No. 4275.

AUGUST 5, 1905.

The Croonian LecturesON

THE CHEMICAL CORRELATION OF THEFUNCTIONS OF THE BODY.

Delivered before the Royal College of Physicians of Londonon June 20th, 22nd, 27th, and 29th, 1905,

BY ERNEST HENRY STARLING, M.D. LOND.,F.R.S.,

FELLOW OF THE COLLEGE; JODRELL PROFESSOR OF PHYSIOLOGY,UNIVERSITY COLLEGE, LONDON.

LECTURE I. t1acDelivered on J1ltte 20th. m

THE CHEMICAL CONTROL OF THE FUNCTIONS OF THE BODY. p1

MR. PRESIDENT AND GENTLEMEN,-From the remotest lt19ages the existence of a profession of medicine, the practice g,

of its art and its acceptance as a necessary part of everycommunity have been founded on a tacit assumption that tl

the functions of the body, whether of growth or activity of corgans, can be controlled by chemical means ; and research

c

by observation of accident or by experiment for such means has resulted in the huge array of drugs which form the bpharmacopoeias of various civilised countries and the o

common armamentarium of the medical profession through-out the world. The practice of drugging rests on the t

supposition that the functions of the body can be in- o

fluenced in a normal direction by such means. I h

propose in these lectures to inquire how far such a e

belief is consonant with our own knowledge of the t

physiological workings of the body, how far, that is to say, a

the activities and growth of the different organs of the r

body are determined and coordinated among each other by tchemical substances produced in the body itself but capable cof classification with the drugs of the physician. If a imutual control, and therefore coordination, of the different ifunctions of the body be largely determined by the pro- tduction of definite chemical substances in the body, the dis- (

covery of the nature of these substances will enable us to i

interpose at any desired phase in these functions and so to ! aacquire an absolute control over the workings of the human Jbodv. Such a control is the goal of medical science. How (

far have we progressed towards it ? How far are we justifiedin regarding its attainment as possible ? c

I hope to be able to vindicate to you the assumption 4

which is at the basis of medical practice and to show thatthe activities of, at any rate, the large majority of theorgans of the body are coordinated among themselves by the Iproduction and circulation of chemical substances, so thatthe results of physiological researches up to the presentjustify us in the faith that within a reasonable space oftime we shall be in the possession of chemical substanceswhich are normal physiological products, and by means ofwhich we shall be in a position to control not only theactivities but also the growth of a large number of the organsof the body.

In man and the higher animals the marvellous adaptationseffected by means of the central nervous system are so muchin evidence that physiologists have been tempted to ascribeevery nexus between distant organs to the intervention ofthe nervous system ; the more so because by this means anadaptation to changes, internal or external, can be effectedin many cases within a fraction of a second. But in theevolution of life upon this earth this method of adaptationis of comparatively late appearance and is confined almostentirely to one division of living beings—i.e., the animalkingdom. In the lowest organisms, the unicellular, such asthe bacteria and protozoa, the only adaptations into themechanism of which we can gain any clear insight are thoseto the environment of the organism and in these cases themechanism is almost entirely a chemical one. The organismapproaches its food or flies from harmful media in con-

sequence of chemical stimuli ; it prepares its food for

digestion or digests it by the formation of chemical sub-

stances, toxins or enzymes. In the lowest metazoa, such asthe sponges, there is still no trace of any nervous system.The coordination between the different cells of the colony is

still determined by purely chemical means. The aggregationof the phagocytic cells round a foreign body is apparentlydue to the attraction exerted on them by the chemical sub-stances produced in the death of the injured tissues.With the appearance of a central nervous system or

systems in the higher metazoa the quick motor reactionsdetermined by this system form the most obvious vital mani-festations of the animal. But the nervous system has beenevolved for quick adaptations, not for the abolition of thechemical correlations which existed before a nervous systemcame into being. A study of the phenomena of even thehighest animals shows that the development of the quicknervous adaptations involves no abrogation of the other moreprimitive class of reactions-i.e., the chemical ones. Wherethe reaction is one occupying seconds or fractions of asecond the nervous system is of necessity employed. Wherethe reaction may take minutes, hours, or even days for itsaccomplishment the nexus between the organs implicatedmay be chemical. Already we are able, in many cases, toprove the existence of such a chemical nexus and to employit in artificially producing a state of growth or activity whichis in normal circumstances merely a phase in a complexseries of physiological changes.The chemical reactions or adaptations of the body, like

those which are carried out through the intermediation of thecentral nervous system, can be divided into two main

classes-(1) those which are evoked in consequence of

changes impressed upon the organism as a whole fromwithout; and (2) those which, acting entirely within thebody, serve to correlate the activities, in the widest sense

