Study of a new piperidone as an anti-Alzheimer agent: Molecular docking, electronic and intermolecular interaction investigations by DFT method

2.1 General

2.1.5

2.1.5 Theoretical studies

The Gaussian 09 software was utilized to do the theoretical computations (Frisch, et al., 2009), using the 6–311++G(d,p) basis set and the DFT theory with the hybrid functional B3LYP. The molecular visualization software GaussView was used (Dennington et al., 2009). All of the computations have resulted in an optimized geometry with the lowest energy.The optimized geometries exhibited positive vibrational frequencies, showing that they were stable.Vibrational assignments were determined using the VEDA4 software (Jamroz and Analysis, 2004) while accounting for possible energy distribution components (PED) ≥ 10%. The band assignments were confirmed using the GaussView program. Inside this study, NBO simulations were utilized to calculate the energies of intermolecular interactions. In order to investigated the biological activities of our molecule, a molecular docking was carried out. The X-ray crystal structure and structural data for the proteins have been deposited in Protein Databank (rcsb, xxxx) and imported into the modeling program Molecular Operating Environment (MOE) (Environment, 2015). The drawn picture of proteins-ligand is minimized and build in MOE using the MMFF94 force field Scheme 1.

Close

Scheme 1

Synthetic scheme of 3-chloro-r(2),c(6)-bis(4-fluorophenyl)-3-methylpiperidin-4-one [CFMP].

Export to PPT

2.2 Synthesis of 3-chloro-r(2),c(6)-bis(4-fluorophenyl)-3-methylpiperidin-4-one

Arulraj et al. described the synthesis of the CFMP molecule (C₁₈H₁₆ClF₂NO) (Arulraj et al., 2016) 0.3-chloro-2-butanone, p-flurobenzaldehyde, and ammonium acetate are obtained with 99 percent purity from Sigma-Aldrich chemical manufacturer and utilized as such any additional refining.A composition of 0.1 mol, ammonium acetate, 0.2 mol, p-flurobenzaldehyde, and 20.4 ml, 3-chloro-2-butanone in ethanol was gently heated at room temperature. After allowing it to cool to ambient temperature, we obtained the colloidal suspension, which is dissolved in diethyl ether (250 ml). The dissolved solution is mixing with 100 ml concentrated HCl. The solidified hydrochloride of 3-chloro-r(2),c(6)-bis(4-fluorophenyl)-3-methylpiperidin-4-one was extracted by filtering and rinsed with a 1:1 mixture of ethanol and diethyl ether. The yellowish-based solution was extracted from an alcoholic solution was prepared by addition of a quantity of aqueous ammonia equal to 30 ml. The solution was diluted with 200 ml distilled water. The portion of crude sample was dissolved in absolute alcohol (100 ml), warmed till the sample dissolved, and animal charcoal (2.0 g) was added to the resulted solution. The heated solution was filtered using Whatman filter paper to achieve a pure solution. The filtered solutions were allowed to settle down for 48 h. The newly formed white crystals are carefully gathered.

¹³C NMR (CDCl₃, 500 MHz): 21.93 (methyl carbon), 45.25 (C-5 carbon), 60.52 (C-6 carbon), 68.85 (C-2 carbon), 71.58 (C-3 carbon), 137.87–114.85 aromatic carbons, 163.92, 163.44, 161.95, 161.48 aromatic carbons (ipso), 201.97 (C = O); ¹H NMR (500 MHz, CDCl₃): 1.42 (CH₃ proton, s), 2.07 (NH proton, s), 2.53–2.49 (dd, H(5a) proton), 3.45–3.40 (dd, H(5e) proton), 3.94 (H(2) proton, s), 4.06–4.03 (H(6) proton, dd), 7.54–7.03 (Aromatic protons, m). Melting point 136.1 °C.Elemental analysis: C: 64.28 (64.39); H: 4.76 (4.80); N: 4.05 (4.17)

3.1 Conformational and molecular geometric analysis

Piperidin-4-one chair conformers are the most stable conformers. It will have two chair conformations that vary in the NH group's axial (A) and equatorial (E) orientations (Sivakumar and Manimekalai, 2011). The chair conformer's equatorial type of NH is perhaps the more stable in the piperidin-4-one molecule. For the CFMP compound, a comprehensive conformational study was investigated to find the most stable conformers. The computational calculations were performed to evaluate the CFMP molecule, and then DFT method was utilized to optimize the overall geometry of these structures.

