Team:IISER-Pune-India/Engineering

Overview

Illustrated below is the pipeline of experiments that we plan to perform in the lab to prepare the cyclotide drugs and verify their efficacy.

Since we had limited lab access for just a three week period due to the pandemic, we were unable to experimentally validate our system or pursue significant work in the lab. However, keeping in mind the engineering design cycle, we have thoroughly planned and designed all our experiments. We have considered the possible problems we might face through the course of the project and have come up with alternatives (in case of unexpected results) and troubleshooting protocols if faced with any failures in our experiments. To evaluate the outcomes and verify them, we have also made measurement standardization protocols for our Assays.


Research

Our research began with looking up statistics on the prevalence of Malaria and drug resistance of the Malarial parasite. Over the past five decades, Plasmodium falciparum, the Plasmodium parasite which causes the deadliest form of Malaria in humans, has gained resistance against commonly used drugs like chloroquine, sulfadoxine, quinine and mefloquine.[1]

As of today, Artemisinin is the most potent anti-malarial drug. There is, however, growing evidence of Artemisinin resistance worldwide. Our last line of defence is Artemisinin Combination Therapy (ACT), and it’s failing fast, potentially making malaria untreatable in the years to come.[1]

We were aware that the Plasmodium parasite spends a part of its life cycle in RBCs. We thought of blocking potential interactions between RBCs and parasites so as to inhibit them from infecting RBCs. Targeting multiple interactions by creating multiple inhibitors could possibly overcome resistance as parasites begin to develop resistance to a particular drug.

The next question was "What system could we use as a delivery vector for these inhibitors?"

Team Heidelberg 2014 had worked on the synthesis of cyclotides, which have previously been used as potential therapeutics for HIV and other diseases.[2] Upon further research, we found that cyclotides could be grafted with a peptide of choice to give them more stability.[3]

That’s how we came up with the idea for a cyclotide based drug. Cyclotides act as a scaffold or backbone for inhibitory peptides that target crucial host-pathogen interactions. They are robust, cost-effective, efficient, orally bioavailable and resistant to degradation, making them optimal candidates for drug design.[3]


Imagine

To begin, we decided to study crucial interactions between the parasite and human RBCs proteins that mediated entry and infection, and accordingly design peptide-based inhibitors for said interactions.

One problem we faced was the fact that there are four species of Plasmodium that infect humans via mosquitoes.[5] After our interaction with Dr. Velavan (which can be found on our iHP page here), we picked Plasmodium falciparum to work with, as it causes the deadliest form of Malaria.[5]

The basic idea was to design a short peptide which would act as a competitive inhibitor for a crucial interaction between the parasite and the human RBC proteins. Being small, the inhibitory peptides would cause minimal immunological reactions. This peptide would then be grafted into a cyclotide backbone, which would be used for delivery.

Cyclotides make an optimal drug delivery system as they are more resistant to heat than linear peptides and hence are easier to transport and require minimal cold storage facilities.[6] This stability at elevated temperatures comes in handy particularly in tropical countries that are the most affected, and it allows for easy and safe storage in rural areas where there is insufficient availability of cold storage facilities. Using cyclotides may also allow us to administer the drug orally.

In order to gauge the levels of public awareness about malaria, we decided to conduct a survey, the results of which were used to effectively design our human practices and public engagement strategy. In order to determine the problems faced by public health workers in the treatment and diagnosis of the disease and find ways to ameliorate them, we also contacted doctors, scientists and public health workers to learn more about the challenges faced by them in successfully eradicating Malaria.



We realised that to design a suitable inhibitor, we would require a computational approach, in order to shortlist possible interactions we could target. We contacted Dr Madhusudhan MS of IISER Pune and he generously offered to help us, and along with Tejashree Kanitkar, a PhD student from his lab, we designed the basic layout of our dry lab and modelling experiments.

Our modelling approach has the following steps:

  1. Extraction of available data from the Protein Data Bank[7] and further analysis and shortlisting of potential PDB IDs that could be used in our project
  2. Identification of the interacting regions of the human protein and the parasite protein.
  3. Computational mutagenesis of the interacting region
  4. Scoring the mutants and doing some advanced studies for stability in silico

We started with extracting data from the Protein Data Bank using advanced search and a custom web-parser script based on python and found over 400 hits which we narrowed down to 47 interactions.[8] We did literature studies and shortlisted 5 interactions to work with. We performed computational saturation mutagenesis for the identified interacting region using a custom python script built on top of Chimera and obtained mutants.[8] The mutants were scored based on their interacting energy using FoldX.[9] We shortlisted 3 mutant interactions with most negative interaction energy for further studies.

We ran Molecular Dynamics simulations for these inhibitors to study their interaction in silico. We got access to the PARAM BRAHMA facility under the National Supercomputing Mission, Government of India at IISER Pune for our computational needs. We performed the simulations as described in the Modelling page of our project and analysed the results.

We found that some inhibitors did not interact with the parasite protein as expected. One reason may have been the small size which leads to structural instabilities. From the iBEC[10] proposal feedback we also learnt that the PfEMP1 interaction which we had targeted so far is highly variable. We also faced problems in doing homology modelling of cyclic peptides.

From our results, we realised that we needed to search for more non-variable interactions of the plasmodium and the human proteins. We analysed how the distance between the centroids of the interacting proteins evolved over time during the MD simulations and the intermolecular hydrogen bonds between them. As for the homology modelling for cyclic peptides, we contacted Dr David Norman, College of Life Sciences, University of Dundee, for help and he generously provided us with tutorials to build a cyclic peptide using homology MODELLER.[11]

We conceptualized cyclotides to contain inhibitory peptides in a suitable loop to competitively inhibit the host-parasite interactions of interest. After extensive literature review, we finalised on the cyclotide Kalata B1 which is 29 amino acids long with 6 loops divided by 3 cysteines as a knot. Based on in-silico modelling, we decided on the insert sequence for potential competitive inhibition (in loop 6) to graft in pET28-a(+) vector.

We also planned to insert host and parasite proteins into the same vector to express and study their interaction in vitro and the effect of the cyclotide inhibitor on their binding affinity. For tagging the parasite and host proteins we planned to use polyhistidine and Strep tags respectively. This differential tagging is to prevent the host protein from non-specifically interacting with the stationary phase (Ni-NTA) and yielding a false positive interaction with 6xHis-tagged parasite protein immobilized on the stationary phase.

We obtained the cyclotide Kalata B1 precursor (OAK1) and competitive inhibitor sequences from the modelling team. We built 1-3 composite biobricks depending on the potential inhibitors. Codon-optimised inhibitors were grafted into loop 6 of OAK1 and loop 3 was replaced with the Strep tag for purification. We also obtained wild type protein sequences from PDB to build basic Biobrick parts.

Building the inhibitor construct:

The sequence of nucleotides from the start to the stop codon was removed and the cyclotide construct containing the inhibitory peptide was inserted into the plasmid. A similar procedure was followed for the host and parasite protein.

Our Biobricks were built on the Benchling platform. Further analysis was done on various softwares including Geneious Prime, Snapgene and Benchling to check for RFC 10 criteria and this was cross checked with our mentors. We tested our Biobricks for the ORFs. The initial codon optimisation was done using the GeneSmart tool of genscript and was tested again with codon optimisation tool in Benchling for better results.

While building Biobricks we encountered various problems. We learned that Biobricks for the split Intein approach needed to be tested for at least 3 loops. Even though we grafted our inhibitors in loop 6, the problem of circularisation still persisted. Another issue that we came across after building the ORF was breaking at both ends of the insert sequence. This problem occurred due to the 7th general nucleotide(N) present at the end of the BSA restriction site i.e (GGTCTCN).

To solve the circularisation issue we took the circularisation construct (BBa_K1362000) and removed the RFP gene from it. We then grafted the Cyclotide KB 1 sequence into the construct which was subsequently grafted into pET 28-a(+). We made sure that ORFs were on a positive strand. The seventh nucleotide was removed from both strands to complete the ORF.

Designing the experiments was the complete second module of our project. During the synthesis of peptides in the lab we aimed to perform a wide variety of experiments from cell cultures, PCRs, gel electrophoresis, mass spectroscopy, and microscopy to accomplish cloning of the desired sequences, expression and purification of proteins. It would also involve a few experiments to validate the results such as SDS-PAGE.

For cloning we would use Gibson assembly techniques. Once cloning is successful, the next step is the expression of the cloned plasmid into the vector. Protein purification would be done by expressing Strep tag for the Host protein and 6Xhis tag for parasite proteins. The purified protein would then be run over SDS-PAGE to check for its purity.

The required plasmid was obtained from the lab and the gene of interest was ordered from IDT with the primers designed using Gibson. Once all the sequences are received, follow the Gibson assembly steps detailed in the Experiments Section.

