Early Response of Mouse Foot Pads to Mycobacterium Leprae

Transcription

1 INFECTION AND IMMUNrrY, Jan. 1973, p Copyright American Society for Microbiology Vol. 7, No. 1 Printed in U.S.A. Early Response of Mouse Foot Pads to Mycobacterium Leprae MICHAEL J. EVANS, HAROLD E. NEWTON, AND LOUIS LEVY Life Sciences Division, Stanford Research Institute, Menlo Park, California 94025, and Leprosy Research Unit, Public Health Service Hospital, San Francisco, California Received for publication 18 August 1972 The purpose of these experiments was to study the early response of mouse foot pads to Mycobacterium leprae. To accomplish this, mice were inoculated in both foot pads with large and small numbers of organisms. The animals were sacrificed at intervals from 2 hr to 27 days after inoculation. The microscopical results, which utilized normal BALB/c and thymectomized-irradiated B6C3F1 mice, showed that the tissue responded first with an influx of polymorphonuclear cells and later lymphocytes and monocytes. The latter formed a diffuse infiltrate in the tissues. Under conditions where growth normally occurred, the mononuclear cell infiltrate did not persist. The organisms were found within phagocytic cells and the interstitial space. They were always contained within a phagosome and often fused with lysosomes. Most of the organisms appeared to be degenerating at all of the times studied. No organisms were observed in striated muscle fibers of tissues studied. The mouse foot pad infection with Mycobacterium leprae has proven to be a suitable system in which to study the host-parasite relationship in leprosy. After inoculation of a small number (103-7) of M. leprae into mouse foot pads, a lag phase occurs, after which the organisms multiply until a "ceiling" of about 106 organisms per foot pad is reached. At this time, histopathological changes in the foot pad take place, multiplication of M. leprae ceases, and killing of the organisms begins (14). It has been suggested that the ceiling effect is the result of an immune response. At this stage of the foot pad infection, morphological changes, interpreted to indicate activation of the macrophages (3), may be seen to occur in macrophages that contain organisms. Rees (12) and Shepard and Congdon (16) have demonstrated that the ceiling may be raised by two or three orders of magnitude if adultthymectomized mice, first lethally irradiated and then transfused with syngeneic bone marrow, are inoculated. If immunologically competent mice are inoculated with a large amount of M. leprae (about 106 organisms per foot pad), no multiplication occurs (13, 14). Multiplication of a large inoculum of M. leprae does occur in immunosuppressed mice (13). Lefford (5) has recently reported a similar relationship between inoculum 76 size and subsequent multiplication of BCG organisms in mice, and concludes that this relationship has an immunological basis. The current study (presented in part at the 7th Annu. Leprosy Conf., Menlo Park, Calif., 29 February-2 March 1972) was undertaken in an attempt to learn whether the morphological changes in mouse foot pad macrophages containing M. leprae, which occur only several months after a small inoculum, are present early after a large inoculum. If this is true, it would lend support to the hypothesis that the failure of multiplication of a large inoculum of M. leprae results from the early initiation of a cell-mediated immune response. MATERIALS AND METHODS Locally bred BALB/c mice and B6C3F, thymectomized and irradiated (T+R) mice obtained from C. C. Congdon, Oak Ridge National Laboratory, Oak Ridge, Tenn., were used in this study. The M. leprae were all from the same strain of organisms originally isolated from a patient with lepromatous leprosy by C. C. Shepard, Center for Disease Control, Atlanta, Ga., and carried since then in mouse passage both in Shepard's laboratory and in San Francisco. The M. marinum culture was a gift from Arthur Back, Microbiology Laboratory, San Francisco City-County Health Department. The mice were inoculated in both hind foot pads. Foot pads were prepared for study both in San Francisco and in Menlo Park (3). Those samples