Arulraj et al (Arulraj et al., 2016) [CCDC: 1508672] described the single-crystal structure of synthesized 3-chloro-r(2),c(6)-bis(4-fluorophenyl)-3-methylpiperidin-4-one [CFMP]. Distorted chair conformation shows in the piperidine-4-one ring of CFMP molecule, [puckering parameters and φ, θ and Q = 182.7 (19)°, 166.5 (4)°, and0.548 (4)Å, respectively], except for chlorine, all substituents have an equatorial orientation. The experimental analysis shows that the mean planes of two fluoro-substituent attached in the benzene rings have a dihedral angle of 49.3 (3)°. The Molecular geometric structure of CFMP as determined using DFT/B3LYP/6–311++G(d,p) level as shown in Fig. 1a. Table 1 shows the bond lengths, angles of the XRD analysis by experimentally and Optimised DFT of CFMP structure. The theoretically predicted structural parameters are much nearer to the values obtained from XRD observations.

Close

Fig. 1

a & b: a) Molecular geometric structure of CFMP obtained by DFT/B3LYP/6–311++G(d,p) level. b) The molecular electrostatic potential (MEP) of CFMP.

Export to PPT

3.2 Molecular electrostatic potential investigation

The MEP is a useful characteristic for determining a molecule's reactivity (Romani et al., 2020). It allows for the identification of nucleophilic and electrophilic sites that enhance hydrogen bond formation. The nucleophilic site implies a strong attraction, whereas the electrophilic site indicates a strong repulsion. (Scrocco and Tomasi, 1978). The Molecular electrostatic potential of 3-chloro-r(2),c(6)-bis(4-fluorophenyl)-3-methylpiperidin-4-one (CFMP) and the 3D graphical surface have been developed and are shown in Fig. 1b. The MEP at the B3LYP/6–311++G(d,p) approach was computed to predict reactive sites for nucleophilic and electrophilic interactions on CFMP. In the sequence, the electrical potential diminishes: blue < yellow < orange < red. We can distinguish the electrophilic and nucleophilic sites of a molecule using these different colors. Further investigation indicates that the piperidin-4-one ring with in chemical structure has a red (negative charge) area that is sensitive to electrophilic attack.The color codes for these surfaces vary from −4.167e-2 (red color) to 4.167e-2 (blue color) (blue color). The extremely positive area is represented by the blue color zone observed surrounding the nitrogen atom. These findings might help to understand hydrogen bonding inside the molecule.

3.3 Analysis of natural bond order

NBO is the efficient approach for analyzing and identifying the particular bonds and energy connected with electrons of a single pair electrons, which are incredibly meaningful in physicochemical processes (Weinhold, 2001). This technique resulted in electronic exchanges, hydrogen bonding among donor–acceptor molecules, and interactions of hyperconjugative. The energy difference between interacting orbitals might be used to calculate the interaction energy among acceptor and donor electrons, as well as the overall stability of orbital interactions. Calculating the second-order stabilization energy E⁽²⁾ can help you figure out which interaction is the most stable. $${\mathrm{E}}^{(2)}=\mathrm{\Delta }{E}_{\mathrm{i}\mathrm{j}}=q_i\frac{{F(\mathrm{i},\mathrm{j})}^2}{{\epsilon }_j-{\epsilon }_i}$$

