The Journal of Experimental Medicine

Gay, Frederick P.; Clark, Ada R.; Street, Julia A.; Miles, Dorothy W.

doi: 10.1084/jem.69.5.607pmid: 19870866

The precise mode of therapeutic action of sulfanilamide on streptococcus can be arrived at only by considering the sum total of factors that inhibit or favor the natural growth of the microorganism under the experimental conditions that obtain, whether in vivo or in vitro . Too sweeping conclusions have hitherto been drawn from the study of a single variable factor, such as an unfavorable temperature or the absence or presence of peptone. We have attempted here to analyze the factors that have hitherto been recognized and some new ones, but particularly the relationship of these factors to one another. The result obtained on adding sulfanilamide to the streptococcus in the test tube is usually bacteriostasis and not complete destruction of even small numbers of bacteria. This is on the condition that the suspending medium is a favorable one for the growth of the microorganism; the more growth-promoting the medium is the less the bacteriostasis. If, on the other hand, the medium is too poor, or one that in itself inhibits growth, the addition of sulfanilamide may lead to sterilization of the culture. The conditions for growth of the streptococcus in the body of the rabbit or mouse, depend on the strain of bacteria used, but are on the whole favorable. Defence, however, in the form of phagocytosis by both polymorphonuclear and by mononuclear cells is attempted even in the susceptible animal. When sulfanilamide is used to treat such an animal, or when sulfanilamide-grown (inhibited) streptococci are employed, phagocytosis is pronounced, whether studied in the test tube or in the animal body. In the rabbit the delay by sulfanilamide and resultant increased phagocytosis by polymorphonuclears allows mononuclear cells to accumulate and recovery may result. Sulfanilamide not only does not completely destroy the streptococcus but does not even impair its innate virulence. It acts upon the streptococcus not only by inhibiting growth but by a temporary inhibition of hemotoxin formation, but only under certain conditions. The drug does not neutralize hemotoxin already formed. No significant effect of sulfanilamide on the formation of leucocidin or fibrinolysin by streptococcus has been evident in our experiments. Sulfanilamide differs in one important respect from other drugs that are destructive either in the test tube or actually in the body, for protozoa and bacteria. Protozoa fix or adsorb arsenicals and acriflavine that kill them variably in vitro and in vivo . Streptococci fix both gentian violet and acriflavine, which dyes have marked destructive action in the test tube but are less effective in vivo . Sulfanilamide is not diminished at all by contact in vitro with large masses of streptococci, nor does the action of this drug render the microorganism more capable than untreated cocci to adsorb gentian violet or acriflavine, or to be destroyed by these highly bactericidal substances. Footnotes Submitted: 13 January 1939

Baker, Lillian E.

doi: 10.1084/jem.69.5.625pmid: 19870867

Experiments designed to ascertain the reason for the cessation of growth of heart fibroblasts when they are cultivated in a plasma coagulum with embryo juice as nutrient fluid have shown that it is due, first, to the gradual removal of serum from the coagulum, and second, to an insufficient supply of embryo juice. In a medium containing embryo extract at 66 per cent concentration and serum at 8 per cent concentration, growth continued until the entire coagulum in a 3½ cm. flask was covered with tissue. The serum is needed to furnish additional nutriment, and also to prevent digestion of the coagulum. Footnotes Submitted: 12 January 1939

Claude, Albert

doi: 10.1084/jem.69.5.641pmid: 19870868

It is known that azoprotein solutions, like testicular extracts, possess the property of causing particles to spread through the dermis. The present work shows that azoproteins exhibit, like testicular extract, the power to increase the size of virus lesions in the skin of rabbits, and the size of tumors in chickens. The results indicate that the extent of the lesion is roughly proportional to the spreading power of the solution. This suggests that the spread of the infective material, over a larger area of skin, is directly responsible for the enhancing effect. The production of extensive lesions by means of spreading agents may have a practical value when large amounts of working material are needed. Footnotes Submitted: 8 November 1938