2 VOL. 7, 1973 RESPONSE OF FOOT PAD TO M. LEPRAE 77 taken in San Francisco were fixed in 2% glutaraldehyde buffered with cacodylate to ph 7.2 for as long as 24 hr. They were then taken to the Menlo Park laboratory, where they were washed in buffer and postfixed in a 1% OsO4 solution containing 7.5% sucrose buffered with Veronal acetate to ph 7.2. The samples taken in Menlo Park were fixed directly in the above OS04 solution. After fixation, the tissues were dehydrated in a graded alcohol series and embedded flat in BEEM capsules in Araldite or Spurr low-viscosity embedding media. Specimens for light microscopy were sectioned with a glass knife on a Porter Blum MT2 ultramicrotome at 1.0 jm, placed on glass slides, and stained with toluidine blue. Specimens for electron microscopy were sectioned with a diamond knife, placed on collodion-coated nickel grids, stained with uranyl acetate and lead citrate, and observed with a Philips 200 electron microscope. In all the animals studied, the cellular response of the tissue to the inoculum involved an infiltrate of polymorphonuclear (PMN) cells and mononuclear (MN) cells. The cellular infiltrate was confined primarily to areas of loose connective tissue around muscles and nerves in the area of the inoculation. Using a light microscope, the degree of this response was estimated in each preparation and scored +, + +, or +++. A score of + denotes a sparse cellular infiltrate, and a score of +++ denotes a dense infiltrate. An example of a tissue preparation scored + + is shown in Fig. 1. Harvests of M. Ieprae from mouse foot pad tissue and the staining and counting of the organisms were carried out in the San Francisco laboratory as described previously (3). Measurement of the proportion of viable organisms in each harvest in experiment 1 was performed by mouse passage (6). The bacterial suspension resulting from each harvest was diluted so as to yield a concentration of organisms per 0.03 ml. Groups of BALB/c mice ("passage mice") were inoculated, each in both hind foot pads, with the diluted suspensions. At intervals thereafter, harvest of the pooled tissues of four foot pads were performed, yields of organisms were counted, and growth curves were constructed, assuming an average doubling time during logarithmic multiplication of 13 days for this strain of M. leprae in these mice. Then, assuming a lag phase of constant duration, the "time to plateau"-the time in days from passage until multiplication of the M. leprae had achieved a level of 106 per foot pad in the passage mice-was calculated for each passage. RESULTS Experiment 1. In the first experiment, the tissue response to a large inoculum of M. leprae in unvaccinated BALB/c mice was compared with that of mice vaccinated with BCG. Mice were randomized between two groups. One group of mice received four weekly subcutaneous injections of to 1058 viable BCG (Glaxo) in the inguinal areas. One week FIG. 1. Cellular infiltrate in the loose connective tissue of mouse foot pads 3 days after inoculation of 106 M. leprae. This infiltrate was graded + +. x 125. after the fourth injection, the mice of both groups were inoculated in each hind foot pad with M. leprae freshly harvested during logarithmic multiplication. At the intervals shown in Table 1, harvests were performed of the pooled foot pad tissues from two mice in each group, and the organisms from each harvest were passaged. Foot pads from two additional mice from each group were prepared for microscopical study at the same time intervals. The passages performed on the day of inoculation (day 0) were not from harvests of mice inoculated with the large inoculum. They are from dilutions of the large inoculum to 5 x 103 ; these passages to unvaccinated mice provided a measure of the proportion of viable M. leprae in the inoculum; the passage on day 0 to vaccinated mice demonstrated that these mice had been effectively vaccinated. The viability did not differ significantly between organisms harvested from unvaccinated mice and those recovered from vaccinated mice 1 day after inoculation. The time to

3 78 TABLE 1. EVANS, NEWTON, AND LEVY INFECT. IMMUNITY Response of unvaccinated and BCG-vaccinated mice to a large inoculationa of M. Ieprae Results of passages in unvaccinated mice Results of passages in vaccinated mice Days after inocula- AFBb AFB AF Time to AFB AFB AFB Time to tion recovered/ inoculated/ Days after recovered/ plateau recovered/ inoculated/ Days after recovered/ plateau foot pad foot pad passage foot pad (days) foot pad foot pad passage foot pad (x10') (x103) (x105) (x105) (X103) (xl10) (days) OC d _ c Data for day 0 are not the results of harvests from mice inoculated with the large inoculum, but represent results of harvests from additional unvaccinated and BCG-vaccinated mice inoculated with a dilution of the large inoculum to provide a measure of viability of the large inoculum, and to demonstrate the effectiveness of vaccination. dorganisms recovered in this harvest were passaged to measure viability; time to plateau for this passage was 160 days, demonstrating viability of a proportion of the harvested organisms, and proving that the M. leprae had indeed multiplied in the vaccinated mice, despite the small harvest. a M. leprae per foot pad. b Acid-fast bacilli. plateau was 127 days for the inoculum, 127 days for organisms harvested from unvaccinated mice after 1 day, and 140 days for organisms harvested from vaccinated mice. Differences of more than 16 days between the times to plateau for two growth curves are significant at the 95% level of confidence. The passages of organisms obtained by harvests from both groups of mice 5 days after inoculation demonstrate that the proportion of viable organisms had decreased (time to plateau was 157 days for organisms from unvaccinated mice and 167 days for those from vaccinated mice). No dramatic change in the time to plateau was encountered in the passages of organisms harvested from either unvaccinated or vaccinated mice more than 5 days after inoculation. The average time to plateau for the seven passages from unvaccinated mice made at 5 days or more after inoculation is 168 days. The difference of 41 days between this value and the values for the inocu-