Where ε_i and ε_j represent the diagonal elements of orbital energies, q_i represents the occupancy of the ε_i orbital, and F(i,j) represents the Fock matrix member outside diagonal. The intensity of the interaction between both the donor and the electron acceptor seems directly proportional to the size of E(2) energy; the greater the magnitude of E(2), the stronger the interaction between both the donor and the electron acceptor. The CFMP computed findings are tabulated in Table 2. In this work, the stabilizing energy E(2) is related to hyper conjugative interactions, LP(2) → π(C₁₆−C₂₄) → π*(C₁₇−C₁₉) and π(C₁₆−C₂₄) → π*(C₂₁−C₂₂), is determined to be 21.34 and 19.26 kJ/mol, respectively. Anti-bonding orbitals of its acceptor π*(C26−C27) and π*(C32–C34), the bond π(C29–C31) stabilizes the energy of 21.19 and 18.11 kJ/mol, respectively. The intramolecular hyperconjugative contacts are caused by the overlap of the π(C−C) and π*(C−C) orbitals, resulting in intramolecular charge transfer within the molecular system (Arulraj, 2022). According to our calculations, the E(2) energy between π(C₂₉–C₃₁) and π*(C32–C34) is about 18.11 kJ/mol. Likewise, the π–π* interaction of C26−C27 → C32−C34, C26−C27 → C29−C31, C32−C34 → C26−C27, and C32−C34 → C26−C27 and C32−C34 → C29−C31 bonds exhibited lower electron densities than phenyl rings with the highest hyperconjugative interaction energy E(2). The NBO characteristic of the CFMP molecule was caused by the polarized mobility of the π–electron cloud inside the molecule and internal charge transfer.

3.4 Non-linear optical properties

The study of how a material's optical properties change when it is subjected to high-intensity radiation is known as nonlinear optics (NLO). These characteristics properties are applied in a wide range of areas, including medical, data processing and storage, sensors, and so on.The values of hyperpolarizability and polarizability are important for determining optical connections and interactions. (Prasad and Williams, 1991). Table S1 shows the optical characteristics of several materials, including the hyperpolarizability (β), the polarizability (α) and the dipole moment (μ). These parameters (β), (α) and (μ) are calculated using the formulae below. $$\mathrm{d}\mathrm{i}\mathrm{p}\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{m}\mathrm{o}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}\mu =\sqrt{{\mu }_x^2+{\mu }_y^2+{\mu }_z^2}$$ $$\mathrm{p}\mathrm{o}\mathrm{l}\mathrm{a}\mathrm{r}\mathrm{i}\mathrm{z}\mathrm{a}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}\alpha =\frac{{\alpha }_{\mathit{xx}}+{\alpha }_{\mathit{yy}}+{\alpha }_{\mathit{zz}}}{3}$$ $$\mathrm{\Delta }\alpha =2^{-1/2}{[{\left({\alpha }_{\mathit{xx}}-{\alpha }_{\mathit{yy}}\right)}^2+{\left({\alpha }_{\mathit{yy}}-{\alpha }_{\mathit{xx}}\right)}^2+{\left({\alpha }_{\mathit{zz}}-{\alpha }_{\mathit{xx}}\right)}^{2}6{\alpha }_{\mathit{xx}}^2)]}^{-1/2}$$

Hyper polarizability $\beta =\sqrt{{\beta }_x^2+{\beta }_y^2+{\beta }_z^2}$ $${\beta }_x={\beta }_{\mathit{xxx}}+{\beta }_{\mathit{xyy}}+{\beta }_{\mathit{xzz}};{\beta }_y={\beta }_{\mathit{yyy}}+{\beta }_{\mathit{xxy}}+{\beta }_{\mathit{yzz}};{\beta }_z={\beta }_{\mathit{zzz}}+{\beta }_{\mathit{xxz}}+{\beta }_{\mathit{yyz}}$$

The hyperpolarizability (β) and polarizability (α) values are in atomic units (a.u.), the computed values were converted to electrostatic units (esu) (1 a.u. = 0.148 × 10^–24 esu and: 1 a.u. = 8.6393 × 10^–33 esu). μ = 0.900 Debye, α = 3.388 × 10^-23 esu, and β = 3.416 × 10^-30 esu are the computed values for CFMP compound. Urea (β = 3.72810^-33esu) is a sample molecule used in research on the NLO characteristic of organic chemicals that would be optimized and calculated using the DFT/B3LYP/6–31++G(d,p) technique. That's why, urea was frequently used as a prototype quantity for comparative purposes. Since the molecule has a greater hyperpolarizability (β) value than urea, it could be used in the scientific investigation of nonlinear optical characteristics.

3.5 Frontier molecular orbital analysis

Gaussian 09 software and the TD-DFT/B3LYP technique with basis set 6–311++G(d,p) were used to compute the detected excited states of a synthesized compound. Table 3 displays the absorption Oscillator force (f), wavelength (λ), and excitation energies, as well as the primary contribution of transitions. Using this method, we identified three absorption peaks on analysed UV–Visible spectra. In our case, of the CFMP compound, the experimental spectrum was obtained with 100 percent ethanol at 293 K (Fig. S1).The examination of this spectrum show that the strong absorption peak was observed at 306 nm. The electronic transition from the ground to excited states was associated to the HOMO → LUMO (94%) excitation at 294 nm (33982 with f = 0.0061). Moreover, the HOMO + 5 → LUMO (32%) and HOMO + 5 → LUMO (19%) and oscillator strengths, wavenumbers, transitions energies are 0.0340 and 0.0240, 306 and 258 nm, 32604.156 and 38413.768 eV respectively. The structure of our compound allows for a robust π–π* transition with a significant extinction coefficient throughout the UV–Visible range. The experimental findings match well with the transition excitation energies.

The frontier molecular orbitals affect the electric and optical properties, as well as the chemical processes. The energy gap for CFMP was calculated using the DFT/B3LYP technique and the 6–311++G(d,p) level. The energy gap and HOMO-LUMO eigenvalues regulate the molecule's bioactivity and chemical environment. As an electron acceptor, the LUMO indicates the capacity to obtain an electron, and the LUMO as a donor shows the tendency to donate an electron. Fig.S2 depicts the HOMO–LUMO orbital of the studied CFMP compound. The calculated energy gap is around 5.11 eV. Table 4 shows the CFMP electronic parameters like global electrophilicity index (w), global softness (S), global hardness (η), chemical potential (μ), electronegativity, energy gap (ΔE_HOMO-LUMO), and frontier molecular orbitals (E_HOMO, E_LUMO) were computed using the DFT method. Those calculated E_HOMO and E_LUMO using the equations shown below.

HOMO energy (E_HOMO) = -6.83 eV

LUMO energy (E_LUMO) = -1.72 eV

Energy gap (ΔE_HOMO-LUMO) = 5.11 eV

The molecule's stability improves as the band gap energy decreases. The molecular charge distribution was estimated using the DFT method. This computation takes into account the charges of each atom in a molecule. According to the diagram, HOMO is placed over the phenyl piperidine ring and has a delocalized electron charge density, whereas LUMO, on the other hand, is located across the piperidine ring and has an electron density that is centred at the ring surface. The negative and positive phases are denoted by the colors red and green, respectively. The molecule's overall softness and hardness, which is a good sign of determines its chemical stability. The value of the energy gap was used to assess if a molecule is hard or soft. The CFMP compound hardness value (η) is 2.56 eV. The CFMP compound is classed as a hard material as a result of this finding, and the electrophilicity index (w) defines the molecule's bioactivity.

3.6 Mulliken atomic charges

The electronegativity normalization and charge transfer mechanisms in the reactions were detected using the atomic charge. Mulliken atomic charges were calculated to analyse the increase in electropositivity owing to its electronegative neighbor and vice versa in brief charge transit in a molecule. Charge transfer and atomic charges are fundamental to molecule activity and chemistry (Mulliken, 1955). The Mulliken population analysis, total atomic charges of CFMP computed using DFT-B3LYP level are shown in Table 5. The MA charges for CFMP are shown in Fig. S3. The negative values −0.552, −0.517, −0.632, −0.266, and −0.277 on the aromatic phenyl rings' C₇, C₁₀, C₁₁, C₁₇, and C₂₇ atoms cause electron density redistribution. Because of their major negative charges, the atoms C₉, C₁₆, and C₂₆ can accept higher positive charges of 0.875, 0.779, and 1.102, respectively, and thus become more acidic. It's worth mentioning that its O₄ atom in the CFMP molecule seems to have a negative atomic charge of −0.216.

3.7 Vibrational IR spectra and assignments

The FT-IR vibrational wavenumbers were calculated using the DFT/B3LYP/6–31++G(d,p) method. To evaluate the vibrational properties of CFMP, we perform vibrational research utilizing infrared spectroscopy at ambient temperature. The discrepancies between estimated and experimental wavenumbers were addressed by adding a scaled field or simply scaling the wavenumbers computed with appropriate factor (Scott and Random, 1996). To adequately conform with experimental results, the normal mode vibration frequencies were calculated at the B3LYP level. Such that, above 1700 cm⁻¹ wavenumbers are scaled by 0.958, whereas below 1700 cm^-1 wavenumbers are scaled by 0.983.

Table S2 displays the scaled wavenumbers that were calculated and experimentally observed. Fig. S4 depicts the calculated FT-IR spectrum and experimental FT-IR spectrum. This table shows the computed complete PED assignments, infrared intensity, and fundamental vibrational frequencies in both unscaled and scaled forms. In our investigation, the calculated wavenumbers range from 4000 to 400 cm⁻¹. The vibrational assignment was completed with the assistance of a scaled quantum mechanical software and a total energy distribution calculation technique. To get theoretical values nearer to calculated values, we used a scaling factor of 0.9608 (Palafox, 2000).

3.8 Hirshfeld surface analysis

Hirshfeld surface (HS) analysis aims to investigate intermolecular interactions, determine inter-nuclear distances, bond angles, crystal packing, and close contact atoms. Fingerprint plots, Hirshfeld surfaces measuring the space occupied by a molecule, and intermolecular interaction were all created with the Crystal Explorer 3.1 tool, which was supported by crystallographic data files (CIF) (Agarwal et al., 1245). Hirshfeld surface analysis gives a great graphical representation of the interactions of hydrogen bonding. HS analysis visually investigates intermolecular interactions in crystal structures by using 2D fingerprint plots and molecular surface contours. The contoured form of a molecule inside a crystal structure produces a van der Waals (vdW) surface. The various vdW radii of atoms define the distances between locations on the surface and nuclei (atom) outside (d_e) and within (d_i) the mean surface, whereby the contact distances di and de may be d_norm(standardized) (Turner et al., 2017; Sagaama and Issaoui, 2020).The van der Waals radii of the external ( ${\mathrm{r}}_{\mathrm{e}}^{\mathrm{v}\mathrm{d}\mathrm{w}}$ ) and internal ( ${\mathrm{r}}_{\mathrm{i}}^{\mathrm{v}\mathrm{d}\mathrm{w}}$ ) and atoms on the surface, van der Waals radii and normalized contact distance d_norm, is represented as (Ben Issa et al., 2020). $${\mathrm{d}}_{\mathrm{n}\mathrm{o}\mathrm{r}\mathrm{m}}=\frac{{\mathrm{r}}_{\mathrm{i}}-{\mathrm{r}}_{\mathrm{i}}^{\mathrm{v}\mathrm{d}\mathrm{w}}}{{\mathrm{r}}_{\mathrm{i}}^{\mathrm{v}\mathrm{d}\mathrm{w}}}+\frac{{\mathrm{r}}_{\mathrm{i}}-{\mathrm{r}}_{\mathrm{e}}^{\mathrm{v}\mathrm{d}\mathrm{w}}}{{\mathrm{r}}_{\mathrm{e}}^{\mathrm{v}\mathrm{d}\mathrm{w}}}$$

Fig. 2a displays Hirshfeld analysis plotted with the d_norm and two neighbouring molecules outside the surface for the complex. In this work, the Hirshfeld surfaces for our compound d_norm, d_e, d_i, curvedness, and shape index are showed in Fig. 2b.

Close

Fig. 2

a & b: a) Hirshfeld analysis mapped with d_norm and showing two adjacent molecules outside the surface for the CFMP. The red colors indicate contacts with distances shorter than the van der Waals radii, and indicate weak inter/intramolecular interactions. b) Hirshfeld surface for CFMP with d_i, d_e, d_norm, shape index, and curvedness.

Export to PPT

Intermolecular distances were depicted as a white color near $({\mathrm{r}}^{\mathrm{v}\mathrm{d}\mathrm{w}}$ ) connections with a d_norm of zero. Blue surfaces are contacts with high d_norm values that are bigger than the sum of $({\mathrm{r}}^{\mathrm{v}\mathrm{d}\mathrm{w}}$ ).

Two-dimensional fingerprint plots of the Hirshfeld surface contact that contributes 100% of CFMP (Fig. S5). The d_e versus d_i plot revealed the existence and amount of several types of intermolecular interactions. The d_e and d_i are also used to generate the 2D fingerprint plot and a summary of intermolecular interactions in the CFMP crystal compound. 2D fingerprint plots illustrating main intermolecular contacts and percentages of different intermolecular interactions. It leads to the Hirshfeld surfaces in the CFMP compound (Fig. S6). Total intermolecular interactions in CFMP, including reciprocal connections: The relative percentage of (a) = H∙∙∙H contacts (33.9%) (a1), (b) = H∙∙∙F contacts (22.2%,) (b1), (c) = C∙∙∙H contacts (19.2%) (c1), (d) = H∙∙∙Cl contacts (11.1%) (d1), (e) = O∙∙∙H contacts (9.0%) (e1), (f) = F∙∙∙F contacts (1.1%), (g) = C∙∙∙F contacts (0.9%), (h) = C∙∙∙O contacts (0.6%), (i) = C∙∙∙Cl contacts (0.6%) and (j) = O∙∙∙O contacts (0.1%). A1 through E1 indicate decomposed views of the weak intermolecular interactions in each of the first five (a1 – e1) 2-D represented views of the molecule. Fig. S7 depicts the significant percentage additions to the Hirshfeld surface area of the CFMP compound's numerous near intermolecular interactions.

3.9 Molecular docking study

Molecular docking is a powerful method used to show the binding activities of drug molecules with protein. The main purpose of this tool has helped us in designing a new inhibitor and drugs in the treatment of Alzheimer illness. These diseases are a degenerative disorder of the brain which affect a large number of people in the world. The exact cause of this disease is still unknown, but it is believed that environmental and genetic factors contribute to it. Therefore, there are some drugs that may delay its progress. In this context, a bibliographic study shows that these proteins VEGFR2 (code :6lvm), hAChE (code: 4ey7), DRD2 (code: 6CM4) and DHFR (code:1DRF) are important targets for developing a new drug against Alzheimer disease (Youssef et al., 2017; Dawood and Dayl, 2020). Molecular docking simulations were performed using MOE.10 software to get detailed information about ligand-receptor binding energy. The CFMP compound was chosen as the ligand and docking studies were done taking 6lvm, 4ey7, 6CM4 and 1DRF as the proteins. For this reason, the native ligand CFMP was re-docked in the binding sites of these enzymes. The 2D and 3D pictorial presentation of the compounds helped us to find the active sites (Fig. 3). This figure shown the best conformations which corresponds to the lowest energy of each complex ligand-proteins. Also, the calculation results such as the binding energy score (S), RMSD, E-config and E-place are summarized in Table 6. The coordinates of interaction parameters of ligand with proteins are tabulated in Table 7. Analysis of Table 6 reveals that all the proteins presented good docking score. In addition, the root mean square deviation (RMSD) values of these compounds range from 0.84 to 1.53 A°. The binding score values for 6LVM, 4ey7, 6CM4 and 1DRF are −5.59, −7.19, −5.93 and −6.72 kcal/mol, respectively. However, the better binding energy score and RMSD value has been obtained from the protein 4EY7 (S = -7.19 kcal/mol, RMSD = 0.84 A°). In addition, The RMSD values facilitates the stability of the system if it is lower than 3 (Ahmad et al., 2017). Furthermore, the RMSD values allow to confirm the reliability of our docking results (Arshad et al., 2020). Also, it is found from the Table 7 that the key contributing amino acid in CFMP with the target complexes are ALA 9, ASP 114, HIS 393, ASP 94, ASP 635, LYS 508 and LEU 478. These interactions allow to stabilize the molecules on the active sites. We note also that the measured distances for both complexes studied vary between 2.62 and 4.36 Å. The 6lvm shows two conventional H-bonds of LEU 478 and LYS 508 at a distance of 4.36 A° (energy value of 0.6 kcal/mol) and 2.86 A° (energy value of 4.0 kcal/mol), respectively. Similarly, the docked conformation 1DRF and 4ey7has two conventional H-bond (3.01 A°/ ALA 9 and 2.62 A°/ ASP 114) and (3.46 A°/ASP 74 and 3.51 A°/ASP 74), respectively. Thus, the docking results show that CFMP has a lot biological activities. Finally, we can conclude that the different interactions observed between the complexes makes it clear that our compound can be a potential inhibitor against Alzheimer's disease.

Close

Fig. 3

3D (left) and 2D (right) molecular docking of CFMP ligandwith 4e7y, 1DRF, 6CM4, and 6LVM proteins.

Export to PPT