, of the term, of the different parts and organs of the body.The first class of adaptations includes those reactions of

: the body to chemical poisons produced by bacteria or higher. organisms and represents one of the most important means: by which the body maintains itself in the struggle for exist-Lence. The complicated phenomena involved in the forma-

tion of antitoxins, of cytolysins, of bactericidal substances,,

and such like means of protection, have been the subject of; much study of recent years and their immediate interest to the practical physician renders it unnecessary for me todevote any time to their discussion, especially as the subject,L is one to which I have not given any personal attention. Thet investigation of the second class, that of the correlation of- the activities of organs, has byreason of its greater obscurity,- or of the greater difficulty of its practical application in3 medicine, fallen largely to the province of the physiologist3 and I therefore propose to deal almost exclusively with thosei members of this class of reactions which have so far beenv definitely ascertained.1 Before, however, entering into details of any particular

correlation it may be profitable to consider what we mayn expect to be the nature of the substance which will in any,t given case act as a chemical nexus between different organs.e We are dealing here with a question of general pharma-ecology. As Ehrlich has pointed out, the chemical substances,t which act on the body or parts of the body, producingIt physiological or pharmacological effects, can be divided)f largely into two main groups. Ehrlich’s conception of the,first ,s first group is bound up with his conception of the nature of)f the living protoplasmic molecule as a living nucleus withIe side chains of various descriptions. Assimilation of food-ts stuffs consists in the linking on of the food molecule

as a fresh side chain to the central nucleus. TheIS common feature among the substances of the firsth class is their close resemblance to an assimilable sub-)e stance or foodstuff. All these substances acquire a closef attachment to, or even identification with, the living proto-,n plasm, and as a rule their effects are apparent only afterid sufficient time has elapsed for their building up into theIe protoplasmic molecule. To this class belong the numerousIII substances closely allied in their chemical character tost the proteids which are designated as toxins. All are pro-al duced by the agency of living organisms. I need onlyis adduce as examples the various products of the pathogenicae bacteria, such as diphtheria and tetanus, the poisonousse toxins of higher plants, such as ricin and abrin, and thoseIe formed as a weapon of offence by higher animals, such asm the active principles of the various snake venoms. Accordingn- to Ehrlich, these all resemble assimilable foodstuffs in thator they possess a haptophore group by which they can anchorb- themselves on to the living molecule, becoming thus part ofas its side chains. The toxophore -group thus introduced inton. the living molecule upsets and disorganises its reactions,is leading by disorder of one or more functions to the death of

F

340

the animal. In most cases the toxophore group, at any rate,is specific for some definite tissue or type of cell. Thustetanotoxin exercises its effect almost entirely on the peri-pheral sensory neurones. It is doubtful, however, whetherthe haptophore group is so specific, if we are to acceptEhrlich’s conception of the mode of formation of antitoxins ;since we may get formation of antitoxins in animals wherethe toxic effect is entirely wanting.

The idea that these toxins ape the part in the protoplasmicmolecule of an assimilable foodstuff does not involve as anecessary sequence the formation of antitoxins or anti-bodies to the normal foodstuffs. That, in fact, the powerof assimilation is independent of the power to produoeantibodies has been shown by van Dungern in a researchspecially directed to determine this point. This observerfound that the proteids of crabs’ blood could be injectedinto the blood stream of the rabbit and undergo assimilation.Being proteids foreign to rabbits’ blood their injection pro-voked the production in the latter of a precipitin for crabs’blood plasma, but the assimilation of the proteid and theproduction of the precipitin were found to be absolutelyindependent phenomena.The first group therefore of pharmacological substances

may be defined as substances presenting many points ofresemblance to proteids, potent like enzymes in infinitesimaldoses, and giving rise as a result of their introduction intothe body to a reaction consisting in the production of anantibody.