4 VOL. 7, 1973 lum and the harvest on day 1 is equivalent to 3.15 doublings, and represents the loss of 89% of the viable organisms present initially. The values for the passages from the vaccinated mice yield an almost identical comparison. The results of the histopathologic study of the foot pad tissues of these mice are presented in Table 2. The cellular response was essentially the same in both unvaccinated and vaccinated mice. At 24 hr, a moderate infiltrate of PMN cells and a slight infiltrate of MN cells occurred. By 5 days, the PMN cell infiltrate had declined, but the MN cell infiltrate had increased. At 9 days, there was a large MN cellular response that persisted for the duration of the experiment. The cells in the MN cellular infiltrate were mainly MN macrophages together with a few small lymphocytes. Experiment 2. Because an inoculum of M. leprae contains a large amount of mouse tissue debris, it is possible that the observed cellular response resulted from the mouse tissue and not from the M. leprae. To test this possibility, inocula were prepared from foot pads containing no M. leprae and also from a suspension containing nonviable M. leprae (organisms stored at 4 C for a number of weeks). An additional inoculum of heat-killed M. marinum was prepared to observe the effect of organisms with no contaminating cellular material. The results of this experiment are summarized in Table 3. In the case of the inoculum containing tissue debris but having no organisms, there was a slight MN cellular response at day 1 which declined and was TABLE 2. Effect of inoculation of 106f M. leprae in foot pads of BALBIc mice Unvaccinated mice Vaccinated mice Days after inoculation PMN MN PMN MN RESPONSE OF FOOT PAD TO M. LEPRAE no longer present at 7 days. In the case of the inoculum containing tissue debris plus M. leprae and that of the inoculum of M. marinum not contaminated by cellular material, the response was the same as that reported in experiment 1-namely, a PMN cellular response followed by MN cells that persisted for at least 7 days. Thus, the cellular response in tissues inoculated with M. Ieprae results from the organisms and not from the mouse tissue debris in the inoculum. Experiment 3. The tissue response to a large inoculum of nonviable M. Ieprae in T + R mice was studied in experiment 3 (see Table 4 for results). Initially, there was a slight PMN and a moderate MN cellular infiltrate. The MN cellular response persisted for 8 days but declined by 7 days and was no longer present at 9 days. This is in contrast to the results in the immunologically competent BALB/c mice, in which the MN cellular response persisted. Experiment 4. Immunologically depressed and competent animals were inoculated with 10-7 M. leprae per foot pad; the results from this experiment are presented in Table 5. In both kinds of animals, a slight PMN and MN TABLE 3. Effect of inoculation of tissue debris into foot pads of BALBIc mice Days Control 106 Dead 10 Dead after (no organisms) M. Ieprae M. marinum inoculation PMN MN PMN MN PMN MN ~~~0++ TABLE 4. Effect of inoculation of 106 M. leprae in foot pads of thymectomized, irradiated mice Time after inoculation PMN MN 2 hr day days days days

