Blunt3 to study the effect of cortisone on granulation tissue formation in the ears of rabbits.They found that 12.5mg of cortisone acetate per day per rabbit for 8 to 11 days produced inhibition. It seemed as though the inhibition of healing might be used for the assay of ACTH if smaller laboratory animals, preferably mice, proved suitable. At first some difficulty was experienced in consistently obtaining good granulation tissue in mice. Among methods tried were burning and the injection of turpentine and formalin, but the following method was finally devised. They failed to obtain good responses in rats. TECHNIQUE- Male albino mice weighing 12 to 18 g were maintained on a diet of Thomson cubes and water. Operation was performed under ether, and an aseptic technique was adopted as far as possible.Fur was removed from the anterior abdominal wall with scissors, a piece of skin about 2 to 3 mm in diameter in the midline of the anterior part of the ventral surface of the abdomen was lifted with fine forceps and clipped off with sharp scissors. The wound was covered with three thicknesses of sterile Vaseline gauze, over which was placed the cap of a bottle about 15mm in diameter. The dressing was held in place with adhesive tape, and the cap prevented any pressure on the wound and any contamination with urine and faeces. Care and attention to detail are essential if consistent controls are to be obtained. Twenty-eight and a half hours later the mouse was killed, and the dressing was carefully removed.The ACTH was given in normal saline as 25 subcutaneous injections, each of 0.1 ml. Injections were given one hour and half an hour before the ulcer was made in the morning, immediately after making it and then hourly, except between 11.30 p.m. and 7.30 a.m. when only two injections were given. With a sufficient dose of ACTH complete inhibition of healing is obtained. By this technique well-defined granulation tissue was consistently obtained. APPEARANCES OF THE TISSUES- two feet beneath and one foot to the side of an electric light. ances of the ulcers. Two to three minutes after removal of the dressing, the ulcers were examined about Figs. 1 and 2 show the appear- The points to be noted are summarised as follows- Control 1. 2. 3. 4. healing ulcer Inhibited ulcer Marked hyperaemia.I.. No hyperaemia. Sloping margins. 2. Sharp, clear-cut margins; this appear- ance is emphasised by the shadow in the third lesion (Fig. 2). Rough, thick floor. 3. Smooth, thin floor. The original vessels in the floor are 4. The original vessels in the floor are obscured by new tissue. still visible. Histological examination of most ulcers was at first carried out. The ulcer and a large piece of the surrounding skin were removed and placed flat on a thick piece of blotting paper. After 1 minute the skin and paper were immersed in formalin. The paper was removed twenty-four hours later, and the skin was trimmed. The ulcers were sectioned in the trans- verse diameter and stained with haemotoxylin and eosin. The histological appearances are shown in Figs.3, 4 and 5. The same differences occur as are seen macroscopically, and it is also found that the thick floor of the control ulcer consists of granulation tissue with marked polymorph infiltration, proliferating fibroblasts and numerous capillaries, while the floor of the inhibited ulcer is very thin and contains only a few polymorphs, some flattened fat cells and no capillaries. ASSAY METHOD In the assays, a preparation of ACTH, Armour 7911, was used as a laboratory standard. Doses of this, ranging between 33 and 62 pg, were given to groups of 10 mice. The percentageFig. 1 . Left, control healing nlccr; rifiht, inhibited ulcer Fig. 2.- I .eft, control healing ulcer; centre, partly inhibited ulcer; right, completely inliiliited ulcerFig. 3 .Section cut immediately after the ulcer was made; stained with H. a i d 17. 1 x 30 Fig. 4, Section of a healing ulcer; stained 14. and 1:. 