Goldblatt, Harry; Kahn, Joseph R.; Hanzal, Ramon F.

doi: 10.1084/jem.69.5.649pmid: 19870869

Constriction of the aorta just above the origin of both main renal arteries invariably resulted in elevation of the carotid systolic and carotid mean pressure. The hypertension was not immediate, but developed in about the same time as after constriction of the main renal arteries (3). Constriction of the aorta just below the origin of both main renal arteries had no significant effect on the carotid systolic or carotid mean pressure. Since these results were first reported (1), Rytand (88, 89) has shown by an indirect method, namely, the demonstration of the development of cardiac hypertrophy, that hypertension in the upper part of the body can be produced in the rat by constriction of the aorta just above the origin of both main renal arteries. The immediate effect of constriction of the aorta either below or above the main renal arteries is a fall of blood pressure (femoral mean pressure) below the site of the clamp, the extent of the fall being directly dependent upon the degree of constriction of the aorta. Of particular interest is the eventual elevation of the femoral mean pressure above the normal in some animals with the aorta constricted or even occluded above the origin of the main renal arteries. This was most pronounced and persistent in those animals in which, in addition, the aorta below the origin of the renal arteries, and, in some animals, the main renal arteries, also were constricted. The most important factors which determined this elevation of blood pressure in the lower part of the body were probably increased flow of blood into the vascular bed below the clamp and peripheral vasoconstriction of renal and humoral origin, as in the case of the hypertension produced by constriction of the main renal arteries alone (2–86). Although elevation of the carotid systolic or carotid mean pressure occurred invariably within 24 to 48 hours after the constriction of the aorta above the site of origin of both main renal arteries, yet there was a tendency, after a variable period, for the elevated blood pressure to become lower or even to drop to the original level. Increased constriction, and finally occlusion of the aorta, above the origin of the main renal arteries, and even constriction or occlusion of the aorta below the renal arteries, in addition, failed to induce hypertension that persisted for a long time at a high level. In order to produce this effect, it was necessary to constrict the main renal arteries as well. The possible explanation of the failure of the hypertension to persist for a long time after constriction of the aorta alone, is that the initial ischemia of the kidneys disappeared due to the improvement of the blood flow through the kidneys as a result of ( a ) the increase of the natural accessory circulation to the kidneys; ( b ) the increased blood pressure above the site of the clamp and consequent increased flow of blood into the part of the aorta below the clamp; ( c ) increased pressure below the site of the clamp due, in great part, to peripheral vasoconstriction, and in part to the increased inflow of blood into the lower part of the body through the aorta and collateral channels. For the dog, this method is not necessary for the production of persistent hypertension. Constriction of the main renal arteries is easily performed and is effective for the production of generalized hypertension (2–11). However, constriction of the aorta in addition to constriction of the renal arteries results in greatly elevated persistent hypertension. Constriction of the aorta alone above the origin of the main renal arteries would be useful in the dog only for the production of relatively short periods of hypertension in the upper part of the body. For small animals it may be a more effective and useful method. In the dog, the only technical difficulty encountered was the erosion of the wall of the aorta by the clamp. This may not