The effect of buffered phosphate solutions upon a thin layer of living vascular tissue in moat chambers introduced into the rabbit's ear.код для вставкиСкачать
THE EFFECT OF BUFFERED PHOSPHATE SOLUTIONS UPON A THIN LAYER O F LIVING, VASCULAR TISSUE I N MOAT CHAMBERS INTRODUCED INTO THE RABBIT'S EAR a RICHARD G. ABELL Awatom~alLaboratory, University of PenwyEvania Sahool of Medicine FOUR FIGURES That phosphate buffers produce toxic symptoms when introduced into the body in sufficient concentration is well known. The cause of this toxicity, however, is somewhat obscure, and has been variously attributed to the phosphate ions in combination with the pH of the solution in which they exist, to the sodium or potassium ions contained in the buffers rather than to the phosphate ions, and to other changes in the blood plasma and tissue fluid. The present technic offers the possibility of submitting a thin layer of living vascular tissue to the action of buffered phosphate solutions, and at the same time of observing with the high powers of the microscope the effect of such solutions upon the tissue. The reaction of the tissue to phosphate solutions having the same pH as the blood, and to others differing in pH from the blood, can thus be determined by direct observation. So also can the effect of solutions of sodium and potassium chloride containing the same concentration of sodium and potassium ions as the buffers, and of others con'This work was aided by a grant made to the department of anatomy at the University of Pennsylvania by the Rockefeller Foundation, under which the author held a fellowship. lBrief reference to some of the results obtained in this study waa made by AbeU and Clark ('33) at the 1933 meeting of the American Association of Anatomists. 61 52 RICHARD G. ABELL taining even greater concentrations of these ions. I n a similar manner, the effect upon the tissue of the osmotic pressure of the solutions can be determined. Many experiments concerning the toxicity of phosphate solutions are contained in the literature, but so f a r as the author is aware, with the exception of a preliminary report by Abell and Clark ( ’33), direct microscopic observations of the sort to be presented in this paper have not previously been described. MATERIAL AND METHODS Information concerning the construction, installation, and care of moat chambers may be secured from the original description of the chamber by Abell and Clark (’32). Briefly, such a chamber contains a very thin, transparent space called a ‘bay.’ Following introduction of the chamber into a hole cut through the ear of a rabbit, living tissue, continuous with the subcutaneous tissue of the ear, grows into the bay through two entrance holes at the proximal end. At its distal end the bay communicates with a reservoir, or ‘moat,’ into which chemical solutions may be placed and from which they may be subsequently removed, through two permanently installed access cannulae, Chemical solutions introduced into the moat pass by diffusion into the bay, and there interact with the living cells and tissue normally present during the vascularization of the chamber (compare fig. 1). Since the bay is transparent, and very thin, the effect of the solution upon the tissue can be observed with the high powers of the microscope. Buffered solutions of phosphate at pH 7.4 and 6.2 were used in the experiments to be described. The one at pH 7.4 was prepared by adding 39.50 cc. of 0.2 M NaOH to 50 cc. of 0.2 M KH,PO,, that at p H 6.2 by the addition of 8.60 cc. of 0.2 M NaOH to 50 cc. of 0.2 MKH,PO,, as described by Wilson (’28). When the hydrogen ion concentration of the phosphate solutions thus prepared differed from that desired-as shown by standard buffers, the pH of which had been determined electrometrically-such small corrections as necessary were PHOSPHATE SOLUTlONS A N D LITING TISSUES 53 made, bromthymol blue and phenol red being u s e d as indicators. The method of installation of the phosphate solutions within the moat and of subsequently removing them from it was the same as that described for soliitioiis of methyleiie blue by the author ( ’34). Before introclizctioii into the moat, the phosphate soliitions were sterilized by boiling. 7JpOli cooling, the pH was again measured. No corrections were necessary a t this time. Records of growth of blood vessels into the chaiiibers were made with a Leitz ‘clrawing eyepiece’ at 2- to M a y intervals during the period preceding tlie iiistallation of the phosphate solutions within the moats, a period wliich varied iii length, in all of the chambers but one, from 23 t o 30 days. Photomicrographs of the tissues in the bays were taken almost every day throughout the prc-introductorj- periods. Microscopic and photomicrographic records were lilicmise made €011o wing int r odu ct i on of the ph o spli at e solut iom s wit liiii t lie moats, i n one instance such records being continued for more than a year. The techiiic emploj-ed in the experiments to he described does not involve the introduction of a needle into the tissue itself. C’onseqnently the microscopic picture produced bp iiitroductioii of the phosphate solntioiis is free from the effect upon tlie tissue of direct meclianical manipulatioii. It is a real pleasure to express to Dr. Eliot R. Clark my deep appreciation for encouragement and advice given throughout the course of these experiments. OBSEBVA TTONS 1. Tntroduction into f fie moat of bicferpd piiospkate solnt i o m at pH 7.4. Phosphate solntions at this pH were intro- duced into the moats of three chambers, designated as chambers 7/30/31L, 1/23/32R, and 2/17/32L. I n each case the installation of the buffer within the moat was followed by certain definite reactions of the tissue. These reactions may, perhaps, best be described by referring to the individual