5 80 EVANS, NEWTON, AND LEVY INFECT. IMMUNITY TABLE 5. Effect of inoculation of M. leprae in foot pads of mice Competent Thymectomy Days after + irradiation inoculation PMN MN PMN MN t t, cellular response occurred at day 1. There was no further increase in cellularity, and the infiltrate was no longer present by 9 days. Types of cells containing organisms. There was no difference in the types of cells containing organisms among the different groups of animals. One hour after inoculation, the organisms were found in MN cells and also lying free in the loose connective tissue. At day 1, when there was a large influx of PMN cells and, to a lesser degree, of MN cells, the organisms were found primarily in MN cells. Very few were found in PMN cells or free in the tissue. As the MN cellular response increased, the organisms were found dispersed throughout the infiltrate within the MN cells (Fig. 2). Most of these cells contained only a few organisms. However, occasional cells were found that contained many organisms. In the T + R mice administered the large inoculum of M. leprae, the MN cellular response declined, and most of the organisms were found in the small groups of MN cells that remained in the tissue (Fig. 3). In the animals administered organisms, only an occasional organism could be found in MN cells. No organisms were observed in striated muscle cells, endothelial cells of capillaries, or Schwann cells in any animals, whether injected with the large or the small inoculum. Organisms within cells. The phagocytized organisms were found in the cytoplasm of MN cells. No difference was observed in the appearance of the organisms or in the host cell (cytoplasm) of specimens inoculated with either freshly harvested or stored organisms. In both cases, some organisms were observed that had not degenerated, but most appeared to be degenerating (Fig. 4). A considerable amount of cell debris, presumably from the inoculum, and lipid droplets were frequently observed in the cells (Fig. 5). Phagocytized M. * ~ ~~ X FIG. 2. Mononuclear cell infiltrate in the loose connective tissue around a large nerve. x ,: N' 0 4- :-W.A..!rt.,<., 9,.. 4 t FIG. 3. M. leprae in mononuclear cells 7 days after inoculation. x 1,250. leprae were surrounded by a phagosome membrane that appeared to consist of two layers (Figs. 5-7). The macrophages that contained organisms usually also contained many lysosome-like structures, and many instances were observed of the association of M. leprae in the phagosome with lysosomes (Fig. 6 and 7). The lysosomes themselves were surrounded by a single membrane and contained an homogeneous, dense granular material. I. w. UAL zwm

6 VOL.7, 1973 RESPONSE OF FOOT PAD TO M. LEPRAE 81 a- V.1, F.1 / I 'U i.,3 /.<.rs *\s W tb' ^^ FIG. 4. Electron micrograph of a cellular infiltrate showing cells containing phagocytized M. leprae (ML), cellular debris (CD), and lipid droplets (L). x8,400. DISCUSSION Previous work in this laboratory showed that changes occur at the plateau of multiplication of M. leprae which are associated with the onset of cellular immunity (3). At the tissue level, the number of MN cells increases, and the cells form a diffuse infiltrate in areas of the foot pads containing organisms. This

7 82 EVANS, NEWTON, AND LEVY INFECT. IMMUNITY I,.*-' s..~~~~~~~~~~~~~~14 _~~~~~ asa FIG. 5. Phagocytized M. leprae within a cell. Organisms are bounded by a phagosome membrane. Most organisms appear to be degenerating. x 22,750. infiltrate persists from plateau to the end of the plateau, when the macrophages become acstudy. tivated, a membrane forms around them that At the cellular level, other changes occur appears to be identical with the membrane (3). Before plateau, the organisms are found usually seen around phagocytized organisms. free in the cytoplasm of MN cells, but at Also the number of lysosomes increases and

8 VOL. 7, 1973 RESPONSE OF FOOT PAD TO M. LEPRAE 83 I. i"-, 4.' f,. IRWAMON"', 3 c luaw Ir I J{ 0! I.., *.., I'd I I..._. LS.L-. I.....,.AI, 'A' 6 Downloaded from on November 24, 2018 by guest FIG. 6 and 7. M. erating x22,7,50. 7 Ieprae associated with lysosome-like structure (LS). Organisms appear to be degenthe breakdown of organisms begins. These histological results are similar to those of the present study in two respects: (i) the persistence of an MN cell infiltrate in both studies during killing of organisms and (ii) the indication that MN cells containing organisms are activated during killing of organisms. Following inoculation of about 106 M. leprae in foot pads of immunologically intact mice, we found an inflammatory response and persis-

