Vaccinated Mice Remain More Susceptible to Mycobacterium tuberculosis Infection Initiated via the Respiratory Route than via the Intravenous Route

Abstract

Mice given Mycobacterium tuberculosis bacilli via the respiratory route succumbed much sooner to infection than mice given 1,000 times more bacilli via the intravenous route. Vaccination provided increased protection to an M. tuberculosis challenge infection; however, mice infected via the respiratory route remained much more susceptible.

It was shown by a previous study (6) that mice given Mycobacterium tuberculosis via the respiratory route die much sooner than mice given 1,000 times more bacilli via the intravenous (i.v.) route. This was the case even though the number of bacilli that implanted in the lungs in each case was the same. It was suggested (6) that the more virulent behavior of M. tuberculosis given by aerosol is the result of the pathogen being directly ingested by alveolar macrophages that are more permissive of its growth than monocyte-derived macrophages, which presumably carry the organism into the lungs from the blood. It was further suggested that a smaller number of bacilli given by aerosol likely causes the development of less immunity, because it represents a much smaller antigenic load and engages less lymphoid tissue than a larger inoculum given i.v. The purpose of the study described here was to determine whether vaccinated mice that already have an acquired state of specific immunity are less capable of defending against infection initiated via the respiratory route than via the i.v. route.

The basic experiment consisted of challenging control and Mycobacterium bovis BCG-vaccinated B6D2 F₁ [(C57BL/6 × DBA/2)F₁] mice from the Trudeau Institute animal breeding facility with 8 × 10² M. tuberculosis bacilli by aerosol or with 5 × 10⁵ bacilli i.v. and monitoring the growth of the pathogen in lungs, livers, and spleens over time. Survival times were also recorded. Mice were vaccinated by giving them 10⁵ BCG Pasteur (TMC 1101; Trudeau Institute Culture Collection) bacilli i.v. and allowing the BCG infection to progress and to then partially resolve over a 50-day period. The mice were then put on a 10-day course of chemotherapy (200 mg of isoniazid and 100 mg of rifampin per ml of drinking water) to reduce BCG numbers. Enumeration of BCG bacilli in major organs of five BCG-vaccinated mice 10 days after commencing chemotherapy showed that chemotherapy had reduced the number of BCG bacilli per organ to below the detection level (10²) of the assay. Treatment was stopped 3 days before challenging the mice with the Erdman strain of M. tuberculosis (TMC 107) prepared for initiating infection via the respiratory or i.v. route as described previously (6). Infection via the respiratory route was performed with a Middlebrook airborne-infection apparatus (Tri Instruments, Jamaica, N.Y.). M. tuberculosis bacilli were enumerated by plating serial 10-fold dilutions of whole-organ homogenates on Middlebrook 7H11 agar.

The effect of BCG vaccination on the ability of mice to resist infection initiated via the respiratory route versus the i.v. route is shown by the bacterial growth curves in Fig. 1. It can be seen that mice infected by aerosol or i.v. inoculation started off with about the same levels of lung infection (8 × 10² to 10 × 10² bacilli) on day 1. In the case of unvaccinated mice, and in agreement with a previous publication (6), M. tuberculosis given by aerosol grew in the lung during approximately the first 20 days about 2 logs more than did M. tuberculosis given i.v. After this time lung infection was stabilized in the case of the former mice, or its rate of progression greatly reduced in the case of the latter mice, almost certainly because of the expression of acquired, specific immunity. The roles of CD4 Th1 and other T cells in anti-M. tuberculosis immunity are the subject of recent articles (1, 5, 7). Vaccination provided mice infected via either route with an ability to express immunity earlier in the lung and to thereby reduce the number of bacilli in this organ at day 40 of infection, by 1.5 logs in the case of mice infected by aerosol and by 1 log in the case of mice infected i.v. After this time lung infection was similar to lung infection in unvaccinated mice; i.e., it was stabilized in the case of mice infected by aerosol or reduced in rate in mice infected via the intravenous route. The situation in the liver and spleen was essentially the same in mice infected via either route, in that vaccination served to significantly reduce the level of infection in these organs. In the case of aerosol-infected mice this was achieved by a reduction in the level at which infection was allowed to establish itself from about day 20 of infection. In contrast, it was achieved in mice inoculated i.v. by inhibition of bacterial growth during the first 10 days of infection. It can be seen in Fig. 1 that in mice infected by aerosol, M. tuberculosis did not disseminate from the lungs to other organs until after about a 20-day delay.

