Chemical Properties of Tissue-Fibrinogen.

Tissue-fibrinogen is slightly soluble both in water and in dilute NaCl solutions, but is soluble to an appreciable extent only in dilute alkaline solutions. Its alkaline solutions are not coagulated on boiling, and the tissue-fibrinogen can be reprecipitated from these solutions by acidifying slightly with acetic acid. Tissue-fibrinogen solutions give the xantho-proteic and the biuret1 reactions in the cold. In the case of the biuret reaction, a certain amount of time has to elapse before the reaction reaches its maximum intensity. Nitric acid in the cold produces a coagulum which dissolves to a considerable extent upon agitation with excess of the acid. This coagulum disappears almost entirely on heating, and falls out to a considerable extent on cooling the solution. When subjected to a peptic digestion, a clear solution of tissue-fibrinogen deposits a heavy precipitate of a substance which is soluble in alkalies, and which is rich in phosphorus.

I pointed out (loc. cit.) that these chemical characters entitle us to classify tissue-fibrinogen among the nucleo-albumens; and I showed that we had some confirmation of this conclusion in the fact that it was possible to obtain intravascular coagulation by the injection of a solution of a nucleo-albumen obtained from yeast.2

It will be seen that certain of the above reactions, notably the biuret and the xantho-proteic reactions obtained in the cold, and the disappearance of the nitric acid precipitate on warming, and its reappearance on cooling, would naturally suggest the presence of

1 That the biuret reaction obtained is a true biuret reaction, and a very distinct one, may be gathered from the following experiment :-A portion of the solution is divided into two equal parts, and one of these is digested with hydrochloric acid and pepsin for twelve hours. At the end of that period it is filtered and neutralized, and its biuret reaction is then compared with that of the original solution, equal portions of the fluids and of the reagents being of course employed in either case. In the case of the digested solution the biuret reaction develops immediately, while it develops only gradually in the case of the original solution. At the end, however, of two hours the reaction is almost quite as marked in the one solution as in the other, and there is no difference whatever in the tints obtained. It is, however, to be noted that there is a shade of purple in biuret reaction when it is observed immediately after the addition of the reagents. As, however, the disintegration of the tissue-fibrinogen by the caustic alkali proceeds the last trace of blue disappears, the shade of red then obtained, of course, varying with the concentration of the solution.

2 Rauschenbach (Dissert. Dorpat., 1882) had already proved that the addition of yeast-cells increases the rapidity of coagulation and the amount of fibrin formed in cold horse plasma.

albumose in our 1 per cent. Na,CO, solution. Certain of the physiological evidence obtained, viz. the occurrence of a negative variation (to be referred to more particularly in a subsequent portion of this Paper) in certain conditions after the injection of these solutions, suggested the same conclusion.

In view of these considerations I therefore took the precaution of very carefully washing out my acetic acid precipitates with salt solutions of moderate strengths, and afterwards with distilled water, in order, by such treatment, to remove any trace of the ordinary albumoses which might contaminate the tissue-fibrinogen. I found, however, that such treatment altered nothing either in the physiological or the chemical reactions of my solution, and it thus became evident that these reactions were either (a) due to an admixture of dysalbumose, which, as described by Kühne, possesses the same solubilities as those which have been determined to characterize tissue-fibrinogen, and which consequently would be inseparable from it, or (b) that they were of the essence of the tissue-fibrinogen.

The latter of these two assumptions appears to me, upon general grounds, to afford the more natural explanation of the facts, and there do not appear to me to be any real chemical difficulties in the way of making the assumption that a substance which is not albumose may give what pass current as the characteristic chemical reactions of an albumose. In support of this view I urged (loc. cit.) the following considerations, and I may perhaps be allowed to recapitulate their substance here :

(a) It would seem to be self-evident that any substance, from which nitric acid would split off an albumose, would give the albumose reaction with that acid in precisely the same manner as free albumose does.

(b) The colour reactions (e.g. the xantho-proteic and biuret reactions) are only by courtesy allowed to rank as evidence of the presence of proteids, as they, strictly speaking, afford evidence only of the presence of certain chromogens, which occur invariably among the disintegration products of the proteids. Further, the distinction which is drawn between these colour reactions, as obtained at ordinary temperatures, and as obtained after the application of heat, is based only upon certain differences of stability which obtain between the various proteids; and it plainly does not therefore follow that a proteid which can be disintegrated at ordinary temperatures by the action of strong acids or alkalies must be always either an albumose or a peptone. Indeed we have direct evidence in the case of both the nucleo-albumens obtained from yeast that both these colour reactions can be obtained in the cold

with proteids which do not belong either to the one or to the other of the classes above referred to.

We may therefore, without doing violence to any facts, assume that these reactions do belong to the essence of the tissue-fibrinogen. We have additional justification for this assumption in the fact that when we acidify our solution and boil it, the filtrate will no longer give any biuret reaction; further, in the fact that when we neutralize, and then precipitate our solution, by the addition of strong alcohol, the coagulum obtained will no longer, after standing some weeks under alcohol, yield up, upon extraction with water, any substance which will give a biuret reaction. These facts appear to me to be strong evidence in favour of the view that our 1 per cent. Na2CO, solution of the washed acetic acid precipitate is not contaminated with any free albumose.

We have now to pass on to the consideration of the

Alterations which are produced by the Action of Heat in the Chemical and Physiological Reactions of a Tissue-Fibrinogen Solution.