9 84 EVANS, NEWTON, AND LEVY INFECT. IMMUNITY ting MN cell response that was typical of that caused by other persisting antigen (17). These results are in agreement with the work of Weddell et al. (18), who also reported a PMN cellular infiltrate followed by an infiltrate of MN cells in the early response to both large and small inocula of M. leprae in CBA mice. Under these conditions, multiplication of organisms does not usually occur, and the first experiment of this study demonstrates the killing of some 90% of the viable organisms present in the inoculum within the first 5 days after inoculation. Under conditions in which multiplication of M. leprae ordinarily occurs, i.e., immunologically depressed mice receiving large and small inocula and immunologically competent mice receiving small inocula, the MN cellular infiltrate does not persist. The formation and persistence of a mononuclear cell infiltrate is associated with the onset of cellular immunity (7, 8). Its presence following inoculation of 106 M. leprae in immunologically intact mice of this study, and when multiplication of M. Ieprae has reached plateau (2), suggests that the immunological basis for the killing of M. leprae is similar in both studies. Two additional points concerning the tissue response to large inocula are the lack of difference in (i) the tissue response of vaccinated and unvaccinated mice and (ii) the response to viable and nonviable inocula. It had been anticipated that vaccinated mice would initially respond more vigorously than unvaccinated ones; however, no difference could be detected. An explanation may be found in a recent report by Lefford (5), who showed that, with increasing doses of BCG vaccination, there was a decreased latency of the immunological response. It is possible that M. leprae represented an inoculum large enough to decrease the latent time, so that a difference in tissue response between vaccinated and normal mice could not be seen at the intervals studied. That no difference was detected in the response of the tissue at the cellular or subcellular level to the viable and nonviable inocula may be explained by the work of McRae and Shepard (7, 15) who showed, by the criterion of the solid ratio, that mouse foot pad harvest material contains a high proportion of nonviable M. leprae. Thus, the tissue response to a viable inoculum may be directed mainly against the nonviable organisms present. Because the proportion of viable organisms is small, the reaction to the nonviable portion of the inoculum might then be as great as the reaction to an inoculum containing only dead M. leprae. The second similarity between the previous study (3) and this one was the indication that MN cells containing organisms appeared activated during killing of organisms. In the present study, most organisms were found in the MN phagocytes of the infiltrate. In an immune response, ordinarily there is participation of both lymphocytes and MN cells resulting in activated MN cells, which in turn destroy the organisms (7, 9). In this experiment, lymphocytes were observed in the infiltrates in all experiments, and the MN cells containing organisms appeared activated (i.e., appearance of lysosome-like structures and degenerating organisms). The M. leprae in the activated MN cells were surrounded by two membranes. The outer membrane shows unit membrane construction and is thought to be the phagosome wall (1). The inner membrane (an electrondense line that does not show unit membrane construction) is thought by Armstrong and D'Arcy Hart (1) to be the outer layer of a capsule that appears to be continuous with the cell wall of the organisms. The origin of this membrane has not yet been established. In a previous study (3), it was shown that during the logarithmic phase of multiplication the organisms were surrounded by a single membrane morphologically the same as the inner membrane just described. (The fate of the original phagosome membrane is not known at this time.) When the plateau of multiplication was reached, macrophages containing organisms became activated, and a second membrane formed around the organisms. This membrane appears to be identical with the phagosome wall. At that time, there were also examples of organisms associated with lysosomes. These organisms, surrounded by a double membrane, appear the same as phagocytized M. leprae reported in the present study. Under these conditions, which have been shown to have an immunological basis, killing of organisms was measured in both studies. Thus the histological evidence suggests there is an early immune response following inoculation of large numbers of M. leprae in mouse foot pads. Although the macrophages appeared to be activated and killing of organisms has been demonstrated, it is not known if the degeneration of organisms reported here resulted from macrophage action, or was because most of the organisms in the inocula were nonviable and al-