Ehrlich’s second group of substances, which includes practi-cally all our common drugs, probably act on the protoplasmicmolecule or part of it by reason of their chemico-physicalproperties or their molecular configuration. It is difficultto give a more definite expression of their mode of action.We know that in many cases slight changes in the molecule,such as the introduction or withdrawal of an ethyl, methyl,or NH2 group into or from a drug or group of drugs, altertheir physiological actions in a regular manner. Weknow, moreover, that substances of the most diverse consti-tution, such as the various anaesthetics, may have little morethan their fat solvent powers in common. All these drugs,however, are more or less stable compounds, generally to beobtained in a crystalline form and not easily destroyed byheat. On introduction into the body the incubation periodof their physiological effects is generally determined only bythe time necessary for their distribution to, and their diffusioninto, the cells which they chiefly affect. Although repeateddoses of them can set up a certain degree of tolerance, in nocase is there any evidence of the formation of a physiologicalantidote or antitoxin to the poison.To which of these two groups of bodies must we assign the

chemical messengers which, speeding from cell to cell alongthe blood stream, may coordinate the activities and growthof different parts of the body ? 7 The specific character ofthe greater part of the toxins which are known to us (Ineed only instance such toxins as those of tetanus anddiphtheria) would suggest that the substances produced foreffecting the correlation of organs within the body,through the intermediation of the blood stream, mightalso belong to this class, since here also specificity ofaction must be a distinguishing characteristic. Thesechemical messengers, however, or

" hormones " (from6pli,dw, I excite or arouse), as we might call them, haveto be carried from the organ where they are producedto the organ which they affect by means of the bloodstream and the continually recurring physiological needsof the organism must determine their repeated produc-tion and circulation through the body. If they belong tothe first class and are analogous to the toxins, each pro-duction of a given substance and its discharge into theblood stream must give rise to the formation of a specificantibody, which must increase in amount with each produc-tion of the substance in question and tend therefore toneutralise its physiological effects. It might be suggestedthat in the case of these chemical messengers the forma-tion of an antibody was a local one and limited to theorgan affected and that, in fact, their physiological effect-e.g., secretion-was actually a pouring out of the antibodyto the chemical messenger. But, as we shall see later,experimental evidence is entirely against this view, which,moreover, is not supported by any known instance of asimilar localisation of antibody as a result of injection intothe organism of any of the substances which belong definitelyto the toxin class. The formation of antibodies appears, infact, to be not a process of value in the normal physiological

life of the organism but one which has been evolved as achemical means of defence to prevent the spread of injurioussubstances from the spot originally affected or attacked.We are therefore forced to the conclusion that if the pro-

cesses of coordination of activities among the organs of thebody are carried out under physiological conditions to anylarge extent by chemical means-i.e., by the despatch ofchemical messengers along the blood stream-these emissarysubstances must belong to Ehrlich’s second order of sub-

6 stances acting on the body and must, in fact, fall into thesame category as the drugs of our Pharmacopoeia. Amongthese, indeed, specificity is not wanting and is the basis of

: their classification by pharmacologists. Thus we have drugsi elevating or depressing the activity of the nervous system ;’ we can excite secretion in all the glands of the body by. pilocarpin ; we can stimulate and finally paralyse the

praeganglionic nerve endings of the sympathetic system bythe injection of nicotine, or arouse the anabolic mechanismof the heart by the administration of digitalis. In all these

! cases we are certainly interfering with normal processes.The methods, however, which we employ are not at variancewith those made use of by the body itself in securing theharmonious cooperation of its various parts.

In discussing the internal chemical reactions of the bodyL it will be convenient to divide them into two classes-viz.,) those which involve (1) increased activity of an organ, andL (2) increased growth of a tissue or organ. In both cases we

must assume that the reaction to the chemical stimulus isitself chemical, the first class including, however, thosechanges which are chiefly katabolic or dissimilative and arealways associated with activity, and the second classinvolving diminished katabolism and increased building upor anabolism. We might, in fact, speak of the two classesof chemical stimulants as augmentor and inhibitor.

REACTIONS INVOLVING INCREASED ACTIVITY OF ORGANS.

The most striking because the simplest of this class ofreactions is that which determines in higher animals theadequate supply of a contracting muscle with oxygen andremoval of its chief waste product, carbon dioxide. Theincreased depth and frequency of respiration contingent onmuscular exertion are familiar to everyone and we know that

L the physiological object of such changes is to secure ther increased ventilation rendered necessary by the enormousL rise of gaseous metabolism which accompanies muscularL exercise. Even moderate work may raise the gaseous) exchanges to between four and eight times their amountL during rest. This increase in the respiratory movements is

entirely involuntary and may, in its earlier stages, when! affecting chiefly depth of respiration, be absolutely unnoticedby the subject of them. How is the respiratory centre

