A Supramolecular cavitand for selective chromatographic separation of peptides using LC-MS/MS: A Combined in silico and Experimental Approach

The chromatographic separation of proteomic standards via a silica immobilized pillararene cavitand has been designed in silico using host–guest binding energy studies and realized experimentally to selectively interact with peptides.


S2.1 Conformer generation
To generate the different conformers for each of the peptides, we began by drawing a 3D linear structure of the amino acid sequence of the peptide using GaussView 6.0 software. The geometry of this input structure was then optimised and the optimised structure was used to generate the most probable conformers whose energies are within 50 Kcalmol -1 of this optimised structure using openbabel. This was done using the following commandline argument.
Obabel <input> -O <output> --confab -nconf 10 --score energy. Details on how to run and use openbabel can be found here. The gas and solvent geometry optimised energies for all conformers can be found in Table 1.

S2.2 Generation of peptides⸦pillar[5] arene complexes
All the peptides⸦pillar[5]arene complexes were generated via a heuristic approach in order to sample all plausible peptides⸦pillar[5]arene binding sites. This was done by manually changing the position of each peptide conformer inside and around the cavity, while ensuring that all sidechains and terminals of the peptides goes inside the cavity. A minimum of 50 complexes were generated for each conformer depending on the number of plausible interactions that could be sampled. The geometry of these complexes were then optimised in the gas phase using DFTB/mio-1-1 with UFF corrections to dispersion as implemented in ADF version r79006 2019-10-03. Once optimised, 20 lowest energy complexes were then selected and reoptimsed in the solvent using the implicit Generalized Born solvation model with Solvent Accessible Surface Area (GBSA). 2030 surface grid points were used in order to ensure smooth solvent phase geometry optimisation with little S3 numerical noise as possible. These calculations were done at thesame level of theory as in the gas phase. The gas and solvent phase optmised energies of these complexes, which are presented in Table 3 and 4 respectively, were used to compute the peptides⸦pillar[5]arene binding energies from equation (1) in the main text. This was done using the optimised energies for each peptide conformer and the energies of the pillar[5]arene (Table 2) .

S17
Computed distances for all the important interatomic interactions between the peptides and the cavity are presented Table  5. These interactions correspond to interatomic bonds that are less than 3.0 Å. The numbering of each of the atoms can be visualized from any 3D molecular view from their 'xyz' structures provided. The interatomic distances for all the 20 complexes from each peptide conformer can be downloaded from http://doi.org/10.5281/zenodo.3995081

S2.4 Coordinates of all molecules
The coordinates of all the optimized structures presented in the above tables can be downloaded from http://doi.org/10.5281/zenodo.3995081

S2.5 Specific peptides⸦pillar[5]arene binding sites
To further elucidate the specific interaction site between the peptide and the cavity, Table 6 presents all the amino acid units that are interacting with the cavity for each conformer. The strength of these interactions can be inferred from Fig 2b in the main text. The glutamic acid (E) unit is fully inserted into the cavity.
The Isoleucine unit (I) at the head of the peptide is fully inserted into the cavity.
The carboxylate unit found on the terminal argenine unit (R ) is partially inserted into the cavity.

The
Isoleucine unit (I) at the head of the peptide is fully inserted into the cavity.

The
Isoleucine unit (I) at the head of the peptide is fully inserted into the cavity.

SAEGLDASASLR
The terminal leucine (L) unit closest to the argenine (R) unit is fully inserted into the cavity.

VGNEIQYVALR
The valine and the glycine head (V-G) are inserted in the cavity.
Here the Valine (V) at the head is halfway inserted in the cavity.
Glutamic acid (E) fully inserted in the cavity. --

VFTPLEVDVAK
The glutamic acid (E) unit is fully inserted into the cavity.
The glutamic acid (E) unit is fully inserted into the cavity.
The Valine (V) unit, which is closest to the glumatic acid (L-E-V), is fully inserted into the cavity. --

S4.1 Synthesis of co-pillar[4+1]arene bonded silica gel HPLC stationary phase
3 g of YMC Triart HPLC silica grade (particle size 5 µm and pore size is 120Å) was stirred overnight at room temperature in a mixture of THF and TEA (1:1, 50 ml). The solvent mixture was evaporated, and the solid residue dried under room temperature in the fume hood. Later 1.0 gm of co-pillar[4+1]arene was dissolved in dichloromethane (50 mL) and added to the dried basified silica for HPLC stationary phase and stirred overnight at room temperature. The dichloromethane was evaporated and the co-pillar[4+1]arene bonded silica gel stationary phase was washed with DCM to remove unreacted materials and dried for overnight in the fume hood. TGA analysis was carried out to find out % of mass loading of copillar[4+1]arene on the surface of HPLC grade silica gel. Thermogravimetric analysis studies confirmed the mass loading of the co-pillar[4+1]arene at 19.1037 % w/w.   LC-MS/MS analysis of five selected peptides from a peptide calibration mix was used to evaluate the performance of the newly synthesised co-pillar[4+1]arene silica bound HPLC chromatographic stationary phase in comparison with RP-C18 and a bare silica phase. Gradient conditions on bare silica phase are the same as those of mobile phase gradient conditions for co-pillar[4+1]arene bound silica gel stationary phase column as bare silica phase has been used as negative control.  Table S2. Chromatographic data and peaks table on RP-C18 stationary phase.

S6.1 Chromatographic separation resolution data
The peak width at 50% (FWHM) and retention times of 5 peptides were substituted in the resolution equation to calculate the resolution of 5 peptides on silica bound co-pillar[4+1]arene stationary phase and RP-C18 stationary phase. 3 Resolution R =  Table S3. Chromatographic separation resolution of peptide standards on co-pillar[4+1]arene stationary phase and RP-C18 stationary phase.

S6.2 Peak asymmetry data
The peak asymmetry 4 of chromatographic peaks on co-pillar[4+1]arene stationary phase and RP-C18 stationary phase was calculated according to Sciex recommended formula as the experiment was carried out using Analyst® Software provided by Sciex.