We are indebted to Wooldridge for the observation that when tissue-fibrinogen solutions are boiled they lose their power of inaugurating intravascular coagulation, while they still retain their power of coagulating peptone plasma. I have repeatedly verified this observation; and I believe that this modification in the physiological reactions is attributable to a partial disintegration undergone by the tissue-fibrinogen, by means of which the coagulative power of the solution is reduced to a point at which it is no longer able to inaugurate intravascular coagulation, although the solution still retains enough coagulative power to bring about the coagulation of peptone plasma. If this view of matter is correct we ought to find confirmation of it, on comparing the chemical reactions of the boiled with those of the unboiled solution. We have already seen (vide supra) that when we precipitate the unboiled solution by the addition of acetic acid, and when we make this precipitation more complete by boiling, we obtain a filtrate which no longer gives us any biuret reaction. When we now repeat this experiment with a solution of tissue-fibrinogen which has been previously boiled for some little time in the presence of free alkali, we still, as before, obtain a precipitate when we acidify with acetic acid (this being a precipitation of presumably unaltered, or at any rate only partially altered, tissue-fibrinogen); but the filtrate obtained by filtering off this precipitate now gives a marked biuret

reaction. I take it that this biuret reaction is probably referable to an albumose1 produced by the disintegration of the tissue-fibrinogen.

We have further evidence of a disintegration produced by boiling in the fact that a watery suspension of the thoroughly washed acetic acid precipitate acquires a distinctly acid reaction on boiling, and we have also evidence which points in the same direction in the fact that free phosphates can be demonstrated in the boiled solutions of tissuefibrinogen.

I would, however, guard against being supposed to assert that the demonstrable chemical disintegration products are at all to be regarded as an adequate measure of the disintegration effected; for the alterations obtained in the physiological reactions are often out of all proportion to the demonstrable chemical changes effected by the application of heat, to which, as will appear from the following protocols, the tissuefibrinogens are peculiarly sensitive:

Rabbit 58, 20th Feb., 1891.—Injected 12 cc. of a watery extract of testicles into the jugular. Cessation of respiration in about one minute after completion of injection.

P.M.-Right heart, one solid clot; aorta and left heart, empty; S. and I. V. C., clotted, firm; portal vein also clotted firmly.

Took another portion of the same watery extract; boiled and filtered off the hardly appreciable coagulum. Injected this boiled extract after cooling into

Rabbit 59.-18 cc. injected into jugular; no apparent effect from this injection. Rabbit killed with chloroform, and no trace of clotting detected anywhere.

Took a third portion of the extract, and heated it in a waterbath to 50° C. Injected this after cooling into—

Rabbit 60.-22 cc. were injected; no apparent result. Continued the injection with 16 cc. of the unboiled watery extract. Death follows instantaneously, and the post-mortem discloses universal coagulation in both the venous and arterial vascular systems.

The above rabbits were chosen almost exactly of the same size, and had been kept under the same conditions.

In addition to the above experiments with rabbits, I made a number of experiments in the same direction with dogs, using tissuefibrinogen solutions which had been subjected to a boiling temperature for many hours. In three experiments with such solutions I observed only a negative phase of coagulability, the blood in each case remaining

1 See note A at end of Paper, p. 146.

liquid for many hours after shedding, and giving all the reactions of albumose blood. I may remark that at the time when I made these experiments I was inclined to emphasize the results obtained as affording evidence of a disintegration of tissue-fibrinogen into albumose under the influence of heat. Further study of the phenomena obtained after the injection of albumose-free solutions of tissue-fibrinogen, however, showed that a negative variation of coagulability is part of the reaction of the system to such an injection, and that it is, in fact, the most noticeable part of that reaction when only a small quantity of tissue-fibrinogen has been injected. The view I am therefore inclined to take of these experiments in the dog with boiled tissue-fibrinogen solutions is that there may have been a certain amount of residual, not-yet-disintegrated tissue-fibrinogen present, and the negative variation observed may have, perhaps, been referable to this.

The Modifications that can be effected in the Reaction of the System to Injections of Tissue-Fibrinogens by Modifications induced in the Physiological Condition of the Animal which is the subject of Injection.

I have already in a previous communication (Jour. of Phys., 1891, No. 2) referred to this question in connexion with a discussion of the causes which determine the coagulation, obtained upon the injection of tissue-fibrinogen solutions, to the portal system. I there endeavoured to show that Wooldridge's tentative hypothesis, which proposed to explain the occurrence of coagulation in this vascular area by assuming an absorption into the blood from the intestinal tract of some substance favourable to coagulation, was an hypothesis which further study of the facts showed to be at any rate a very improbable one.

I there also endeavoured to show that the determination of the coagulation to the portal tract was probably due to an excess of CO, present in the blood of that system. In support of this view I adduced the protocols of certain experiments with injections of tissuefibrinogen, which went to show that a rise of the general CO2 tension of the blood, whether brought about by clamping the trachea, or by causing the animal to breathe an atmosphere containing an abnormally large percentage of CO2, was followed by a coagulation throughout the whole vascular system; and I also showed that causes which might be assumed to cause a local increase of CO2 tension gave rise to the occurrence of intravascular coagulation in the particular vascular areas which were involved.

I have made a few further experiments with the view of collecting additional evidence as to the effects of modifications of the CO, tension