The current set of experiments was performed to determine if the adhesion molecule, α1 integrin, played a significant role in the pulmonary pathology during M. tuberculosis
infection. Our data have shown that α1β1 integrin does indeed contribute to the structure of the granuloma and to the cytokine milieu, particularly during the chronic phase of infection. The absence of the α1-integrin subunit resulted in a relative increase in the number of pulmonary IFN-γ-producing CD4 T cells, but there was a reduction in the concentration of TNF-
and MMP-9 when compared to wild type mice. However there was no significant effect on the bacterial burden in any of the organs examined.
Previous studies by others have shown that the α4-integrin played a significant role in the migration of lymphocytes into the lungs of M. tuberculosis
infected mice, which compromised their ability to reduce the number of CFU 19
. In contrast, our data show that the α1-integrin is not as absolutely required as the α4-integrin. This may reflect the different roles that each integrin plays in the process with the latter required more for leukocytes migration into tissue, and the former for adhesion to the ECM after leukocytes have extravasated into tissue. Others that have also examined adhesion molecules during M. tuberculosis
infection have shown that LFA-1 was required for protective immunity and efficient T cell trafficking to the lung, and LFA-1 knockout mice had prominent neutrophil accumulation associated with early central necrosis 24
. Interestingly, lymphocyte trafficking into the lungs was not impaired in the α1-null mouse strain as shown in ; and while the granulomas were smaller overall, there was little difference in the amount of lung affected between the two groups according to morphometric analysis. Thus, although the α1-null mice had smaller lesions, there may have been a greater number of these lesions, which totaled the same area as the lesions in the WT mice. Taken together the data would suggest a sequential involvement of each integrin whereby leukocytes use the α4-integrin and LFA-1 to migrate into M. tuberculosis
infected lungs and then α1-integrin for maintaining the integrity of the granulomatous lesion possibly through the regulation of MMP-9 activity. Reduced MMP-9 levels in α-1 null mice may have also caused the thickening of alveolar septae that was observed in these mice (), a finding that was similar to MMP-9 KO mice during infection 25
. It is possible that decreased MMP-9 levels in the α1-null mice may have also been associated with increased collagen deposition in , although further studies would need to be performed to determine potential involvement of other MMPs. It has been shown that α1-null mice accumulate more collagen in the dermis than wildtype mice due to a loss of feedback regulation of collagen synthesis 26
, and likewise, α1-null mice in the present study also produced more collagen than the wild type strain. In light of this finding and the more recent finding by Richter et al. 27
that T cell subsets localize differently according to collagen distribution and integrin markers in influenza-infected lung tissue, future experiments in this laboratory will be performed to investigate possible differential expression of α1- and α2-integrins on CD4 and CD8 T cells, respectively, as well as localization of these cell populations according to collagen distribution within the granulomatous environment during infection with M. tuberculosis
. More recently, semaphorin (SEMA) 7A has been shown to interact directly with the α1β1 integrin on T cells and SEMA7A knockout mice are deficient in their T cell mediated immune responses 28
. Given that the α1 integrin interacts with multiple ligands, it will be very interesting to determine if the interaction of SEMA7A with α1 integrin plays a role in the immune response to tuberculosis.
Our data suggest that during infection, α1β1 integrin is required for effective activation of MMP-9 (), which contrasts with previous findings in a tumor vascularization model that showed α1-null mice had increased MMP-9 activity, suggesting a down-regulatory role for this molecule 22
and thus may reflect the diverse functions of this integrin in specific pathologies depending on factors such as the cytokine milieu. Certainly, in the absence of the α1-integrin subunit during M. tuberculosis
infection, there were increased numbers of activated T cells from day 60 of infection producing IFN-γ (), although the concentration of TNF-α was consistently lower during this same period (). Indeed, immunohistochemical analysis showed that the greatest accumulation of TNF-α occurred in the pulmonary lesions of wild type mice (). The greater number of effector T cells in the lungs of α1-null mice may have contributed to increased cellular death due to the presence of activated macrophages, suggesting a TNF-α-independent mechanism. Thus, we presume that the increased numbers of effector T cells in the α1-null mice to be located within the granulomatous regions. Indeed others have shown that CD4 T cells tend to be localized within granulomatous regions 29
. Other studies using influenza virus infection in α1-null mice showed that virus-specific memory CD8 T cells were decreased suggesting that α1β1 integrin is responsible for retaining protective memory CD8 T cells in the lung via attachment to the ECM 30
. The current study would suggest that during infection with an intracellular pathogen that persists in the lung for long periods, α1β1 integrin expression on T cells is not required (). The ability of M. tuberculosis
to persist in the lung, while influenza virus is cleared may be the determining factor as to whether T cells can be retained in the lung via the ECM binding integrin. The fact that there were increased numbers of effector T cells may be due to multiple factors that could be attributed to the continuous presence of antigens made by the organism or that the absence of signaling through the α1β1 integrin may maintain the T cells in an effector state. Increased effector T cells may also be due to the decreased levels of TNF-α in these mice, as other have shown that TNF-α can induce T cell apoptosis and thus less TNF-α would result in fewer T cells undergoing apoptosis 31
. Alternatively others have shown an increase in regulatory T cells associated with lower levels of TNF-α 32
, and this may also be related to the ability of α1-null mice to limit granulomas to smaller well focused lesions. Further studies would be needed to elucidate which of these was true. Interestingly, the presence of increased numbers of effector CD4 T cells did not affect the mycobacterial burden, and may suggest that the number of CD4 T cells in the wild type was sufficient to cause a plateau in CFU and any increase above this level had no effect.
In general the current data suggest that α1β1 integrin may be utilized by T cells during the chronic phase of infection to down-regulate their activity and thus may be involved in conversion to a memory phenotype as suggested by others 30, 33
, since in its absence, T cells were significantly more activated than in wild type mice. In addition, our data suggest that it may play a part in the sequential use of integrins by T cells during their trafficking into and through infected tissue. Indeed, within the environment of an infected lung, it would seem likely that leukocytes use multiple integrins, depending on the stimuli received and incorporated. Taken together, it would seem probable that integrins are utilized by multiple cell phenotypes as they traffic from the vasculature into infectious foci. For tuberculosis it would be of great interest to determine what functions, other than adhesion and trafficking, that these molecules play throughout the infectious process.