experiments. T l I H A N A T O M I C I L RhCORD, VOL. 6 4 , KO. 1, AND STJPPLEMENT N O 1 54 RICHARD G . ABELL The phosphate solution MYIS introduced into the moat of chamber 2/17/32L 23 clays following the entrance of the first blood capillaries into the bay. During this 23-day period drawing ocular and photomicrographic records allow that the vessels advanced uninterruptedly toward the moat, the most distal capillaries being approximately 0.53 mm. from the moat at the time of iiistallation of the phosphate solution mithin it, (compare fig. 1,a). A drawing eypiece record showing the peripheral capillary pattern was completd just before iiitroduction of the solution into tlie moat. lmmediately following introduction of the buffer, the image of the peripheral Iressels was projected upon this record. No alteration of the position of the capillaries had occurred, and careful microscopic esamination showed no irritation of any of tlic tissue in the bay. Similar results were also secured in the case of the other chambers in these experiments. Alicroscopic observatioiis made 22 hours following the installation of the phosphate solution within the moat showed that the most distal of the capillaries in the bay had been injured for a distance of approximately 0.2 mm. Manj- erythrocytes and leukocytes were present in front of the new line of functional peripheral capillaries, and sticking of leukocytes to the enclothelium of the inost distal capillaries and to that of tlie veiiulcs and veins collecting blood from the periphery was extensive.3 &licroscopic examination showed that the diame ters of tlie most distal of tlic capillaries, arid of the venules collecting blood from these capillaries, were greater than hefore introduction of tlie bnffer. Oiie of the two large arterioles which snppliccl blood to the peripheral tissne was more widely distended than at ariy previous time, aiid the rate of hlood flow through it, and in general all of the vessels in the bay, was more rapid than before tlie introduction of the phosphate solution into the moat. It appears probable that the increase in Observations on the sticking of leukocytes t o the cndothelml wall h a w been recently presciited and discussed by E. R. and E. L. Clark ( ' 3 5 a and h). They produce evidence which indicates t h a t progressive grades of leukocyte sticking and emigration are associated with progressive changes in endothelium. PHOSPHATE SOLUTIONS h N D LIVING TISSUES 55 diameter of the arteriole proximally was due t o the effect of tlic phosphate solution upon its distal ciid, dilation at the periphery decreasing the resistance to blood Bow, thus allowing more blood to enter the arteriole. The passage of a greater amount of blood into the arteriole was evidently responsible for its proximal dilation, which was therefore probably secondary in nature, and not due to the direct effect of the phosphate solution, although this altcrnative caiinot be completely ruled out. Approximately 70 p in front of the most distal of the functional capillaries, and exteiidiiig from this point toward the , numbers of small moat for a distance of about 1 6 0 ~ large granules and irregular masses of granular material could he observed extending from one side of tlie bay to the other (compare fig. 1,b). Tm7o days later these granules and irregular masses were still present. At this time, the phosphate solution having becn lcft continuously within the moat, with free access to the tissue, thc most distal of the functional capillaries on tlie right side of the bay were 180 p farther from the moat than on tlie day following introduction of the buff er-on the opposite side of the bay 46 1-1. I n the space thus left betweeii the masses of granular material preseiit the day after introduction of the phosphate solution, and the iiem position of the distal vessels in the bay, fresh, graiiular masses of precipitate could be observed. During the succeediiig 3 days gradual precipitate formation continued, the masses of precipitate increasing in size by fusion. It was observed that at this time tlie precipitate was of glistening appearance. Precipitate formation continued until the time of rcmoval of the phosphate solution from the moat, 20 days after its installation within it (possibly to some extent even after this time), and resultcd in the formation of three roughly distinguishable TOWS of precipitate, corresponcling to different stages in the witltdrawal of the peripheral circiilatioii away from the source of phosphate solution, i.e., the moat (compare figs. 1,b and 2). This recession of thc peripheral circulation 56 XICHARD G. ABELL was evidently due iii part t o actual iiijnry, possibly destruction, of thc most advanced capillaries bj- the phosphate solution, hut probably also in part to mecliaiiical tension exerted upon the peripheral tissue by migration of a part of tlie proximal tissue toward the eiitraiice holes, particiilarly on the right side of the bap (compare fig.2, c). Xasses of precipitate formed in the bay of chamber 2/17/32L are shown in figures 1,b, 2, aiid 3. The photomicrographs shown in figures 1,b a i d 2 were taken during the process of precipitate formation. Figure 3 is a photomicrograph taken 65 days following tlie removal of tlie phosphate solution from the moat. Dnriiig the iiitcrrd betwcen tlic time of removal of the phosphate solution and tlic time this last photomicrograph ix-ils taken, tlie precipitate Iiacl undergone practically 110 change in appearance, aiid formed a harrier which prevented forward growth of the ressels. The precipitate formed in this cliamber persisted as a burrier which blood capillaries did not penetrate tliroughont the entire survival period of the chamber-more thaii 13 months following the time of precipitate f o r r n ~ t i o i i - a l t h o ~ isprout ~ ~ ~ s from maiiy of the peripheral capillary loops grew up t o this barrier. Tissue in contact with the precipitate garc no evidence of being irritated by it. Owing to ihc shallowiiess of the hay of chamber 2/17/33L (depth approximately 40 p), the precipitate could be observed with tlie greatest clearness. Two forms could be disFig. 1 a, A ~~hotoinic.rojira~~h, showing blood vessels and other strw turcs in the bay of chamber 2/17/32L, just before introduction of a phosphate buffer (pII 7.4) into the moat. Eraekets E are placed 1 x 1 0 ~tlie edges of the two entrance holes through which tlic new tissue inIaded the bay. The bay extends from the proxiinal line D t o the distal line HB. The m o a t communicates with t h r bay a t t h e line BB. The arrowq pointing u p ~ a r dare placed next a1terioles, ,4; tllovc pointing downnard next veins, V. X 19.3. b, h photomicrograph, showing blood resscls and other structures in the bay of chamber 2 / l i / 3 2 L , 22 houri following installation of the phosphate buffer ( p I I 7.4) within the moat. The extcnt of injury of the tissue a t this t i m e can b e seen bp comparing the position of the peripheral capillary loops i n b, x i t h those in a. The dark masses just beyoiid the most advanced capillaries, which form two more or less distinct rows, x and y, are composed chiefly of granulai material prwipitatcd by the phosphate huffcr, and of emigrated Icukocytes. X 19.3. PHOSPEIBTE SOLTJTIONS ANT) LIVING TISSUES 57 58 EICIiARU G. ISELL tinguished. Oiie consisted of irregular, glistening, granular, amorphous bodies, clumped together t o form larger masses. The other mas composed of rounded structures, glassy in appearance and showing conceiitric rings, as though laid do.lr.11 in layers. Iiicomplete fusion of these rounded, spherule-like bodies had resulted in the formation of double and multiple structures (fig. 3). Results similar to those described for chamber 2/17/32L were also secured with chamber 7/30/31L. In the case of this chamber, the phosphate solution was not iiitroducecl into the moat until 11months and 3 weeks after the appearance of the first vessels in the hay. The bay of this chamber was so thin at the end toward the moat, approximately oiily 18 p deep, Fig. 2 a, A photomicrograph, showing the distal tissue in the bay of chamber 3/17/32L and the precipitate being formed jiist in front of thc distal tissue. This photomicrograph was taken G days following installation of a phospliate buffer (pH 7.4) wthin the moat. C’ompare with figure 1,b, a photomicrograph taken 22 hours subsequent to introduction of t h r sanie buffer into the moat. X 23.3. h, -4 photomicrograph, sliolving tlic distal tissue in thc bay of chamlm 2/17/32L and the prec4ipitate being formed in front of the distal tirsue. This photornjcrograph was takcn 10 days followjng the installation of tlic phosphatc buffer mithin the moat. X 23.3. c, A photomicrograph, showing the distal tissuc i n the bay of chamber 0,/17/32L and the precipitate formed in front of the distal tissue b y reaction of the buffcr (pH 7.4) with some substance or suE,stanccs in the bay. This photomicrograph tyas taken 20 days following installation of the phosphate buffer (pH 7.4) writhin the moat. The letters x, y and z ifidicnte roughlv distinguishable rows of precipitate. The buffered phosphate solution was removed from t h e moat immediately after this photomicrogra~ihwas taken, and RingerLocke’s solution introduced in its place. x 23.3. Fig. 3 A pliotoinicmgraph, showing a t high magnification t h a t part of the precipitate and peripheral tissue contained in chamber 2/17/3211 which is within the rectangle in figure 2 , c. One of the spherule-like rnaaws of precipitate is designated by M. The positiori of sonre of the granular, auiorphous bodies is shown by N. This photomicrograph was taken 65 daTs after removal of the phosphate buffer, pH 7.4 at the time of its introduction, Prom the moat. The precipitate reniained undissolved by the tissue. Thirteen and one-half months after the time of formation of tlic precipitate thc chamber w a y removed froin the ear of the rabbit, and the nixsses of precipitate analyzed. They contained phosphorus. X 307. Fig. 4 A niass of graiiiilar, amorphous precipitate formed withiii the hay of chamber 1/23/32R, subsequent to introduction of a phosphate buffer (pH 6 2 ) into the moat. Living tissue has coniple+ely surrounded this mass of precipitatc, which later dissolred. This photomicrograph mas taken 23 days following the reninval of the buffer from the moat. X 307. PHOSPHATE SOLUTIOXS AND L I V I N G TISSGES 59 60 JXICHAIlD G . A4EELL that the blood capillaries were unable to approach closer to the moat than 1.37 mm. In spite of this extreme thinness of the bay, between the vessels and the moat, precipitate formatioii occurred. It was observed that precipitate formation was not limited to the region of injured tissue proximal to the original position of the peripheral capillary loops. Precipitate was also formed in the region just in front of these capillai-y loops. The distance beyond tlic original peripheral capillary loops to the line at which such precipitate has been obserrecl to form varies from approximately 75 to 230 p. This region is one which coiltailis sprouts from the distal capilla1.27 loops, many living white blood cells, and large numbers of fibroblasts. Some of these structures were injured, others possiblp destroyed, by the buffers during precipitate formation. It is not a t present l~iiowiiwhether any precipitate mas formed within uninjnrecl tissue either distal or proximal to the peripheral capillary loops. 2. I&roduc.tioiz i t i t n t h e woat of u liujf"ered phosphate solutio+z at pH 6.2. The phosphate solution at this pH was introduced into chamber 1/23,/32R, the bay of wliieli was approximately 100 p deep. Both drawing eyepiece and photomicrographic records show that the tissue advancccl uninterruptedly before introdiiction of the phosphate solution into tlie moat. At the time of introdidion of tlie phosphate solution, two main crests of vascular tissue, h and C, extended for a considerable distance in front of the djstal tissue in other parts of the bay. The most advanced capillary loops of the larger of the crests, crest A, %[-ere0.33 mm. from the moat a t the time of the introduction of tlie phosphate solution into the chamber, those between the crests 0.73 mm., while those adjacent to the smaller of the crests, C, were 1.22 mm. from the moat. Microscopic observations made 2 days following the installation of the buffer vvithiii the moat showed that some of the peripheral capillaries in the bay had been injured, others probably destroyed, during this time, this result being similar to that secured following introduction of the phosphate solu- PHOSPHATE SOLUTIONS AND L M N G TISSUES 61 tion at pH 7.4 into chamber 2/17/32L and 7/30/31R, the injury and destruction being, however, somewhat more extensive in places in the present case. Destruction of the capillaries of the crests, which had been closer to the moat than capillaries in other parts of the bay, was most extensive. The vessels of the crest that had been closest to the moat, crest A, had undergone disintegration for a distance of 1.086 mm. proximal to the tip of the crest. Similar destruction of vessels of crest C was observed. Between the crests, injury to the vessels was slight, while beside the base of the smaller of them, C, none of the vessels showed any visible effect of the phosphate solution, the peripheral capillaries beside the base of this crest having been farther from the moat than those in any other parts of the bay. Arterioles which supplied blood to the tissue of the crests were more widely distended than at any previous time, and the more peripheral of the capillary loops and the venules collecting blood from these loops were noticeably dilated. The circulation through the arterioles supplying the peripheral crests, the distal capillaries, and the venules and veins collecting blood from these capillaries was much more rapid than before the introduction of the buffer into the moat. I n contrast to the change in size of these arterioles, the arteriole which supplied the peripheral tissue beside the base of crest C, where no destruction of vessels had occurred, had undergone little or no change in diameter. Microscopic examination of the vessels showed extensive sticking of leukocytes to the endothelium of the peripheral capillaries and to that of the venules and veins collecting blood from these peripheral capillaries. As in the cases of introduction of the buffered phosphate solutions a t pH 7.4, large numbers of erythrocytes and leukocytes were present in front of the new line of functional peripheral capillaries, and for a distance of about 0.8 mm. proximal to this new line of functional capillaries many freshly emigrated leukocytes could be seen in the tissue outside of the vessels. 62 RICHARD 0. ABELL The chamber contained several lymphatic vessels, all of which were separated from the moat by several rows of intact blood capillaries. No injury to any of these lymphatics could be observed. The number of leukocytes in them was no greater than normal, and none contained any erythrocytes. Just in front of the most peripheral of the intact vessels of the crests, and between these crests, masses of irregular, granular, clumped precipitate could be observed. None of this precipitate appeared to be present in the form of spherules. Little or no precipitate could be seen in front of the vessels at the base of crest C, and none was observed in any of the tissue separated from the moat by circulating capillaries. The phosphate solution was removed from the moat 50 hours after being installed within it, and Ringer-Locke’s solution introduced in its place. The pH of the phosphate solution taken from the moat was measured, and was found to have remained unchanged while the solution was in the moat. Twenty-four hours after removal of the phosphate solution the rate of circulation had decreased considerably, and the sticking of leukocytes to the endothelium of the blood vessels was not as pronounced as on the preceding day. During the next 48 hours the rate of circulation and size of arterioles returned to the normal condition. At the end of this time sticking of ledocytes to the endothelium was no more pronounced than before the injection of the phosphate solution into the moat. Five days following the removal of the phosphate solution from the moat new sprouts could be seen growing from the peripheral capillaries, but during the succeeding 10 days the vessels of the crests grew forward only approximately 78 p. This is 27 p less than the average distance that the vessels advanced per diem during the 30-day period preceding the introduction of the buffer into the moat. I n contrast to this abnormally slow rate of forward growth, the peripheral capillaries beside crest C, in front of which little or no precipitate had been formed, and which had given little evidence of being PHOSPHATE SOLUTIONS AND LIVING TISSUES 63 affected by the phosphate solution, advanced 1.1mm. during the 15-day period following removal of the buffer from the moat. At the end of this time they had grown beyond the peripheral capillaries in the other regions of the bay distal to which extensive precipitates had been formed. As the tissue grew forward, it surrounded many of the masses of precipitate. The presence in the tissue of these masses of precipitate did not seem to irritate the vessels in any way. Many cases were studied in which capillaries had grown close to such precipitated masses, yet showed no abnormal stickiness of their endothelium toward leukocytes, or other signs of irritation. Figure 4 shows such a mass of precipitate within the living tissue. Many of these masses of precipitate within the tissue were gradually absorbed, so that 42 days following the removal of the phosphate solution from the moat a considerable portion of the precipitate had disappeared. Some of the masses of precipitate were less soluble than others. 3. Ilztroductiom ilzto the moat of solutions of sodium and potassium chloride. I n order to determine whether the sodium and potassium ions present in the buffers had been responsible for the injury of the tissue and formation of the precipitate, a solution of sodium and potassium chloride containing the same concentration of sodium and potassium ions as the phosphate solution at pH 7.4 was installed within the moat of chamber 1/23/32R after much of the precipitate previously formed in this chamber by the phosphate solution at p H 6.2 had been dissolved. None of the vessels were destroyed, and no precipitate was formed. A phosphate solution at p H 7.4, subsequently introduced into the moat, injured the peripheral tissue for a distance of 0.77 mm. proximal to the most advanced capillaries, and caused the formation of a precipitat e. A solution of sodium and potassium chloride containing approximately twice the concentration of sodium and potassium ions as did the phosphate solution at pH 7.4 was introduced into chamber 7/30/31L, before injection of any phosphate 64 RICHARD G. ABELL solution into the moat. No precipitate was formed and no vessels were destroyed. As in the case of the experiment noted above, a phosphate solution at pH 7.4,subsequently introduced into the moat, produced typical injury to the tissue and resulted in the formation of a precipitate. 4. AfiPzalysis of the precipit,ate formed i.n-the bay subsequaat to iw%vductiofiinto the moat of a buffered phosphate solution at pH 7.4. In the case of one chamber, chamber 2/17/32L, the precipitate formed subsequent to introduction of the phosphate buffer at pH 7.4 was analyzed for phosphorous. This was done in the following way. Thirteen and one-half months after the formation of the precipitate, the chamber was cut from the ear of the rabbit, cocaine being used as the local anaesthetic, and care being taken not to disturb either the precipitate or the tissue in the bay. The top of the chamber was removed, thus exposing the precipitate and the tissue in the bay. The bay was then flooded with 2 M nitric acid, followed by a solution of ammonium nitrate and a 3 per cent solution of ammonium molybdate (Wilson, '28). The masses of precipitate turned yellow. So also did the tissue, probably because of the production of ammonium phosphomolybdate. The color of the masses of precipitate was slightly more intense than that of the tissue. On the following day, the masses of precipitate were removed from the chamber. Care was taken not to include material other than the masses of precipitate that had been formed by interaction of the buffer with some substance or substances in the bay. The masses of precipitate were then washed with water. The yellow color paled decidedly, but did not disappear entirely. Following the addition of 2 M nitric acid and solutions of ammonium nitrate and ammonium moIybdate, the masses of precipitate became as yellow in color as they had been before washing with water. It would thus appear that the precipitate formed as the result of introduction of the buffered phosphate solation into the moat contained phosphorous. PHOSPHATE SOLUTIONS AND LIVING TISSUES 65 DISCUSSION That the phosphate solutions introduced into moat chambers in the present experiments actually passed to the region of living cells and tissue in the bay, is indicated by the increased stickiness of certain of the vessels toward leukocytes, the injury to a part of the peripheral tissue, and the formation of masses of precipitate. The work of Starkenstein ( '14), Binger ('17), Tisdall ( '22), and others, demonstrates that solutions of sodium phosphate, injected into animals, may produce toxic symptoms. Binger ('17) suggests that the toxicity may be due to the specific action of the phosphate ions in combination with the reaction of the solution in which they exist, and states that it is intimately associated with a decrease in serum calcium, but not dependent upon this alone. Greenwald ('18) asserts that the phosphate ion is not particularly toxic, and attributes the effects of injection of sodium phosphate solutions to the changes in reaction and osmotic pressure produced by them in the blood plasma and in the tissue fluids, and "to the increased concentration, both absolutely and relatively to the other cations, of the sodium ions." Tisdall ( '22) considers the increased sodium concentration combined with the lowered concentration of calcium as the most important factor in the production of the toxic symptoms, while Underhill, Gross and Cohen ('23) believe that the phosphate ion plays no specific role in the production of the toxemia, except in so f a r as it may aid in causing a diminution in calcium and thus disturb the normal ionic equilibrium. They agree with Greenwald and Tisdall that the symptoms are due to sodium, and in their experiments also to potassium poisoning. On the other hand, Elias and Kornfeld ('23) report that the tetany produced in patients by the injection of solutions of sodium phosphate is due to the phosphate ions, and not to the hydrogen ion concentration or to the sodium ions contained in the solutions. I n the present experiments it is clear that the sodium and potassium ions were not responsible, at least not directly, for the injury to the living tissue, since solutions of sodium and THHl ANATOMICAL RECORD, VOL. 64, NO. 1, AXD SUPPLEMENT NO, 1 66 RICHARD 0. ABELL potassium chloride containing even greater amolxnts of sodium and potassium than did the phosphate solutions failed to produce such an effect. The injury of the tissue by the phosphate solutions at pH 7.4 as well as at pH 6.2, indicates that the toxicity of such solutions, when introduced into moat chambers, does not depend upon the hydrogen ion concentration within the limits used in the present experiments, though it is possible that the somewhat greater extent of tissue injury in parts of chamber 1/23/32R by the phosphate solution at p H 6.2 may have been due to the greater concentration of hydrogen ions contained in this solution. I n a series of experiments, the author has introduced into moat chambers Ringer-Locke 's solutions of greater and of less osmotic concentration than the phosphate solutions described in ihis paper. Such solutions did not visibly injure any of the tissue in t.he bay. It would thus appear that the phosphate ions in the phosphate buffers introduced into moat chambers in the present experiments exerted a toxic effect upon the living tissue. Injury to the tissue concomitantly with the formation of masses of precipitate suggests that, as in the case of the experiments reported by Binger ('17), the toxicity may have been associated with a diminution of the calcium ion concentration, such diminution having been caused by the excess of phosphate ions. Whether or not the toxicity was due to an effect of the phosphate ions exerted in some other manner than by a disturbance of the normal ionic equilibrium is not evident. There appears to be little doubt that the phosphate ions contained in the phosphate solutions were responsible for the formation of the precipitate, since masses of precipitate were produced by introduction of phosphate solutions at pH 7.4 as well as at pH 6.2, and since none were formed subsequent to injection of sodium and potassium chloride solutions containing even greater concentrations of sodium and potassium ions than did the phosphate solutions. According to Peters and Van Slyke ( '32), the serum and other body fluids are probably normally effectively saturated with Cas(PO,),, so that any excess of either calcium or phosphate ions would, in time, cause precipitation, were not such FIBERS IN TISSUE CULTURE 67 excess removed. I n the case of the plasma, such removal according to these investigators, is normally presumably effected through excretory channels or by precipitation of the excess Caa(P04)2in the bones. I n the present experiments the circulation passing through the most advanced vessels was evidently unable to remove the phosphate solution with sufticient rapidity to prevent injury to the peripheral tissue, the phosphate ion concentration becoming sufficiently high to cause precipitation even in the absence of bone. Since each of the phosphate solutions used contained a large physiological excess of phosphate ions, it appears probable that the precipitates produced were composed partly of calcium phosphate. Such a conclusion is borne out by analysis of the precipitates formed in chamber 2/17/32L for phosphorous. No analysis for calcium was made, but it is well known that excess of phosphate ions precipitates calcium within the pH range used in the present experiments, and vice versa. Sendroy and Hastings (’27) found that the addition of solid Cas(P04)2to horse serum ‘in vitro’ resulted in a rapid precipitation of CaCO,. This suggests that a part of the precipitate formed following introduction of the phosphate solutions into moat chambers may have been calcium carbonate. I n this connection the experiments of Wells and Mitchell ( ’10) are of interest. These investigators observed that subsequent to calcium phosphate or calcium carbonate implantation into animals, in time carbonate was taken up by the phosphate and phosphate by the carbonate. The amorphous, granular, clumped precipitates formed in the present experiments resemble those of calcium phosphate described by Watt (’23), precipitated ‘in vitro’ in colloidal solutions. Such precipitates were found by Watt to be constantly granular and amorphous in character, the amorphous granules gathering in clumps and masses. Calcium carbonate, precipitated in the presence of colloids, was observed by Watt to separate out and persist in the form of spherules, some of which had slightly irregular peripheries and appeared to be built up of concentric layers. The similarity between such spherules and those observed in the ‘PHB ANATOMICAL RECORD, VOL. 64, NO. 1, AND BUPPLEXRNT N O . 1 68 RICHARD Q. ABELL present experiments is suggestive. It should be noted that spherules were not observed in chamber 1/23/32R following introduction of the phosphate solution at p H 6.2. This would seem to be of some interest, since Watt reports that spherules are less likely to form at higher hydrogen ion concentrations. Areas of pathologic calcification taken from the pineal gland, the choroid plexus, and various arteries were observed by Watt ( '27) to contain calcium phosphate and calcium carbonate in forms which resembled those previously described for these salts precipitated in colloidal solutions. The precipitates formed in the present experiments are strikingly similar in appearance to those observed by Watt in these areas of pathologic calcification. Wells ('11) refers to the highly insoluble character of the calcium salts of pathologic calcifications, while Watt ('25) states that the deposits of calcified areas, once formed, tend to persist. Shelly ('31) described calcified nodules in the subcutaneous tissues of the arm and buttocks. The history of the patient suggests that the deposits may have been highly insoluble. Many of the masses of precipitate formed in moat chambers in the present experiments were extremely insoluble, some persisting for several months, others remaining undissolved throughout the entire