It is well documented that Mtb metabolizes cholesterol, though the role of this metabolism in pathogenicity remains unclear. Various Mtb mutants defective in the ability to transport or degrade cholesterol have been previously investigated in respect to possible attenuation of the infection process. The deletion of mce4, which affects cholesterol uptake by Mtb, attenuated the Δmce4 mutant in IFN-γ-stimulated, but not resting, mouse bone marrow MØ
. Disruption of the gene encoding 3-ketosteroid 9α-hydrolase attenuated the growth of ΔkshA and ΔkshB mutants in both resting and IFN-γ-activated, mouse bone marrow MØ
. The inhibition of side chain degradation by inactivation of fadA5 decreased the virulence of mutant during the late stage of mouse infection
. It was also previously shown that Δigr knock-out strain of Mtb was attenuated in mice during the early phase of infection
. The igr of Mtb was identified as required for degradation of the 26-propionate side chain fragment
[21, 22]. The above data suggest that ability of Mtb to catabolize cholesterol is important during both early and late stages of the infection. In contrast, Yang et al.
 reported that replication rates of wild-type Mtb CDC1551 and its mutant ∆hsd were similar in the lungs of guinea pigs and concluded that cholesterol was not an essential source of nutrient for Mtb during infection. On the other hand, Mtb H37Rv ∆hsd mutant (as well as double mutant ∆hsd∆choD) were able to utilize cholesterol suggesting that both HsdD and ChoD are not essential for cholesterol degradation
. All above-mentioned examples described the activity of Mtb mutants in animal models. However, the intracellular replication of mutants defective in the ability to degrade cholesterol and their effects on the functional activity of human MØ are less well understood. Therefore, the aim of our study was to determine whether the ∆kstD mutant can multiply in human MØ and assess its capacity to modify the functional activity of the phagocytes. As we demonstrated previously, KstD is an essential enzyme in the metabolism of cholesterol by Mtb; therefore, the ∆kstD strain is unable to use cholesterol as a primary source of carbon and energy, and accumulates the non-toxic derivatives of cholesterol, AD and 9OHAD. Moreover, the in vitro growth of ∆kstD strain is not affected in rich medium compared to the wild type
Herein, we found that the lack of a functional kstD gene did not influence the ability of resting or IFN-γ-activated MØ to ingest Mtb. However, we observed that the intracellular replication of ∆kstD mutant was attenuated in both resting (statistically significant) and IFN-γ-activated (statistically insignificant) MØ compared to the wild-type strain. The attenuation of cholesterol degradation mutants was previously observed in IFN-γ-activated MØ
[9, 11]. Our data suggest that cholesterol degradation ability is important for Mtb at multiple stages of the infection in resting and IFN-γ-activated MØ. The significant attenuation of the mutant observed in our study in resting MØ may result from experimental model used - human cell line THP-1.
Tubercle bacilli after penetration into MØ reside predominantly in a cholesterol rich region of cell plasma membrane and the ability of Mtb to degrade cholesterol would give tubercle bacilli an advantage within the host. However, it is unclear whether the only reason of attenuation of cholesterol degradation mutants in MØ is due to their inability to use cholesterol as a source of carbon and energy. It was previously found that a mutant lacking an intact hsaC gene accumulated catechol derivatives that appeared to be toxic to Mtb
. The attenuated growth of the ∆kstD mutant in resting MØ, used in the current study, was not due to the accumulation of toxic compounds, suggesting that cholesterol degradation ability per se is essential for the replication of tubercle bacilli inside MØ
. On the other hand, the lack of a functional copy of kstD might modify the basic metabolism affecting pathogenic features of the bacilli. The mutant ΔkshB revealed unusual change in the structure of the cell wall which was thickened and loosened as a result of the synthesis of lipid types other than those in wild-type Mtb
. Such modification of the cell envelope can influence the pathogenicity of Mtb.