FIG. 1.

Course of M. tuberculosis infection in the lungs, livers, and spleens of BCG-vaccinated (IMM) and unvaccinated (CONT) mice infected with 8 × 10² CFU of M. tuberculosis Erdman by aerosol or with 5 × 10⁵ CFU of M. tuberculosis Erdman i.v. Results are given as means ± standard errors (error bars) for five mice per group per time point.

The effects of vaccination on the survival times of mice infected via the respiratory route and the i.v. route are shown in Fig. 2. It can be seen, in agreement with a preceding study (6), that unvaccinated mice died much earlier from infection initiated via the respiratory than via the i.v. route. The results show that vaccination significantly extended survival times of mice infected via either route. However, vaccinated mice infected by aerosol died much sooner than vaccinated mice and unvaccinated mice infected i.v.

FIG. 2.

Survival times of 10 BCG-vaccinated (IMM) and 10 unvaccinated (CONT) mice infected with 8 × 10² CFU of M. tuberculosis Erdman by aerosol or with 5 × 10⁵ CFU of M. tuberculosis Erdman i.v.

Macroscopic examination of the organs of control and vaccinated mice infected via either route suggested, in agreement with previous observations (2, 6), that the cause of death in all cases was loss of pulmonary function due to infection-induced lung pathology. In all mice, lesions were visible on the lungs but not on other major organs. The extensiveness of infection-induced pathology in the lungs of the different groups of mice at day 50 of infection can be appreciated from an examination of photographs of crystal violet-stained, 20-μm-thick sections of the left lung lobe of a mouse from each group, as shown in Fig. 3. It is clear that the shorter survival times of aerosol-infected mice were associated with more extensive pathology and that vaccination served to retard the development of pathology. It is known from previous studies (3) that lung pathology continues to develop during the stationary phase of infection.

FIG. 3.

Photographs of 20-μm-thick, crystal violet-stained sections of the left lung lobes of nonvaccinated (a and b) and vaccinated (c and d) mice infected with M. tuberculosis by aerosol (8 × 10² CFU) (a and c) or i.v. (5 × 10⁵ CFU) (b and d), as described for Fig. 1 and 2. The darkly stained areas represent the most densely consolidated areas of lung tissue. There was more consolidation in nonvaccinated mice infected by aerosol. Vaccination served to reduce the rate of development of lung consolidation. Magnification, ×4.2.

The levels of lung protection afforded mice against M. tuberculosis challenge infection by BCG vaccination as shown by this study are similar to those shown previously by others for mice (4) and guinea pigs (9). The common finding has been that vaccination enables the host to reduce the level of lung infection but not to resolve it. An important point illustrated by this study is that BCG-induced immunity was not capable of being immediately expressed in the lungs and was not fully expressed in this organ until approximately day 20 of the challenge infection, which was somewhat earlier than the expression of immunity in unvaccinated mice. A similar observation was made with guinea pigs (9) some years ago. As pointed out by Smith (10), the duration of this delay is important because it provides the pathogen time to undergo numerous generations at sites of implantation and to reach relative large numbers at these sites before immunity is fully expressed. It is apparent that a similar delay occurs in DNA-vaccinated mice (4). Because bacterial growth was faster in vaccinated and unvaccinated mice infected by aerosol before immunity was fully expressed, the level of lung infection reached in these mice was higher than those in the lungs of vaccinated and unvaccinated mice infected via the i.v. route, hence, the reason for more rapid development of pathology and earlier death of the former mice. By enabling immunity to be expressed earlier, vaccination served to proportionally reduce the level of infection reached in each case. Thus far, these experiments have only been performed with B6D2 F₁ mice.