Recombinantly purified PPE2 protein of M. tuberculosis (rPPE2) inhibits differentiation of 3T3-L1 cells into adipocytes.

(A) 3T3-L1 cells were treated with different concentrations of rPPE2 and induced to differentiate using DMI cocktail (dexamethasone, insulin and IBMX). At day 10 post-induction of differentiation, the accumulated triglycerides in adipocytes were stained with Oil Red O and image of differentiated adipocytes were taken. (B) For quantitative measurement of lipid accumulation, lipid-bound Oil Red O stain was extracted with isopropanol and absorbance of the extracted solution was measured at 490 nm using isopropanol as blank. (C) Differentiation of rPPE2-treated 3T3-L1 cells in the absence (Medium-Isotype control) or presence of anti-PPE2 Ab. (D) 3T3-L1 cells were differentiated in the absence or presence of 3 µg/ml rPPE2 and changes in the expressions of PPAR-γ, C/EBP-α and FAS genes at various days post-differentiation were measured by qPCR. GAPDH transcript level was used as an internal control. The results shown are mean ± SEM of 3 different experiments. ns = non-significant.

Infection of mice with M. smegmatis expressing PPE2 protein of M. tuberculosis alters adipose tissue physiology.

(A, B) Balb/c mice were infected with 100 million of either Msmeg-pVV or Msmeg-PPE2 bacilli and sacrificed at 2nd and 7th day post-infection (dpi) to measure the weight of perigonadal (A) and visceral (B) fat tissue. (C, D, E, F) Expression levels of PPAR-γ, C/EBP-α and adiponectin in perigonadal tissue and in visceral tissue were quantified by qPCR at day 2 (C and E) and day 7 (D and F) post-infection (dpi). GAPDH transcript level was used as an internal control. (G and H) Histological analyses of perigonadal (G) and visceral (H) adipose tissue sections were performed at day 2 and day 7 post-infection by staining with hematoxylin and eosin (H&E) and photographs of representative sections are presented at 40X magnification (scale bar = 100 µm). Black arrows indicate adipocyte hypertrophy (enlarged adipocyte) and the red arrows point to the infiltrated immune cells into the adipose tissue. (I and J) Adipocyte cell area for perigonadal (I) and visceral (J) fat tissue was measured using ImageJ software. (K and L) Bacterial load in perigonadal (K) and visceral (L) fat tissue was measured at day 2 and day 7 post-infection. For (M and N) For glucose tolerance test at day 7 post-infection, mice were fasted overnight followed by intraperitoneal injection of glucose (2 mg/kg body weight) and blood glucose levels were monitored for 120 minutes (M) and the Area under the curve (AUC) was calculated (N). The results shown are mean ± SEM of 4 mice.

Administration of rPPE2 protein leads to adipose tissue loss and impaired glucose tolerance in mice.

Balb/c mice were intraperitoneally administered with 3 mg/kg of either BSA or recombinant PPE2 protein (rPPE2) on alternate days for 12 days and mice were sacrificed at day 14 post-treatment. (A) Representative photographic images of perigonadal and visceral fat tissues are shown. (B and C) Weight of the total perigonadal fat mass (B) and visceral fat mass (C) are recorded. (D) Transcript levels of PPAR-γ, C/EBP-α and adiponectin genes are quantified in perigonadal fat tissue using qPCR where GAPDH transcript level was used as internal control. (E) Histopathology analyses of perigonadal and visceral fat tissue (representative photographs of hematoxylin and eosin (H&E)-stained sections) is shown at 40X magnification (scale bar = 100 µm). Black arrows indicate adipocyte hypertrophy (enlarged adipocyte) and the red arrows indicate infiltrated immune cells into the adipose tissue. (F and G) Mice were starved overnight and glucose tolerance test was performed for 120 minutes at day 14 post-treatment (F) and the mean Area under the curve (AUC) was calculated (G). The results shown are mean ± SEM of 4 mice.

Pre-immunization with rPPE2 protects the mice from detrimental effects of PPE2 on adipose tissue.

Mice were immunized with 3 doses of recombinant PPE2 protein with Freund’s incomplete adjuvant at 15 days interval. Next, non-immunized and rPPE2-immunized mice were infected with Msmeg-pVV or Msmeg-PPE2 and a change in fat content was observed after day 7. (A) Photographs of perigonadal and visceral fat tissues are shown. Weight of perigonadal (B) and visceral (C) fat tissues of various groups is recorded. (D) Percentage change in total body fat as determined by Dual-energy X-ray Absorptiometry (DEXA) is shown. (E) At day 7, sections of perigonadal and visceral fat tissues were prepared and stained with hematoxylin and eosin and photographs of the representative sections were visualized at 40X magnification (scale bar = 100 µm). (F) A glucose tolerance test was performed on 7th dpi and (G) Area under the curve (AUC) of glucose concentration is shown. The results shown are mean ± SEM of 4 mice.

PPE2 alters adipose tissue physiology during M. tuberculosis infection in mice.

Balb/c mice were infected through aerosol route either with M. tuberculosis clinical strain CDC1551 (wild-type) or CDC1551-ΔPPE2 (ppe2KO) or CDC1551-ΔPPE2::PPE2 (complemented) strains and were sacrificed at day 30 (A and C) and day 60 (B and D) post-infection (dpi). Perigonadal (PG) and visceral (VS) fat tissues were harvested and weight of PG fat tissue (A and B) and VS fat tissue (C and D) was measured. (E) Also, representative photographs of hematoxylin and eosin (H&E)-stained sections of PG at both day 30 and day 60 post-infection were visualized at 40X magnification (scale bar = 100 µM). Black arrow shows adipocyte hypertrophy (enlarged adipocyte) and red arrow represent infiltrated immune cells into the adipose tissue. (F-I) Graphs showing cell area of adipocytes in PG (F and H) and VS (G and I) fat tissues at 30 dpi (F and G) and 60 dpi (H and I) respectively. (J-M) Number of infiltrated CD3+ T-cells were counted using ImageJ software and plotted as cells per field of view for PG (J and L) and VS (K and M) adipose tissue sampled at 30 dpi (J and K) and 60 dpi (L and M) respectively. (N and O) Changes of expression of PPAR-γ and CD36 in perigonadal tissue at day 30 (Ni, Nii) and day 60 (Oi, Oii) post-infection were measured by qPCR and GAPDH transcript level was used as an internal control. (P) Number of bacilli in PG fat tissue at 30 (Pi) and 60 (Pii) dpi respectively. (Q and R) IPGTT performed on mice along with corresponding AUC data obtained at day 30 (Qi, Qii) and day 60 (Ri, Rii) post-infection respectively. (S) Fasting blood glucose levels of mice at 60 dpi and (T) HOMA-IR scores in each mice group were calculated at day 60 post-infection. The results shown are mean ± SEM of 5 mice.

PPE2 modulates the adipose tissue transcriptome during infection.

Perigonadal fat tissues were harvested from mice (n = 3) infected with wild-type, ppe2KO and complemented strains at 60 dpi and subjected to RNA sequencing using Illumina platform. (A) Venn diagram showing all the differentially expressed genes (DEGs) in all subsets (B) Venn diagram showing significant DEGs after applying ±1.3-fold change cutoff. (C) Volcano plot showing significant DEGs in ppe2KO versus wild-type-infected mice in PG adipose tissue. (D) Gene set enrichment analysis (GSEA) for gene ontology analysis (E) KEGG pathway analysis in ppe2KO versus CDC1551 adipose tissues. (F) Heat map of DEGs for chemokine, metabolic pathways, cytokines and TB susceptibility genes in PG adipose tissues. (G) Heat map showing differentially expressed ribosomal biogenesis genes (H) Heat map showing differential expression of phospholipase C activity genes. (I) Perigonadal fat tissue were harvested from mice either left uninfected or infected with various strains of M. tuberculosis at day 60 post-infection and used for cDNA synthesis. Transcript levels of ATF4 and GRP78 as measured by qPCR with β-actin as an internal control. The results shown are mean ± SEM of 4 mice.

PPE2 induces lipolysis in adipocytes by targeting the cAMP-PKA-HSL signaling axis.

(A) 3T3-L1 cells were differentiated into mature adipocytes and then treated with various concentration of rPPE2 and after 24 hours medium was removed from cells and amount of glycerol released was quantified. (B) 3T3-L1 mature adipocytes were incubated with 3 µg/ml rPPE2 or TNF-α cytokine in the absence or presence of TNF-α inhibitor and the free glycerol released was quantified. (C) Mature 3T3-L1 adipocytes were treated with 3 µg/ml of rPPE2 and after 24 hours, cells were harvested for semi-quantitative reverse transcriptase PCR analysis (Ci) and densitometric quantification was carried out using ImageJ software (Cii). (D) In a separate experiment, rPPE2-treated 3T3-L1 mature adipocytes were harvested after 6 hours and 24 hours, and the levels of phosphorylated hormone-sensitive lipase (p-HSL) and total HSL were assessed by Western blotting using anti-p-HSL Ab and anti-HSL Ab respectively (Di). Densitometric quantification of the Western blots was carried out using ImageJ software (Dii). (E) Glycerol release was quantified in rPPE2-treated (3 µg/ml) mature 3T3-L1 adipocytes at 24 hours post-treatment in the absence or presence of H89, a PKA-specific inhibitor. (A-E) All the data represent mean ± SEM of 3 independent experiments. (F) Levels of non-esterified free fatty acid (NEFA) measured in sera of mice either left uninfected or infected with various strains of M. tuberculosis at day 60 post-infection. The results shown are mean ± SEM of 5 mice. (G) NEFA levels in the plasma of healthy controls and TB patients.