STUDIES IN CALCIFICATION.

Transcription

1 STUDIES IN CALCIFICATION. III. A QUANTITATIVE STUDY OF THE EQUILIBRIA CONCERNED WITH THE CALCIFICATION OF BONE. BY L. EMMETT HOLT, JR. (From the Department of Pediatrics, the Johns Hopkins Unioersity, Baltimore.) (Received for publication, April 22, 192.) In discussing the problem of bone calcification one must bear in mind the anatomical relations between the bone matrix, the osteoblasts, and the blood capillaries. The osteoblasts may be considered as resting in tissue spaces which are in contact with the amorphous bone matrix on the one hand, and with the walls of the blood vessels on the other. The bone matrix consists of a protein substance, ossein or collagen, which constitutes about onethird of its mass. The remaining twothirds consist of inorganic matter: calcium, phosphorus, carbonate, and magnesium, with perhaps small traces of alkali salts. Only the calcium and phosphorus are present in relatively large amounts, and Dhe ratio in which these elements exist is almost exactly that required to form tertiary calcium phosphate Ca(PO&. Thus it is usually stated that about 8 per cent of the inorganic matter of bone consists of this salt, about 14 per cent is calcium carbonate, and the remainder magnesium carbonate and various alkali salts. There have been two chief theories about t he product ion of the calcified bony matrix. Some authors claim that both the collagen and the lime salts are secreted by the osteoblasts. Others, however, while admitting the likelihood that the collagen is produced by the osteoblasts, believe that the lime salts are laid down in the matrixeither separately or as a double saltby a simple process of precipitation. In either case it is quite obvious that the inorganic salts must ultimately be derived from the blood stream, and it is consequently not surprising to find, as Howland and Kramer (1) have demonstrated in infants, that 70

2 80 Studies in Calcification. III when either the calcium or the phosphorus or both of these elements are sufficiently diminished in the blood serum, a condition of impaired calcificationactive ricketsis always present. It is not our purpose to discuss the evidence for or against these two theories; however, it may perhaps be worth while to point out that, in the complete absence of any data on the solubility of CaC0 and Ca(PO& in biological fluids, to postulate a process of precipitation is at best but a shrewd surmise. The ultimate proof of such a theory must rest until it can be shown that calcification of bone proceeds only when the ion products of the two insoluble salts in the tissue fluid exceed the solubility product constant for these salts in that particular medium. At present it seems quite impractical to investigate the composition of the fluid in contact with bone matrix. We have therefore to fall back on the blood serum as offering the nearest approach to this. To correlate the solubility data and ion products in serum with the process of calcification would at least offer suggestive, if not conclusive, evidence for the deposition of these bone salts by precipitation. The present study was confined to that salt which constitutes the major part of the inorganic matter of bonetertiary calcium phosphate. The product of the ions which constitute this salt, [Ca++1 X [PO: 12, was calculated from data of Howland and Kramer (1, 2) on the blood serum of infants and rats with active and healing rickets. Some normal cases have been included for the sake of comparison. The object in view was to ascertain the relation between the ion products and the solubility product constant for Ca(PO& in blood serum when calcification was proceeding normally or when it was impaired. The ion products were calculated as in Paper I. The calcium was assumed to be completely dissociated. [POT ] was calculated from the equation derived in Paper I: [PO?] = [PI K1 Kz KS [H+] + [H+l K, f [H+lK& t KIKZKJ Where [PI = the molar concentration of total phosphorus. KI, K2, and K = the first, second, and third dissociation constants of HsPO,.

3 L. Emmett Holt, Jr. 81 [H+l = hydrogen ion concentration. This was assumed to be p1i = 7. in all cases, as it has never been shown to deviate from the normal in rickets. The ion products thus calculated are stoichiometric ion products, and are, therefore, entirely comparable to the stoichio TABLE I. Ion Products in Active and Healing Rickets and Normal Infants. d mff. ; cc T d t 2.: 2.( 2.4! 2.4: 2.4: ( : 1.: 2.4: 1.4! 2.6; c!& _ i i 1 1 I: ic 72x7. Per 100 Xl% cc s i q:::; ( : :.2 1.6t.8 1.2: ;. 1.7; : ,0.26 CI X ;;;: 6 L mg. per AZ 100 x 100 CC $ x jo i : ) $ co z X ) 2 _ : : : : : : 1.2 1, , C I C I C.9 I C I E 2.70 I ( j ! ( 1.2: il Z.8 2.6!2. 2.6! ! ~ , : 0.2.4! L 1 Healing I rickets. I I I I I Normal. Healing Normal. Active rickets. rickets. I I I I metric solubility product constant evaulated in Paper I. Results for children and for rats are given in Tables I and II. 1 A few of these cases have been taken from unpublished data of HOWland and Kramer.

