In the preceding pages we have presented evidence which we believe furnishes new light on the disease process in avian tuberculosis. From a well known strain of avian tubercle bacillus, A1, four variants have been dissociated, each manifesting distinct colony topography and physical and chemical characteristics. From these studies we have learned that the variants are sometimes unstable, not only in vitro but also in vivo, and that this characteristic is one of the prominent factors influencing both the advancement and retrogression of the disease.
The four variants remain fairly stable in vitro provided they are cultivated on proper culture media. About 80 per cent smooth S and flat smooth F.S. colonies will develop true to type on egg media; and rough R and the chromogenic Ch when cultivated on gycerine agar media, in about the same percentage.
An early non-specific eosinophilia followed inoculation, no matter which variant was used. This usually subsided by the 18th hour.
The early stages of tubercle formation produced by all the variants appeared similar. First there appeared an aggregation of eosinophiles and their ingested bacilli within the tissues and then followed replacement by large mononuclear cells which wandered in and phagocyted both eosinophiles and bacilli. After the formation of tubercles composed only of large mononuclear cells, certain differences between the virulent and avirulent variants became apparent.
There was also a direct relationship between the dosage and the extent of the disease. In our experience from 0.16 to 0.25 mg. gave the most uniform results.
If S variant was used, the early non-specific eosinophilia was followed by a second rise which continued to ascend, running parallel with the total leucocyte count. There was a slow increase of monocytes and a corresponding decrease of lymphocytes. The microscopic lesions, 2 weeks after inoculation, were composed of irregularly shaped clumps of necrosing large mononuclear cells. The margins were clear-cut and bordered by few or no lymphocytic cells. By the 3rd and 4th weeks eosinophiles began to migrate into the centers of the tubercles and abscess formation became evident. Death usually occurred after 5 or 6 weeks, although this varied with the dosage. The appearance of the lesions suggested an acute "toxic" nature, as manifested (1) by marked enlargement of the spleen and liver, (2) by the short fatal course of the disease, which never became very extensive, (3) by the presence of few organisms within the tubercles, (4) by destruction of reticulum and (5) by the marked response of the blood leucocytes.
After inoculation with F.S. variant the early non-specific eosinophilia disappeared, returning again at the terminal stage, 5 to 6 weeks later. The number of eosinophiles then ascended sharply, accompanied by a marked leucocytosis which continued until death. During this eosinophilia a monocytosis occurred and lymphocytes fell away rapidly.
During the 2nd week, clumps of closely packed large mononuclear cells containing masses of phagocyted bacilli were seen. In the 3rd and 4th weeks there appeared a prominent peripheral zone of hyalin, collagenous-like material associated with an increased reticulum formation which apparently walled off the lesion. The bacilli within the tubercle multiplied rapidly, it seemed as though they were more resistant to destruction than the S organisms, perhaps due to their higher lipin content. As a result of their rapid increase, progression of the tubercle continued, with necrosis and abscess formation, accompanied by the dissemination of the organisms both locally and distantly to new uninvolved tissue. Wide extension of the disease followed due to the increase in numbers of bacilli, terminating with death usually during the 5th and 6th weeks. The manifestations of "toxin" damage shown in S disease were not so marked until the approach of the end-stage. We interpret the type of lesion produced by the F.S. variant as a foreign body tubercle to distinguish it from the acute "toxic" nature of the S. A quantitative chemical determination showed that this organism contained 20.14 per cent lipin whereas in the S variant it was only 15.03 per cent.
Following inoculation with R variant the initial eosinophilia returned to normal within 24 hours. There was then an increase of both lymphocytes and monocytes, the latter exceeding the former for a time. Later, usually after the 2nd month, the lymphocytes again became predominant over the monocytes, approaching the normal base line. Eosinophiles failed to react. This blood picture indicated a successful resistance against the infecting organisms, and was supported by the pathological findings. None of the chickens of this group died during the period of study (79 days) although infecting doses of 0.25 mg. were used. Necropsies on four chickens revealed no macroscopic evidence of disease approximately 11 weeks after inoculation. Microscopic lesions were present in the spleen and liver of some cases. In chickens inoculated 11 weeks previously these tubercles consisted only of very small, discrete clumps of degenerating mononuclear cells often surrounded by a border of lymphocytes. Necrosis, abscess formation or caseation was not found. Their small size and appearance was evidence of their retrogressive character. Isolated giant cells were rarely seen. It was impossible to find bacilli within these tubercles. In one instance it was possible to show complete innocuousness of this organism in the tissues, as far as producing recognizable tuberculous lesions. Reticulum played little part in the development of the lesion and it was neither increased or destroyed.
Following inoculation with the chromogenic Ch variant the initial eosinophilia declined by the 24th or 42nd hour count and from then on remained low. The lymphocytes showed little decrease. A prominent monocytosis of prolonged character usually occurred during the 3rd to 6th weeks and was then displaced by a return of the lymphocytes. By the end of the period of observation the differential count had closely approached the normal values. Again the blood picture was truly representative of the relatively benign character of the lesions. The tubercle formation, even after 93 days' duration of the disease, remained limited to small clumps of two or three large mononuclear cells showing some degeneration. Lymphocytic cells were usually present within and about the lesion. There was never any evidence of abscess formation or actual necrosis. Bacilli could not be found within these tubercles. The reticulum in relation to the lesion showed neither increase nor destruction. None of the chickens died though a dose of 0.5 mg. was used.
In three chickens infected with S variant, the blood picture did not follow the pattern set by the others similarly inoculated. Either a high monocytosis replaced the usual eosinophilia, or there was a marked recession of the total leucocyte count followed by a return of lymphocytes towards normal percentage. The blood picture indicated the conversion of an acute process into a subacute or chronic affair. These chickens survived for 78, 79 and 110 days, respectively. An explanation for this occurred to us later when cultures recovered from them, instead of consisting of S colonies, showed the topography of the intermediate type, closely resembling the F.S. type. Evidently this was due to a reversion; i.e., a loss of virulence of the organism occurring within the animal body. At necropsy the spleen and liver were somewhat enlarged and contained prominent, discrete yellow nodules resembling small shot in their size and shape, which, in one case could be lifted from the tissue by the point of the knife. Microscopically, they possessed small centers of caseation surrounded by numerous giant cells and large mononuclear cells with a thick border of lymphocytic cells. This tubercle formation could not be classified in any special group as it was too sharply circumscribed and walled off by lymphocytic cells to be considered as truly malignant in nature as the S type of lesion. Even the younger tubercles in these cases, which consisted of clumped epithelioid cells with prominent lymphocytic cell borders, could not be called acute or "toxic" in character.
At present we make only limited deductions from our observations. In the bacterial dissociation phenomenon we have at least a new line of investigation, and different variants of tubercle bacilli must be taken into consideration when planning new tuberculosis studies. With the stabilization of the human type variants, experiments such as reported here may bring about the proper clinical interpretation, of the course of human tuberculosis.