survival period of the chamber -in one instance more than 13 months. Some of the masses of precipitate were absorbed within a few weeks, however, particularly in the case of those formed by the phosphate solution at pH 6.2. Wells refers to the work of von Werra (1882) as evidence that calcification deposits are not necessarily permanent, and to that of Maximow ('06)' who observed that even if such deposits have undergone secondary ossification they may disappear within a year. The present experiments suggest that differences in solubility of the deposits of pathologic calcifications may be related to the pH at which the deposits were formed. The studies of Klotz ( '05), Wells ( '06), and others show clearly that pathological deposition of calcium generally occurs only in tissues that have suffered either a partial or a PHOSPHATE SOLUTIONS AND LNINQ TISSUES 69 complete loss of vitality. This may also have been the case with the formation of masses of precipitate in the present experiments. The cells and tissues in the regions in which masses of precipitate subsequently appeared were living and active before introduction of the phosphate solutions into the moat. In most instances they were evidently injured during the process of precipitate formation. I n the case of ‘metastatic calcification’ Wells (’11) states that “whenever from any cause the proportion of calcium present in the blood is so great that it requires the effect of both the colloids and of the CO, in maximum concentration to keep it in solution; then the calcium salts are deposited in those points in the body where the CO, content of the fluids is least.” The present work suggests that a similar precipitation might occur without increase of the calcium ion concentration if the phosphate ion concentration became sufliciently high. By means of the present technic it is possible not only to observe, with the high powers of the microscope, including oil immersion lenses, the actual formation of precipitates which resemble closely those of pathologic calcification, but also to modify at will the chemical environment of the tissue, with the purpose of throwing further light upon the factors which are responsible for such calcification. The formation of masses of precipitate for a short distance in front of the original peripheral capillary loops, as well as throughout the region of injured tissue, is of interest in the light of the observation, reported by the author (’34), that granules in certain cells in this same general region stain a brilliant blue following introduction of solutions of methylene blue into moat chambers. At a greater distance beyond the peripheral capillary loops than approximately 293 p the methylene blue was reduced within all of the living cells, indicating a decrease in oxygen concentration with increased distance from the distal capillary loops. The present experiments might indicate a similar decrease in calcium ion concentration. 70 RICHARD Gi. ABELL That methylene blue, following introduction into moat chambers, is prevented from diffusing proximally throughout the tissue in the bay because of removal of the dye by blood passing through the most advanced rows of peripheral capillaries, was shown by the author ('34). Sticking of leukocytes to the endothelium of the venules and veins that collected blood from the peripheral capillaries, but not in others, suggests that the phosphate solutions used in the present experiments also entered the peripheral capillaries. Removal of the solutions by the blood passing through these capillaries, possibly also the formation of masses of precipitate which interfered with the free access of the phosphate solutions to the tissue, evidently prevented more general diffusion of the solutions throughout the bay, and accounts for limitation of precipitate formation and injury of vessels to a narrow strip of peripheral tissue. That the concentration of phosphate solution which reached those capillary loops in the bay that were closest to the moat was greater than that which reached other peripheral capillary loops located in the bay at greater distances from the moat, is shown by the experiment in which the phosphate solution at pH 6.2 was introduced into chamber 1/23/32R, since the extent of tissue injury and precipitate formation was greatest in those regions of peripheral tissue that were closest to the moat, and since terminal arterioles and peripheral capillary loops closest to the moat became dilated, evidently because of the effect of the phosphate solution upon them, while the more proximally placed of the terminal arterioles and capillary loops changed little, if any, in caliber. Similar but not such striking results were also secured with the other chambers. SUMMARY 1. This paper describes experiments in which buffered phosphate solutions at pH 7.4 and 6.2 were introduced into moat chambers. Such solutions, placed within the moat, pass by diffusion into the bay, and there interact with the thin layer of living cells and tissue normally present during the PHOSPHATE SOLUTIONS A N D LIVING TISSUES 71 growth of blood vessels and other structures in the chamber. The effect of the phosphate solutions upon the living tissues within the chambers was observed with the high powers of the microscope. 2. Phosphate buffers, of the pH and concentration used in the present experiments, injure the peripheral tissue in the bay and cause the formation of masses of precipitate in the region of injured tissue. 3. The most advanced of the distal tissue was injured proximally for a distance of 1.22 111111. following introduction of the buffer at pH 6.2. Subsequent to installation of the buffer at pH 7.4, the distal tissue was injured in no case for more than a distance of 0.77 mm. 4. The distance beyond the original distal capillary loops to the line at which precipitate has been observed to form varies from 75 p to 230 p. It is not at present certain whether precipitate formation occurs in the uninjured tissue proximal to the most advanced of the peripheral functional capillaries, although masses of precipitate have been frequently observed within the proximal living tissue. 5. Injury to the peripheral tissue was accompanied by dilation of the arterioles which supplied this tissue with blood; by increased stickiness toward leukocytes manifested by the endothelium of the capillaries, venules and veins which collected blood from the periphery; and with emigration of leukocytes from the distal capillaries. 6. The results indicate that the phosphate solutions entered the most advanced capillaries and were removed by the blood circulating through them. The solutions were not removed with sufficient rapidity, however, to prevent the phosphate ion concentration-possibly also in the case of the buffer at pH 6.2, the hydrogen ion concentration-from becoming injuriously high. 7. The concentration of phosphate solution that reached those capillary loops closest to the moat was greater than that which reached other distal capillary loops, located at greater distances from the moat. 72 RICHARD G. B E L L 8. Injury to the living tissue was limited to a narrow strip of peripheral tissue, evidently because of removal of the phosphate solutions by the most advanced capillaries, possibly also because of the formation of masses of precipitate which interfered with the free access of the phosphate solutions to the tissue. 9. Masses of precipitate formed subsequent to installation of the buffer at pH 6.2 were gradually absorbed by the living tissue, but those produced following introduction of the buffer at pH 7.4were much less soluble, many remaining in contact with living tissue practically unchanged f o r months-in the case of one chamber, for more than a year. Once formed, the masses of precipitate exerted no observable injurious effect upon the tissue. Such masses of precipitate resemble closely those of pathological calcification. Analysis of the precipitate formed by the phosphate solution at pH 7.4 showed that it contained phosphorous, and it appears likely that the precipitates formed by the phosphate solutions at both pH 6.2 and 7.4 were composed partly of calcium phosphate. They may also have contained calcium carbonate. LITERATURE CITED ABFLL,RICHARDG. 1934 Studies of the reaction between methylene blue and living cells and tissues in the transparent moat chamber introduced into the rabbit’s ear. Anat. Bee., vol. 60, p. 161. ABELL, &CHARD G., AND ELIOT R. CLARK 1932 A method of studying the effect of chemicals upon living cells and tissues in the moat chamber, a transparent chamber inserted i n the rabbit’s ear. Anat. Rec., vol. 53, p. 121. 1933 Some studies on the reaction between living cells and chemical substances in the ‘moat’ chamber introduced into the rabbit’s ear. Anat. Rec., vol. 55, supplement, p. 1. BINGEE,CARL 1917 Toxicity of phosphates in relation to blood calcium and tetany. J. Pharmacol. and Exp. Therap., vol. 10, p. 105. CLARK,ELIOTR., AND ELEANOR LINTONCLARK 1 9 3 5 s Changes in blood-vessel endothelium. Am. J. Med. Sci., vol. 189, p. 399. 1935 b Observations on changes in blood vascular endothelium in the living animal. Am. J. Anat., vol. 57, p. 385. ELIAS,H., AND F. KORNWLD1923 ?iber die Wirkung Saurer PhosphateLosungen bei Tetanie. Klin. Wchnsehr., Bd. 2, s. 1206. GREENWALD,IsmB 1918 Observations on the effect of intravenous injections of some sodium salts with special reference to the supposed toxicity of sodium phosphate. J. Pharmaeol. and Exp. Therap., vol. 11, p. 281. PHOSPHATE SOLUTIONS AND LIVING TISSUES 73 KLOTZ,OSKAR 1905 Studies upon calcareous degeneration. I. The process of pathological calcification. J. Exp. Med., vol. 7, p. 633. MAXIMOW, ALEXANDER 1906 Ueber experhentelle Eraegung von Bhochenmarkgewebe. Anat. Am., Bd. 28, 8. 609. PETERS, JOHN P., AND DONALDD. VAN SIJYKE 1932 Quantitative clinical Chemistry. Vol. I. Interpretations. Williams & Wilkins Go., Baltimore. SHELLEY,HAROLD 3. 1931 Calcification in the subcutaneous tissues; report of a case associated with chronic atroph,hic arthritis. Arch. Burg., vol. 23, p. 513. STARKENSTEIN, E. 1914 @her die pharmakologische Wirkung kalziumfallender Sauren und der Magnesiumsalze. Arch. f. exp. Path. u. Pharmakol, Bd. 77, S. 45. BENDROT,JULIAS,JR., AND A. B A I ~ DHASTIWGS 1927 11. The solubility of tertiary calcium phosphate in salt solutions and biological fluids. J. Biol. Chem., vol. 71, p. 783. TISDALL, FREDERICK F. 1922 The influence of the sodium ion in the production of tetany. J. Biol. cfiem., vol. 54, p, 35. UNDEUEILL, FRANK,ERWIN G. GROSSAND WILLIAM&HEN 1923 Studies in inorganic metabolism. 111. The significance of the phosphates in the production of tetany. J. Metab. Res., vol. 3, p. 679. Warn, JAMES C. 1923 The behavior of calcium phosphate and calcium carbonate (bone salts) precipitated in various media, with applications t o bone formation. Biol. Bull., vol. 44, p. 280. 1925 The deposition of calcium phosphate and calcium carbonate in bone and in areaa of calcification. Arch. Surg., vol. 10, p. 983. 1927 Deposition of calcium salts in areas of oalciflcation. Arch. Burg., vol. 15,p. 89. WELLS,H. GIDEON 1906 Pathological calciflcation. J. Med. €tea., N. S., vol. 9, p. 491. 1911 Calcification and ossification. The Harvey Lectures, 19101911, p. 102. W ~ L GH. , GIDEON, AND JAMEEI H. MITCHELL 1910 Studies on caldcation a d ossification. 111. J. Med. Res., vol. 22, p. 501. VON WEZt8A, JOSEPH 1882 Ueber die Folgen des voriibergehenden und dauernden Verschlusaes der Nierenarterie. Virchows Arch. f . path. Anat., Ed. 88, 8.197. WILSON, D. WFLIGHP1928 A laboratory manual of physiological uhemiatry. Williams & Wilkins Go., Baltimore.