4 82 Studies in Calcification. III Comment. Tables I and II show clearly that the ion product in healing rickets, or in the absence of ricket.s, is always definitely larger than when the disease is active. When the disease is healing or TABLE Ion Products in Ache and Healing Rickets and Normal Rats. II. Rat No mg. mg. mg. Per Per 100 Xl% Fi Xl% zoo x 1: cc. cc. cc Normal ; Healing rickets, : : S Normal : Healing rickets ! : Active rickets ; \ absent the ion product is usually greater than 8.0 X 1O2 (p product less than 24.10), while in active rickets the product is usually below this figure (p product greater than 24.10). The dividing line is not, however, an absolutely sharp one.

5 L. Emmett Holt, Jr. 8 When, however, one compares these products with the solubility product constant for Ca(PO& in blood serum, which was found to be about 1 X 1O28 (pk,.,. = 26.0), it is quite clear that even in active rickets the product is considerably greater than that required to precipitate tertiary calcium phosphate. At first glance it might seem surprising that ion products in excess of the solubility product constant are found associated with a condition in which the deposition of Ca(P04)2 is so clearly at fault. It was, however, pointed out in the preceding paper of this series that the precipitation of tertiary calcium phosphate is a comparatively slow process. Although this salt is precipitated with moderate rapidity when the ion product is greatly in excess of the solubility product constant, precipitation proceeds with great slowness when the ion product is only slightly in excess of the value of the constant. The rate of precipitation depends on how much the value of the solubility product constant is exceeded. Thus it would seem quite possible that even in active rickets calcification is proceeding, but at a rate so slow that it is not ordinarily detected. Certain observations recently made by Shipley2 on the bones of rats which have been kept for a long time on McCollum s () Diet 14 offer very suggestive evidence of a slow gradual deposition of Iime salts, even during active rickets. These bones have a broad rachitic metaphysis. During the first 6 weeks of the condition, which may be regarded as the acute stage of the disease, this part of the bone is free from calcium. As the condition becomes chronic, however, the metaphysis not infrequently becomes peppered throughout with fine deposits of lime salts. Thus there is histological as well as chemical evidence for believing that even in active rickets calcification is proceeding, but at an extraordinarily slow rate. When the deposition of Cas(PO& is retarded beyond a certain point, the growth of new bone exceeds it in rapidity, and an inadequately calcified tissue, osteoid tissue, is formed which has been taken as pathognomonic of rickets. We have then a new conception of the immediate cause of the * Shipley, P. G., personal communication.

6 84 Studies in Calcification. III bone changes in rickets, and we can define rickets: not as a state in which the concentrations of calcium and phosphate are so low that Ca(PO& cannot be precipitated, but as a state in which, as a result of lowered ion concentrations, Caa(PO& is deposited so slowly that new bone production exceeds it in rapidity, and consequently uncalcified bone or osteoid tissue is produced. One can thus readily understand the fact that the serum of normal adults contains concentrations of calcium and phosphate similar to those found in active rickets in infants. In adults the growth of bone is either at a standstill or is exceedingly slow; consequent,ly an exceedingly slow rate of deposition of Ca(PO& is required to keep pace with it, and no pathological changes occur. It is only under special conditions of strain, such as occur in uniting fractures when rapid calcification is demanded, that the individuals with the lowest concentration experience dieicuhy (4). In the light of these observations it may be well to consider the significance in rickets of the product of the calcium X total phosphorus concentrations when expressed in mg. per 100 cc. This product was first suggested by HowIand and Kramer (1) in 1922 as an empirical criterion of the activity of a rachitic process, and it has been widely and quite justifiably used as such during the past few years. When, however, one compares this product with the stoichiometric ion product [Ca++J X [PO: I2 in Tables I and II, it is apparent that these two functions are not parallel. In general it holds true that a high ion product is associated with a high empirical Ca X P product and vice versa, and hence the empirical product remains a rough guide to the true ion concentration and to the activity of the rachitic process. It is, however, a less accurate guide than the stoichiometric ion product for several reasons, the chief of which is that the empirical product does not give enough weight to the calcium concentration. Thus in Table II it will be noticed in those cases of active rickets associated with low calcium that the empirical product Ca X P is only slightly reduced, and remains between 0 and 40, the range which has been considered an intermediate range, sometimes associated with active and sometimes with healing rickets. In all these cases, however, the ion product is reduced to a much

7 L. Emmett Holt, Jr. 8 greater extent, well beyond the doubtful range, which is in harmony with the observation that rats with such coneentrat,ions exhibit signs of active rickets with great regularity. Table I shows this same variation in a somewhat less striking manner. The stoichiometric ion product has the additional advantage that physiological and pathological changes in ph are entirely accounted for. As can be seen from Table III comparatively small variations of ph produce considerable change in that fraction of the total phosphorus ionized as PO,. The chief objection to the use of the ion product as a criterion of the activity of rickets is the inconvenience of making the TABLE PH [PO, ] when [PI = I. p [PO, ] when [P] =l x 10e x 10G x 10e x 10S X IO X 1O x 10e X 1O X lo@ x 10c X 1O X lo X 1O 4.78 calculation. The difficulty is largely due to the habit of recording concentrations of calcium and phosphorus in mg. per 100 cc., units without any physicochemical significance. Were such concentrations recorded in terms of mols or, better still (following the custom so well established for hydrogen ions), as negative logarithms of molar concentrations: p[ca++] and p[p], such calculations would then become exceedingly simple. PW,l which can be read from the third column of Table III could easily be obtained from p[p] by the addition of a constant. Then by adding p[ca] and 2p[P@TJ the negative logarithm of the ion III.

8 86 Studies in Calcification. III product, p product, is directly obtained. A sample calculation is given below: Ca = 2. X 10d molar (= 10 mg., per 100 cc.) log [Cal =.40 p [Cal = 2.60 P = 1. X 10d molar (= 4 mg. per 100 cc.) log [P] =.11 p [PI = 2.89 p[po:], at ph 7., = = 8.10 p Product = ( X 2.60) + (2 X 8.10) = From the data which we have presented it seems clear that the degree of supersaturation is a factor of the greatest importance in governing the deposition of Ca(PO& from fluids containing this salt in solution. The reduced degree of supersaturation of the blood serum in active rickets is found so consistently and is of such a magnitude that it is difficult to escape the belief that this is the immediate factor which hinders the deposition of Caa(PO& in the bones in rickets. On the other hand, it should be emphasized that this is not necessarily the only factor influencing the deposition of this salt. In the preceding paper we have referred to the work of Marc () who has shown that a great variety of substances may affect the rate of deposition from supersaturated solutions. Some substances act as catalysts and increase the rate of deposition, while others may act as poisons and delay the process very greatly. It seems quite possible that some such phenomena may explain why Ca8(POJ2 j.s SO readily deposited in certain tissues, and not at all in others. SUMMARY. 1. The ion product for tertiary calcium phosphate, [Ca++1 x [PO4 12, in the blood serum is smaller in active rickets than when the disease is absent or healing. 2. Even in active rickets this ion product is greater than that required to precipitate Ca(PO&.. It is suggested that calcification is not completely arrested in rickets, but that it is so retarded that bone growth exceeds it in rapidity. Thus the inadequately calcified osteoid tissue, SO characteristic *of rickets, is produced.

9 L. Emmett Holt, Jr. 4. The ion product, [Ca++1 X [P0,]2, is a more accurate criterion of the rate of deposition of Cas(PO& and hence of the activity of rickets than is the empirical Ca X P product. BIBLIOGRAPHY. 1. Howland, J., and Kramer, B., Tr. Am. Ped. Sot., 1922, xxxiv, 204; Monatschr. Kinderheilk., 192, xxv, Kramer, B., and Howland, J., Bull. Johns Hopkins Hosp., 1922, xxxiii, 1.. McCollum, E. V., Simmonds, N., Shipley, P. G., and Park, E. A., J. Biol. Chem., 1921, xlvii, Petersen, H. A., Bull. Johns Hopkins Hosp., 1924, xxxv, 78.. Marc, R., 2. physilc. Chem., 1908, Ixi, 8; 1909, Ixvii, 470; 1910, Ixxiii, 68. Marc, R., and Wenk, W., 2. physik. Chem., , Ixviii, 104.

10 STUDIES IN CALCIFICATION: III. A QUANTITATIVE STUDY OF THE EQUILIBRIA CONCERNED WITH THE CALCIFICATION OF BONE L. Emmett Holt, Jr. J. Biol. Chem. 192, 64:7987. Access the most updated version of this article at Alerts: When this article is cited When a correction for this article is posted Click here to choose from all of JBC's alerts This article cites 0 references, 0 of which can be accessed free at ml#reflist1