It was also suggested that cholesterol metabolism of Mtb may contribute to the production of specific virulence factors and/or disruption of host cell signaling
Moreover, the in vivo cholesterol degradation by Mtb can affect the activity of MØ. In our studies the ∆kstD failed to inhibit ROS and NO production in resting MØ compared to wild-type and complemented strains. It is generally accepted that ROS and RNIs kill or inhibit intracellular growth of Mtb
[8, 25, 26]. Similar to previous report
, we found that Mtb induced ROS production in MØ immediately after phagocytosis (data not shown). The increased oxidative response in MØ infected with ∆kstD unable to metabolize cholesterol can be directly related to cholesterol degradation process (e.g. if cholesterol metabolite modifies the signaling of enzymes involved in NO and ROS production) or can be a derivative of attenuation of bacilli inside MØ. To clarify this issue we used two different Mtb mutants, not related to cholesterol degradation process and showing attenuated growth in THP-1, to test them in respect to inhibition of ROS/NO production in macrophages (data not shown). Only one of them was able to inhibit ROS/NO production to the level of the wild type strain. Therefore the most likely interpretation of our result is that ROS/NO over-production in resting MØ infected with ΔkstD results from the attenuation of the mutant’s growth inside MØ, however the specific role of cholesterol degradation intermediates cannot be excluded. Changes in the cholesterol level in plasma membrane modulate the activity of the proteins and the receptors located in the lipid rafts. The components of NADPH oxidase are known to migrate to the plasma membrane of newly formed phagosome. The recruitment of NADPH oxidase subunits and their assembly in the membrane are necessary for an oxidative burst execution
. It was found that methyl-β-cyclodextrin, a chelating agent of cholesterol, inhibited H2O2 production by neutrophils and in the absence of lipid rafts neutrophil NADPH oxidase activity was changed
We did not observe differences in oxidative response in IFN-γ induced MØ infected with wild type and mutant strains. However, the IFN-γ induces iNOS expression initiating the production of NO by MØ prior to their infection with Mtb (data not shown). The high level of NO reached in IFN-γ treated MØ cannot be subsequently lowered even by wild type Mtb at least within the period of the experiment. Therefore, IFN-γ-activated MØ produced a similar, high amount of NO in response to the infection with wild-type or mutant strains.
Phagocytosis of Mtb initiates the production of both TNF-α and IL-10 by MØ. It has been demonstrated by others that TNF-α together with IFN-γ participate in the killing of Mtb through the induction of NO and ROS production. TNF-α is also essential for granuloma formation
[30–32]. We found here that the infection of resting and INF-γ-activated MØ with wild-type Mtb or ΔkstD mutant caused the release of equal amounts of TNF-α. At the same time however, we observed a greater increase in the production of IL-10 by IFN-γ-activated MØ infected with the ΔkstD strain compared to those infected with the wild-type or complemented strains. It has been reported that pathogenic strains of Mtb stimulate lower levels of TNF-α production by MØ than non-pathogenic species
. IL-10 is an immunosuppressive cytokine that blocks phagosome maturation and antigen presentation and also limits the Th1 response
. Thus, our finding that MØ infected with the ΔkstD strain produce higher level of IL-10 than MØ infected with wild-type Mtb and that similar amount of TNF-α is released by MØ after infection with both strains may suggest that certain aspects of the virulence activity of the wild-type strain are in fact not affected in the ΔkstD mutant.
Interestingly, we found that blocking the TLR2-mediated signaling pathway prior to infection restored the phenotype of the ΔkstD mutant in resting MØ to a level similar to that of the wild-type strain. However, neither anti-TLR2 blocking mAb nor IRAK1/4 inhibitor altered the response of MØ to wild-type Mtb. These results suggest that TLR2 signaling is disrupted in MØ infected with wild-type Mtb, but not in MØ infected with the mutant strain. The essential role of the TLR2-mediated pathway in the production of NO and ROS in Mtb-infected MØ is well documented
[5, 6, 26, 34]. Further study is needed to elucidate the complete mechanism by which Mtb affects TLR2 signaling whether the ability of Mtb to catabolize cholesterol might be important for this process. It has been demonstrated by others that Mtb is able to modulate macrophage signaling pathways by stimulating phosphorylation of the Bcl-2 family member Bad as well as AKT kinase
. AKT kinase serves a pro-survival function by inhibiting apoptotic processes
, whereas Bad is a pro-apoptotic member of Bcl-2 family involved in initiating apoptosis