10 VOL. 7, 1973 RESPONSE OF FOOT PAD TO M. LEPRAE ready degenerating. It is probably a combination of both factors, because macrophages may be much more capable of action on nonviable M. leprae than on viable organisms. This was indicated in recent studies by Armstrong and D'Arcy Hart (1) who showed that phagosomes around nonviable M. tuberculosis fused readily with lysosomes, whereas fusion was encountered infrequently with viable organisms. Studies of this sort concerning M. leprae are not yet available, however D'Arcy Hart et al. (4) reported similar findings for M. microti but not M. lepramurium. We found no M. leprae in muscle cells during this study. In the previous study (3), they were found in muscle only during the plateau phase and then only in small numbers. It has been shown that M. leprae may enter muscle tissue soon after inoculation but only in small numbers (2, 18). Esiri et al. (2) reported only a few M. leprae lying in and between muscle fibers of normal and T+R mice for the first 3 months of the infection. However, by the 7th month, they had increased in both types of animal. Although the number of organisms in muscle relative to other tissues was not described in these studies (2, 10, 18), their data support studies indicating that there are very few organisms in muscle of mouse foot pads relative to the numbers found in other tissues (3). This interpretation is also consistent with the work of Pearson et al. (11) who reported that in skin biopsies from leprosy patients 100 to 1,000 times more M. leprae were found in homogenates from epithelial tissue compared with muscle. The data presented in Table 1 suggest that only a fraction of inoculated M. leprae is recovered in the harvest, and that foot pad samples that we routinely used for microscopy contained only about 30% of the inoculated organisms. The fate of the remaining organisms is unknown. To determine what happened to the portion of the inoculum not recovered from mouse foot pads, an intensive study of the fate of inoculated M. leprae has been recently undertaken. ACKNOWLEDGMENTS This work was supported by the U.S. Leprosy Panel ot' the U.S.-Japan Cooperative Medical Science Program administered by the Geographic Medicine Branch, National Institute of Allergy and Infectious Diseases (grants Al and AI-07801), Bethesda, Md. LITERATURE CITED 1. Armstrong, J. A., and P. D'Arcy Hart Response of cultured macrophages to Mycobacterium tuberculosis, with observations on fusion of lysosomes with phagosomes. J. Exp. Med. 134: Esiri, M. M., A. G. M. Weddell, and R. J. W. Rees Infection of murine striated muscle with Mycobacterium leprae: a study by light and electron microscopy. J. Pathol. 106: Evans, M. J., and L. Levy Ultrastructural changes in cells of' the mouse foot pad infected with Mycobacterium leprae. Infect. Immunity 5: Hart, P. D'Arcy, J. A. Armstrong, C. A. Brown, and P. Draper Ultrastructural study of the behavior of macrophages toward parasitic mycobacteria. Infect. Immunity 5: Lefford, M. J The effect of inoculum size on the immune response to BCG infection in mice. Immunology 21: Levy. L Death of Mycobacterium leprae in mice and the additional effect of' dapsone administration. Proc. Soc. Exp. Biol. Med. 135: MacKaness, G. B Cellular immunity. Ann. Inst. Pasteur 120: MacKaness, G. B Resistance to intracellular infection. J. Infect. Dis. 123: McRae, D. H.. and C. C. Shepard, Relationship between the staining quality of Mycobacterium leprae and infectivity for mice. Infect. Immunity 3: Palmer, E. R., R. J. W. Rees, and A. G. M. Weddell Site of' multiplication of human leprosy bacilli inoculated into the foot pads of' mice. Nature (London) 206: Pearson, J. M. H., R. J. W. Rees. and A. G. M. Weddell Mycobacterium leprae in the striated muscle of' patients with leprosy. Leprosy Rev. 41: Rees, R. J. W Enhanced susceptibility of thymectomized and irradiated mice to inf'ection with Mycobacterium leprae. Nature (London) 211: Rees, R. J. W., and A. G. M. Weddell Experimental models for studying leprosy. Ann. N.Y. Acad. Sci. 154: Shepard, C. C The experimental disease that follows the injection of human leprosy bacilli into foot-pads of' mice. J. Exp. Med. 112: Shepard, C. C., and D. H. McRae Mycobacterium leprae in mice: minimal infectious dose, relationship between staining quality and infectivity. and effect of' cortisone. J. Bacteriol. 89: Shepard, C. C.. and C. C. Congdon Increased growth of Mycobacterium leprae in thymectomizedirradiated mice after foot pad inoculation. Int.,J. Leprosy 36: Spector, W. G The granulomatous inflammatory exudate. Int. Rev. Exp. Pathol. 8: Weddell. A. G. M.. E. R. Palmer. and R. J. W. Rees The fate of Mycobacterium leprae in CBA mice. J. Pathol. 104: