3.1 Molecular docking analysis:
Docking studies were performed to investigate the interaction of Indolamine 2,3-dioxygenase (IDO1). Different drugs used for the treatment of metabolic diseases have been used as ligands. These are Metformin, Alpha-tocopherol (Vitamin E) and Pioglitazone as shown in the figure 1A. This docking study revealed that all these three ligands bind successfully with IDO1 with binding energies were greater than –4 kcal mol-1. Table 1-3 suggested the binding kinetics of the amino acids of the ligand IDO1 with the corresponding drugs. Molecular docking study of IDO1 with three drugs were tabulated in Table 1-3. Among all these three drugs, Pioglitazone had the greatest interactions with the target IDO1. These interactions between Pioglitazone and IDO1 were mainly stabilized by hydrogen bond, hydrophobic interactions and pI interactions. The highest binding energy of Pioglitazone with IDO1 was -7.4 kcal mol-1.
Table 1. Prediction of noncovalent interactions for PDB structure of alpha-tocopherol binding with IDO1
|
Hydrophobic interactions
|
Amino acid
|
Position and chain number
|
Bond distance
|
Ligand atom
|
Protein atom
|
LYS
|
238A
|
3.73
|
13
|
2216
|
ALA
|
260A
|
3.61
|
24
|
2428
|
GLN
|
290A
|
3.71
|
11
|
2671
|
PHE
|
291A
|
3.60
|
11
|
2682
|
PHE
|
291A
|
3.67
|
18
|
2688
|
ASP
|
383A
|
4.00
|
4
|
3391
|
LEU
|
384A
|
3.54
|
16
|
3402
|
PHE
|
387A
|
3.64
|
14
|
3429
|
PHE
|
387A
|
3.63
|
16
|
3432
|
PHE
|
387A
|
3.60
|
18
|
3434
|
pi stacking
|
Amino acid
|
Position
|
Chain number
|
PHE
|
291
|
A
|
Table 2. Prediction of noncovalent interactions for PDB structure of Pioglitazone binding with IDO1
|
Hydrophobic bonds
|
Amino acid
|
Residue position and chain number
|
Bond distance
|
Ligand atom
|
Protein atom
|
PHE
|
163B
|
3.78
|
7
|
5042
|
PHE
|
226B
|
3.75
|
7
|
5637
|
ARG
|
231B
|
3.97
|
18
|
5681
|
LEU
|
234B
|
3.61
|
4
|
5722
|
LEU
|
234B
|
3.95
|
18
|
5720
|
PHE
|
291B
|
3.68
|
17
|
6232
|
VAL
|
350B
|
3.53
|
15
|
6817
|
ILE
|
354B
|
3.95
|
8
|
6862
|
PHE
|
387B
|
3.65
|
17
|
7016
|
LEU
|
388B
|
3.99
|
16
|
7026
|
pi stacking
|
Amino acid
|
Position
|
Chain number
|
HIS
|
346
|
B
|
Hydrogen bonding
|
Amino acid
|
Residue position and chain number
|
Distance
H-A
|
Distance
D-A
|
Donor angle
|
ARG
|
231B
|
3.08
|
3.90
|
138.63
|
GLY
|
236B
|
2.63
|
3.59
|
156.11
|
GLY
|
261B
|
2.06
|
2.97
|
147.67
|
SER
|
263B
|
2.60
|
3.13
|
112.56
|
GLY
|
265B
|
2.92
|
3.53
|
119.03
|
Table 3. Prediction of noncovalent interactions for PDB structure of Metformin binding with IDO1
|
Hydrophobic bond
|
Amino acid
|
Residue position and chain number
|
Distance
H-A
|
Distance
D-A
|
Donor angle
|
GLN
|
281C
|
2.29
|
2.80
|
111.02
|
ALA
|
283C
|
3.08
|
4.02
|
154.40
|
THR
|
390C
|
2.54
|
3.42
|
144.38
|
THR
|
394C
|
2.50
|
3.09
|
116.39
|
Fig. 2A illustrates the normal mode analysis (NMA) of Pioglitazone -IDO1complex. Fig. B. The deformability graph of the complex illustrates the peak in the graphs correspond to the regions in the protein with deformability. Fig. C. illustrated the variance map of the Pioglitazone- IDO1 complex. Fig. D. depicts The eigenvalue of the complex. The docked complex generated eigenvalue of 3.68530e-06, the low eigenvalue of the complex shows the ability to deform the structure as this value is directly proportional to the energy of deformability. In Fig. E. the B factor graph of the complex gives easy visualization and understanding of the comparison between the NMA and the PDB field of the complex and it shows the residues that are deformed after binding. Figure F. illustrates the co-variance map of the complexes where the correlated motion between a pair of residues is indicated by red color, uncorrelated motion is indicated by white color and anti-correlated motion is indicated by blue color. The elastic map of the complex shows the connection between the atoms and darker gray regions indicate stiffer regions Fig. G. In vitro experiments.
3.2 Stress reduced body weight gain
Figure 3A illustrated the experimental design. As shown in Figure 3B, 6 weeks of feeding with HFD the body weight of the HFD fed mice significantly increased compared to the normal diet fed mice. After the mice underwent 21days of chronic restraint stress protocol ND+CRS showed delayed body weight gain than the control group while the food intake was decreased. On contrast HFD+CRS showed a different pattern on body weight by decreasing about 10 fold compared to the HFD+CTL despite of altering food intake as shown in Fig. 3C. Whereas decreased food ingestion in ND+CRS mice can be reasonable fact to its decelerated body weight gain. In Fig. 3D. both groups of stressed mice showed increased water drinking. Increased food intake without altering body weight can be indicative of having high energy expenditure by reducing fat mass in both HFD groups. The energy expenditure could attribute into accelerated physical activity while using fat and carbohydrate as fuel to compensate more physical activity.
3.3 Behavioral tests confirmed the successful chronic stress model
We performed a series of behavioral tests for consecutively 3 weeks to assess the depressive-like behavior in mice. As shown Fig. 4A, 4D both of the stressed groups showed significantly decrease in drinking 1% sucrose containing water compared to the distilled water, which implies anhedonia, a typical characteristic for depressive like behavior. Next we performed the tail suspension test in Fig. 4B, 4E the longer latency period of tail suspension test confers the reluctances of the stressed mice to move more often than the control groups. However, in Fig. 4C the group of both HFD mice showed no significant differences in latency time might be due to their overweight. On the other hand, significantly longer immobility time in forced swim test in Fig. 4F suggested the outcome of depressive like behavior in both HFD+CRS and ND+CRS groups. Consistent increase of the immobility time throughout the intensive stress protocol indicates the successful establishment of stress model in experimental mice.
3.4 Alteration of glucose homeostasis in response to stress
Assessments of GTT and ITT suggested improved glucose and insulin sensitivity in both the stressed groups. As expected HFD induced obsessed mice showed impaired glucose tolerance and insulin tolerance than the normal mice according to the Fig. 5A -D. To investigate the glucose utilization process in both the stressed groups we observed the relative m RNA expression levels in glucose metabolizing genes in liver considering the fact that liver is one of the major metabolic organs for glycogen storage which is shown in the Fig. 5E-G . G6pase releases glucose from the liver by glycogen breakdown and provides free glucose into the blood stream during starvation. Downregulated G6pase expression in ND+CRS resonate decelerating Glucose release into the blood from liver hence lowering the glucose level, conversely on set of obesity the stressed mice had higher expression G6pase and reduced expression of GCK which indicates lesser conversion of glucose by glycolysis to promote glycogen conversion in liver. From reducing body weight collectively, we believe that stressed mice are resistant to HFD induced obesity, the resistance had resulted from elevated energy expenditure of utilizing carbohydrate as a form of glycogen.
3.5 Quantification of biological stress markers and lipid profile of serum
Dysregulation of stress hormones and proinflammatory cytokines are the key factors in terms of regulating energy dynamics in depression and anxiety model. IDO-1 is a key enzyme in tryptophan metabolism plays a pivotal role in controlling stressful events. The serum concentration of IDO-1 synergistically increased in HFD+CRS while the increment ND+CRS is 1.8 fold than the ND+CRS group as shown in the Fig. 6A. Collectively, the quantification of serotonin and corticosterone concentrations in serum in both ND+CRS and HFD+CRS groups showed significantly lesser amount of serotonin and higher amount of corticosterone than their respective control groups. Precisely the alteration of serotonin concentration of serum in ND+CRS group decreased by 22.5% than ND+CTL. Contrastingly serotonin concentrations in serum between HFD+CTL and HFD+CRS groups was 8.63% shown in the figure 6B. On the other hand, corticosterone concentration in serum was significantly higher in each groups than the ND+CTL, this result augments the induction of stress by increasing concentration of corticosterone in both the ND+CRS and HFD+CRS group by around 30% and 50% respectively compared to the ND+CTL. The effect of HFD+CTL on corticosterone has increased by 28% compared to the ND+CTL shown in the Fig. 6 C. Moreover, the decreased serotonin concentration and elevated concentration of corticosterone levels in the serum HFD+CTL without the stress exposure is indicative of diet induced stress effect.
In addition, we examined the serum lipid profile .TG, TC and NEFA is shown in the Table 4, which did not change significantly in ND+CTL and ND+CRS groups. Increased TC level in HFD mice is a prominent indication of hyperlipidemia, while TC decreased by 1.74 fold, and increased by 0.9 fold in HFD+CRS compared to the HFD+CTL group.
Table 4. Quantification of lipid profiles in serum
|
Groups
|
Serum TC (mmol/L)
|
Serum TG (mmol/L)
|
Serum NEFA (mmol/L)
|
ND-CTL
|
2.656777
|
± 0.47
|
0.674728
|
± 0.060
|
1.370652
|
± 0.29
|
ND-CRS
|
2.641997
|
± 0.30
|
0.659851
|
± 0.10
|
1.160266
|
± 0.16
|
HFD-CTL
|
6.865074
|
± 0.42
|
0.758623
|
± 0.11
|
1.94529
|
± 0.26
|
HFD-CRS
|
3.998919
|
± 0.67
|
0.726201
|
± 0.13
|
2.077174
|
± 0.15
|
3.6 Alteration of hepatic morphology
Over production of stress hormones can damage the liver physiology while obesity is also associated with a spectrum of liver abnormalities. As expected Fig. 7A H&E staining of the liver sections indicated the ballooning and ORO staining in Fig. 7 B showed the larger lipid droplets accumulation in HFD mice compared to the HFD+CRS group. No substantial changes in hepatic phenotypes appeared in ND+CTL and ND+CRS in either H&E staining or ORO staining.
3.7 Analysis of hepatic gene regulation in lipid metabolism
Next we focused to investigated the hepatic physiology and mRNA expressions of lipid metabolic genes in the Fig. 8A-E. Acaca expression is down regulated in the ND+CRS group than the ND+CTL whereas, in both the HFD groups Acaca was up regulated than the ND groups. Since Acaca takes part in lipogenesis HFD had an effect on its expression level. Srebp1 and FasN are other two lipid metabolic genes which showed similar expression pattern in the ND+CTL group, and significantly decreased expressions were observed in ND+CRS group, which may confer that the stress has a rate limiting effect on their expression level. On the other hand, HFD+CTL showed elevated expression level for both Srebp1 and FasN but the expression is lowered in the HFD+CRS group.
CD36 takes part in the fatty acid transport, which is downregulated in ND+CRS than the control group. In both of the HFD groups the CD36 expression was elevated than the ND+CTL, but comparing to the HFD+CTL, HFD+CRS had significantly lower expression. CPT1a was significantly down regulated in the ND+CRS group than the ND+CTL, but for both HFD groups no significant changes were observed.
3.8 In vivo experiments
3.8.1 Suppression of IDO-1 expression by sh-RNA increased FFA in aml-12 cells
To identify the inhibitory effect of IDO-1 in lipid metabolism in hepatocytes, knock down the IDO-1 was analyzed by shRNA construct plasmid by targeting IDO-1. As shown in the Fig. 9. IDO-1 expression has inhibited by 80% with shIDO-1 transduction in aml12 cells. To establish whether increased IDO-1 influence lipid accumulation in liver, we stimulated FFA and transfected IDO-1 for 48hr and assessed the lipid droplets accumulation and lipid metabolic gene expressions. Sh-IDO-1 transfected cells showed higher count of lipid droplets after ORO staining represented in Fig. 10. The mRNA expression of lipid metabolic genes was upregulated than the control GAPDH gene. SREBP1 showed the highest expression with 1.5-fold increase.