L aroused to an increased activity which is absolutely! proportional to the increased metabolism of the distant[ muscles ? 7 A nervous path is at once excluded by. the fact that hyperpncea or even dyspncea may be’ excited in an animal after division of the spinal

cord by tetanisation of the muscles of the hind limbs. and Zuntz and Geppert came to the conclusion that the exciting agent in this increased activity was some! acid substance or substances produced by the contracting. muscles and transmitted from them through the blood! stream to the respiratory centre. The subject has been

lately investigated in this country by Haldane and Priestley.. In a series of masterly experiments these observers show con-! clusively that the chemical messenger in this case is none

other than carbon dioxide. The contracting muscle, when properly supplied with oxygen, takes up this gas and gives

out carbon dioxide in direct proportion to the energy of its! contractions. The carbon dioxide diffusing rapidly into the! blood stream raises its percentage and, what is still more

important, its tension in this fluid. The respiratory centre differs from the other parts of the central nervous system. in having developed a specific sensibility to carbon dioxide.

Its normal activity is determined by the normal tension of’ this gas in the blood and lymph bathing the centre. Diminu-

tion of the tension of this gas depresses the activity of ther centre, causing slackening of respiration or even the total

cessation of respiratory movements known as apnoea.This work by Haldane may be regarded as finally deciding

, a question which has been the subject of debate for nearly half a century. The dyspnoea caused by the circulation ofr venous blood through the brain and by the deprival of thei respiratory centre of the means of maintaining its normal. gaseous interchanges has been variously attributed either to

341

oxygen starvation or to carbon-dioxide intoxication of thecentre. Haldane shows that the centre is very little sensitiveto changes in the oxygen tension of the blood. The oxygentension in the pulmonary alveoli may be altered from 20 percent. to 8 per cent., without any increase in the depth orfrequency of the respiratory movements. In these circum-stances the heart or circulatory system may feel the

deprivation of oxygen before the respiratory centre has

responded to it. On the other hand, a rise of only 2 per cent.in the tension of carbon dioxide in the alveolar air andtherefore in the blood circulating round the respiratorycentre will increase the volume of air respired 100 percent.

This simplest of all examples of a coordination bychemical means of two widely separate organs may, perhaps,give us a clue to the mode in which the more complex ofsuch correlations have been evolved. In this case thechemical messenger is a product of activity which is commonto all protoplasm and must be excreted by the cell as a

condition of its further activity. The special adaptation inthis case therefore is not the formation of a special sub-stance which shall exert a specific influence on some distantorgan but the development in this distant organ of a specificsensibility to the common product of excretion of the firstorgan. We may, perhaps, assume that the more specialised,messengers which we shall have to consider in detail laterwere at first accidental by-products of the selfish activity,of the organ producing them, the first step in the develop-ment of a correlation being the acquisition of a sensibilityto the substance in question by some distant organ.The only other example of such a reaction, in which we

know both the source and nature of the chemical messengerand the exact nature of the effects which it produces, isthe suprarenal gland. Since the time of Addison we haveknown that atrophy of these glands in man leads to a diseasecharacterised by the three cardinal symptoms of bronzing,vomiting, and extreme muscular weakness. Most of theattempts to reproduce this disease in animals have failed,owing to the fact that death follows the excision of bothglands within 24 hours ; the extreme muscular weakness iscertainly produced, and this is attended by a profound fallin the general blood pressure. In 1894 Oliver and Schilfershowed that from the medulla of the suprarenals a substancecould be extracted which, on injection into the circulation,caused marked rise of blocd pressure and increased strengthof the heart beat. Since the publication of these observa-tions our knowledge on the nature and actions of this sub-stance has progressed rapidly. The researches of Jowett inthis country, of Abel in America, and of von Furth inGermany, have shown that the active substance is a definitechemical compound derived from pyrocatechin and havingthe formula-

Takamine, by the elaboration of a method for its prepara-tion from the gland in a state of purity, has placed in thehands of the druggist a means of supplying the substance inbulk to the medical profession for therapeutic purposes.The exact knowledge of the constitution of adrenalin thusacquired has paved the way for the actual synthetic forma-tion of this substance. Here, again, there has been a keeninternational rivalry and the credit of its synthesis must bedivided between this country and Germany. It is gratifyingthat the only original investigation of the subject which hasyet been published as a contribution to science is theadmirable account by Dakin of his synthesis not only ofadrenalin but of a whole array of substances which are

closely allied to this body in their chemical structure as

well as in their physiological influence on the animalorganism.