occur in small animals. In previous studies (2–11) that have dealt with the constriction of the main renal arteries, this accident rarely occurred. When the constriction of the aorta above the origin of the main renal arteries was of moderate degree, or was gradually made very great, the resultant hypertension was not accompanied by impairment of renal excretory function, as determined by urea clearance or by the quantity of urea, creatinine or non-protein nitrogen in the blood, the benign phase of hypertension (3). When the constriction of the aorta was suddenly made very great, impairment of the renal excretory function usually followed, and the animal developed fatal convulsive uremia and characteristic vascular lesions, the malignant phase of hypertension (9). These facts, are all indicative of the renal origin of the hypertension which results from the constriction of the aorta just above the origin of both main renal arteries. Hypertension did not persist for a sufficiently long time to permit any conclusive comparison between the effect of the high and low pressures on the structure of the vascular system, above and below the site of the clamp, respectively. During the period of survival of these animals, no significant differences were observed between the appearance of the vascular system of the upper part of the body and that of the lower part of the body, and significant cardiac hypertrophy did not develop. In the aorta and large arteries, intimal arteriosclerosis was not observed. In the aorta of one old animal several small plaques of calcification were found in the media, but these were present in the portion of the aorta below, as well as above the clamp, and they were no larger or more abundant than were observed in some old dogs with normal blood pressure. Dogs 3–50 and 3–83, that are still alive, with very high blood pressure above the site of the aortic clamps, and relatively low pressure (though greater than normal) below the site of the aortic clamps, will be valuable for the determination of possible differences between the effects of the two levels of blood pressure in the large and small blood vessels. In these dogs also, it will be possible to determine the effect of the persistently high blood pressure on the myocardium. The possible application of the results of this study to the problem of the pathogenesis of human eclampsia is mentioned here for consideration. Since this condition occurs in pregnancy only at a time when the uterus is greatly enlarged, it is at least possible that the mass may press on the aorta or both main renal arteries sufficiently to produce renal ischemia. The suddenness with which the uremic convulsive phase of eclampsia develops is in keeping with this idea. In the dog, an aggravating effect of pregnancy on an already established hypertension has not been noted. As a matter of fact, most of the hypertensive dogs that have become pregnant, have shown a slight or moderate fall, rather than an increased rise of pressure. Since the dog stands with the body in a horizontal position, and does not lie on its back, pressure of the pregnant uterus on the aorta and blood vessels is less than in human beings who stand erect and frequently lie on their backs. The soundness of this suggestion could be tested by placing pregnant women, in the early stage of eclampsia, in a position which could relieve possible pressure on the aorta and main renal arteries. A possible explanation of the fall of pressure in the pregnant hypertensive dogs is the compensatory effect of the normal kidneys of the pups, as in the case of an animal with one main renal artery constricted and the other kidney normal. As has been shown (3, 31, 72), the presence of one normal kidney in an animal hypertensive due to constriction of the other main renal artery, results, after a variable period, in a return of the blood pressure to normal. How the normal kidney acts to produce this effect is not known. Footnotes Submitted: 18 January 1939