1 x 30August, 19511 FOR THE BIO-ASSAY OF ACTH 477 healed in a group was plotted against the logarithm of the dose and gave a sigmoid curve, as shown in Fig. 6. A probit line was fitted, and it was calculated that the approximate 95 per cent. limits of error would be 78 to 128 for 10 animals on each of two preparations, 84 to 119 for 20 animals, and 90 to 112 for 50 animals. The results of assays on two preparations of ACTH, one from Organon Ltd. and one made by Dr. M. Reiss, are shown in Table 111. Thus the potency ratio of Organon ACTH in terms of standard 7911 is estimated at 0-033 with approximate limits of error, 0.071 and 20 - 10’ 1 I I LOG DOSE 1.6 I .I 1.8 I.! I I DOSE 33.7pg 61.3pg 7.0 6-5 6.0 7 5.5 g a 5.0 4.5 10 4.0 C 2 CID U 3.5 3.0 Fig.6. Dosage - healing curve with adrenocorticotrophic hormone (7911) : 7 to 10 mice per group 0.097. The potency ratio of Reiss ACTH in terms of standard 7911 is estimated at 0.265 with approximate 95 per cent. limits a t 0.228 and 0.309. This may be compared with estimates by the method of Sayers et aZ.l Owing to shortage of mice these figures were not all obtained on the same day; they are based on the assumption that the dose - response curve did not change in any respect over the period of assay. The test for differences between the three slopes shows no evidence that the slope of the probit line has changed during the period of assay. These preliminary observations appear to be encouraging.The method is simple and accurate for assaying preparations known to consist of ACTH, and may therefore be of value in controlling its production. The method lacks the specificity of that of Sayers et aL,l but, in view of its simplicity, may be useful for certain purposes. TABLE I11 RESULTS OF A MOUSE HEALING INHIBITION ASSAY Organon Reiss Dose in pg animals healed Dose in p g animals healed 418.0 4 75 135.0 4 100 469.0 4 75 151-5 4 50 526.0 4 50 169.4 4 50 590.5 4 50 190.6 4 25 Standard 79ll/Organon = 0.083 (Sayer’s method 0.070; Limits = 75 to 132 per cent.). Limits* (P = 0.95) = 0-071 to 0-097. Standard 7911/Reiss = 0.265 (Sayer’s method 0.234; Limits = 46 to 212 per cent.). Limits* (P = 0.95) = 0.228 to 0.309. L A 7 \ r \ Kumber of Per cent.Number of Per cent. * These limits are approximate 95 per cent. fiducial limits.478 VOGT : ASSAYS OF ADRENOCORTICAL HORMONES [Vol. 76 ACKNOWLEDGMENTS We are grateful for statistical assistance given us by Mr. P. Armitage of the M.R.C. Statistics Unit; for technical assistance from Miss J. Hammant in carrying out the assays; from Mr. A. E. Clark in histological work and from Miss J. Dewe in the preparation of the colour drawing. We acknowledge with thanks gifts of ACTH from Dr. J. R. Mote (Armour Company), Dr. W. J. Tindall (Organon) and Dr. Max Reiss (Bristol). REFERENCES 1. 2. 3. 4. Sayers, M. A., Sayers, G. S., and Woodbury, L..A., Endocrinology, 1948, 42, 379. Roe, J. H., and Kuethner, C. A., J . Biol. Chem., 1943, 147, 399. Moms, C. J. 0. R., personal communication. Howes, E. L., Plotz, C. M., Meyer, K., and Blunt, J. W., Proc. SOC. Exp. B i d . Med., 1949, DEPARTMENT OF CHEMICAL -PATHOLOGY 72, 718. ST. THOMAS’S HOSPITAL LONDON, S.E.1 DISCUSSION DR. PRUNTY commented that it was important to stress the remarks made by Dr. Overbeck in the discussion on tlie preceding paper, about the need for repeated injections of ACTH to produce any given response. They had noted in their laboratory that a patient being treated with 40mg of ACTH daily, divided into four doses of 10mg intramuscularly at 6-hourly intervals, gave about l/lOth of the bio- chemical response elicited by the continuous intravenous infusion of 40 mg of ACTH over a 24-hour period. In his paper, Dr. Morris stressed the need for hypophysectomy in any ACTH assay. We, too, held the same opinion until we obtained the results we have reported. We do not know of assays based on multiple injections of ACTH at short intervals. One wonders if this technique does not, in fact, produce a “biological hypophysectomy” in the animal, perhaps by direct action on the pituitary. If this were so, it would explain the constant results obtained in their experiments. This stresses the importance of continuous administration.