In order to comprehend the point of attack of adrenalin,the specific secretion of the medullary part of the supra-renal glands, we shall do well to go back to the mode ofdevelopment of these organs. It was shown by Balfour thatthe suprarenals have in the foetus a twofold origin, thecortex being derived from the mesoblastic tissue, known asthe intermediate cell-mass, while the medulla is formed bya direct outgrowth from the sympathetic system, and con-sists at first of an aggregation of neuroblasts. In someanimals-e.g., teleostean fishes-the two parts of the glandthus formed remain separate throughout life, but in the

higher vertebrates the sympathetic outgrowth becomes sur-rounded by the cortex and the cells rapidly lose all traces ofresemblance to a nerve cell. But the medulla is geneticallypart of the sympathetic system, and its specific secretion,adrenalin, has an action which is apparently confined to thesympathetic system. In whatever part of the body we testthe effects of adrenalin we find that they are identical withthe results of stimulating the sympathetic nerve fibres whichrun to that part. Thus, in all the blood-vessels of the bodyadrenalin causes constriction ; the contraction of the heartmuscle is augmented, the pupil is dilated, while the in-testinal muscle, with the single exception of the small ringof muscle forming the ileo-colic sphincter, is relaxed. Theaction of the sympathetic on the bladder differs, as

shown by Elliott, markedly in various animals ; but, what-ever its effect, a similar one will be produced in the sameanimal by the injection of adrenalin. I have already men-tioned that excision of the suprarenal bodies causes a

profound fall of blood pressure, which continues until thedeath of the animal, and it has been stated that when thisfall is well established it is impossible to raise the bloodpressure by stimulation of the splanchnic nerve or,indeed, to produce any effect at all on stimulation ofthe sympathetic nerve. Thus not only does adrenalin excitethe whole sympathetic system in its ultimate terminationsbut its presence in the body as a specific secretion of thesuprarenal bodies seems to be a necessary condition for thenormal functioning, by ordinary reflex means, of the wholesympathetic system. We are dealing here with a problemwhich, betraying, as it does, an intimate relationship betweennerve excitation and excitation by chemical means, promisesby its solution to throw a most interesting light on thenature of the nerve process and of excitatory processes ingeneral.Our knowledge of certain other members of this group of

chemical reactions is so shadowy that a mere mention ofthem will suffice. As an antithesis to the vaso-constrictoraction of adrenalin we find that every organ when active is

supplied with more blood in consequence of a vaso-dilatationof the vessels which supply it. In certain instances Baylissand I have found that boiled extracts of organs wheninjected into the circulation may evoke vaso-dilatation ofthe same organs of the animal under investigation and wehave suggested that the normal vaso dilatation accompany-ing activity is brought about in consequence of the specificsensibility of the arterial walls to the metabolites of theorgan which they supply. Too much stress, however, cannotbe laid upon these experiments since a more extended seriesby Swale Vincent has failed to give a general confirmationof our results.The severe diabetes which, as shown by Minkowski, can

be produced in nearly all animals by total excision of thepancreas has been held to denote the normal production inthis organ of some substance which is indispensable for theutilisation of carbohydrates in the body. All efforts toobtain a more exact idea of the nature of this pancreaticsubstance or influence have so far proved in vain. Ordinarysugar, when placed in contact with extracts of musculartissues, undergoes oxidation ; and Cohnheim states that thisprocess is much accelerated if an extract of pancreas beadded to the extract of muscle. A repetition of Cohnheim’sexperiments by other observers has shown that the effect isso small as to be almost accidental ; and we must thereforeregard the nature of the pancreatic influence on carbohydratemetabolism and the causation of pancreatic diabetes as

problems still to be solved.So far (except in the case of the muscle-respiratory centre

reaction) we have been dealing with isolated phenomenaoccurring in different parts of the body in which the re-action affects a whole series of tissues. The chemicalstimulus in these cases might be considered analogous toalterations in the composition of the surrounding mediumand the reactions lack that definite and localised characterwhich we have learnt to regard as distinguishing the adapta-tions or reflex actions brought about by the intermediationof the central nervous system. We have now to discuss themechanism of a whole series of chemical reactions wherethis feature of a nervous reaction is not wanting, so that, infact, the reactions, up to quite recently, were regarded asundoubtedly nervous in character. I refer to the chain ofprocesses in the alimentary canal by which the secretion ofone u_ice._s.uaoeeds that of another as the food progressesalong this tube. This series of reactions may well form thesubject of a separate lecture.