King, Lester S.

doi: 10.1084/jem.69.5.675pmid: 19870870

After inoculation with equine encephalomyelitis virus by various routes, guinea pigs were sacrificed at early stages, before symptoms were apparent. The brains were studied histologically, with serial sections; all lesions were noted, and subjected to topographical analysis. Nine cases are presented in detail. With any given mode of inoculation the distribution of lesions varied very widely from one instance to another. In some cases, affected regions bore a striking and definite anatomical relationship to each other. These distributions can be explained only by the assumption that the anatomical pathways played some rôle in the spread of the virus. In other instances lesions were present in areas, the anatomical connections of which were entirely normal. Attention is called to the frequency of lesions in the neocortex, with intact subcortical centers. Such distribution is held to render nerve spread extremely improbable. The only satisfactory explanation of such random distributions is by direct passage of virus from the blood stream into the brain tissue. There is no histological difference between lesions which result from blood spread and those resulting from nerve spread. Footnotes Submitted: 4 January 1939

King, Lester S.

doi: 10.1084/jem.69.5.691pmid: 19870871

The behavior of a fixed strain of Eastern equine encephalomyelitis virus was studied in guinea pigs after intraocular inoculation. Such inoculation concerns the central and not the peripheral nervous system. The susceptibility to intraocular injection lies midway between the highly virulent intracerebral and the quite avirulent peripheral routes. The virus must act for 10 to 13 hours in order to induce a fatal infection. Removal of the inoculated eyeball before this interval almost always prevents fatality although it may allow immunity to develop. The virus, at suitable intervals after injection into the eye, may be recovered from successive and appropriate optic centers before it is demonstrable in non-optic portions. Approximately 24 hours are required for the virus to reach a significant concentration in the contralateral geniculate body, 36 hours in the contralateral visual cortex. Significant amounts of virus may be present in the optic chiasm and tract prior to involvement of the higher centers. Virus placed in contact with the retina produces an insignificant, essentially non-specific reaction comparable to that produced at the site of direct intracerebral inoculation. In the retina there is no ganglion cell necrosis unless there is a complicating intraocular infection. In the cerebral visual centers the first reaction is inflammatory and interstitial, and may appear in the lateral geniculate body as early as 24 hours after injection. Neuronal necrosis is not the primary action of the virus on the nervous system in these experiments. The distribution of lesions in the brain is in excellent agreement with the method of direct testing for virus content, and is far more accurate than the latter. The virus in its primary distribution through the nervous system follows the nerve pathways of the optic system. This occurs within the central nervous system, where presumably there is first an involvement of the nerve cell body and then a spread along the cell process or axone. Footnotes Submitted: 23 January 1939

Landsteiner, K.; van der Scheer, J.

doi: 10.1084/jem.69.5.705pmid: 19870872

Experiments are described dealing with immune sera to pentapeptides and peptide amides. Absorption and inhibition tests gave no indication of the presence in the immune sera of special antibodies for portions of a peptide molecule but the antibodies appeared to be specific for an entire pentapeptide even though the sera contained qualitatively different fractions. Marked disparity was found between the reactions of peptides and corresponding amides indicating differences between acid and other polar groups in their influence on serological specificity. Footnotes Submitted: 15 February 1939

Madden, S. C.; Noehren, W. A.; Waraich, G. S.; Whipple, G. H.

doi: 10.1084/jem.69.5.721pmid: 19870873

When blood plasma proteins are depleted by bleeding with return of the washed red blood cells (plasmapheresis) it is possible to bring dogs to a steady state of hypoproteinemia and a uniform plasma protein production on a basal low protein diet. These dogs are clinically normal. By the introduction of variables into their standardized existence insight into the formation of plasma proteins can be obtained. The liver basal diet maintains health in such hypoproteinemic dogs during periods as long as a year. 17 to 27 per cent of its protein content (entirely liver protein) is presumably converted into plasma protein. Gelatin alone added to the liver basal diet causes very little if any extra plasma protein production. The addition to gelatin of cystine, or tyrosine, or tryptophane, or of both tyrosine and tryptophane has little or no effect on its potency for plasma protein production. When gelatin is supplemented by cystine and either tryptophane or tyrosine, 25 to 40 per cent of the protein content of the combination is converted into plasma protein—an efficiency equaling that of any protein hitherto tested. Preliminary experiments indicate that methionine cannot substitute for cystine nor can phenylalanine substitute for tyrosine in the efficient combination of gelatin plus cystine plus tyrosine. Laked red blood cells given by vein afford little or no material for plasma protein formation. When the reserve stores of plasma protein building material are exhausted the dog can form little if any plasma protein during protein-free diet periods. Footnotes Submitted: 12 January 1939

Hahn, P. F.; Bale, W. F.; Lawrence, E. O.; Whipple, G. H.

doi: 10.1084/jem.69.5.739pmid: 19870874

Artificially produced radioactive iron is an extremely sensitive agent for use in following iron in the course of its changes in body metabolism, lending itself to studies of absorption, transport, exchange, mobilization, and excretion. The need of the body for iron in some manner determines the absorption of this element. In the normal dog when there is no need for the element, it is absorbed in negligible amounts. In the anemic animal iron is quite promptly assimilated. The plasma is clearly the means of transport of iron from the gastrointestinal tract to its point of mobilization for fabrication into hemoglobin. The speed of absorption and transfer of iron to the red cell is spectacular. The importance of the liver and bone marrow in iron metabolism is confirmed. Footnotes Submitted: 17 February 1939