Protein gene overexpression would be done in E.coli BL21(DE3) strain using the T7 promoter system. The module 2 of experiments involves Purification by 6xHis tag and Strep tag. 6xHis tag purification is done by preparation of lysates after cytoplasmic expression of His-tag protein under native conditions followed by batch purification under native conditions. Purification of strep tag proteins is done by gravity flow columns.

Experiments Page

Cloning would be tested by Single restriction digestion based on length and number of fragments in reaction and PCR amplification using gibson primers and sequencing of the cloned plasmids. The positive control should yield ampicillin resistant colonies when transformed into E.Coli. Transformed E.coli BL21 DE3 culture for incubation and purified protein would be run over SDS-PAGE analysis.

Experiments Page

During planning of the experiment design we also came up with the problems we might face while performing the experiments. The protein might not get expressed, it might be degraded or there could be insoluble protein in the column.

General difficulties one might encounter in purification are no, or weak binding of proteins to the strep tag column, contamination of proteins, protein precipitation during purification. We have also listed possible solutions for the above problems.

If we face any problems in Gibson cloning we have considered homologous cloning as an alternative which is very similar to Gibson. For the detailed troubleshooting of the individual sections refer to the respective experiments section on the project page.

Experiments Page

After designing the drug, another essential step in our project would be to test its efficiency in combating the disease. Assay building is a molecular biology procedure for testing or measuring a drug's activity in an organic sample. Various molecular biology experiments and assays can be performed to test, verify, and incorporate the computational predictions. The goal is to select a method that requires the least amount of manipulation and pretreatment of the samples to accommodate substances that intervene with the assay.

In our project, the main focus was to develop experiments for testing the interaction between inhibitory drugs and the binding epitomes of the plasmodium proteins.

The following assays were employed for testing competitive binding of the designer cyclotide:

  1. Solid-phase protein interaction assay

    In vitro by performing affinity chromatography on a Ni-NTA column with the parasite protein immobilized to it and flowing the host protein through it with and without the cyclotides. The difference between the amount of protein left in the supernatant as compared to what was added indicates the bound fraction. This method assesses specific protein-protein interactions at physiological conditions, utilizes relatively small amounts of protein, is free of protein modification, does not require specialized instrumentation and results in quantitative measurements.

  2. In vivo in Plasmodium falciparum parasite cultures
    • Half maximal inhibitory concentration (IC50) assay is commonly used to measure the potency of a drug in pharmacological research. IC50 is a measure of the potency of a substance in inhibiting any specific biological or biochemical function. This assay is fast, relatively inexpensive with no involvement of radioactive isotopes. Due to high affinity towards DNA ,the assay is very sensitive and IC50 values of known drugs obtained were comparable.
    • RBC invasion assay is a flow cytometry-based assay for measuring invasion of RBC by P. falciparum merozoites. Although comparatively expensive, the key advantage of flow cytometry is that a very large number of particles can be evaluated in a very short time.
Experiments Page

Assays Standardization

Standardization aims to ensure that assays of the same analyte in the same samples, done at different places or at different times or both, are readily comparable. Standardization is especially desirable because all external quality assessment surveys have shown that this type of method involves a much greater variability than traditional assays. In our project, we have three assays. The RBC invasion assay is a flow cytometry-based assay for which the general standardization protocol is available in IGEM’s standard measurement protocols. We will contribute the standardization protocols for Solid-phase protein interaction assay and IC50 assay to the same as a part of our project.

Assays can yield unexpected or failing results for a variety of reasons. Even simple assays include multiple steps where mistakes are likely to happen, which significantly increases the difficulty of finding the error sources.

Some potential error sources could include: incorrect, contaminated or expired reagents, equipment error, technician error, and many more. Therefore, resolving the cause of unexpected results is often a time consuming and expensive task in laboratory research. After studying the protocols for each assay in detail, we found some possible errors listed below.

  1. Solid-phase protein interaction assay

    We need a particular concentration of proteins for proceeding to protocol 3 in this assay. But we might end up getting different concentration values for the bicinchoninic acid (BCA) assay.This has to be changed to the needed concentration for accurate results.

  2. Half maximal inhibitory concentration (IC50) assay

    Due to handling errors, we may get less growth in the zero drug well than in any well with some drug concentration. This might be due to the addition of more cells in a well or more dye in a well.Also,the solvent we use might contribute to the inhibition effect.Our results may contain some background noise which have to be deducted from readings for more accuracy.

  3. RBC invasion assay

    Treating infected erythrocytes with α-2-3,6,8– Vibrio cholera neuraminidase trypsin and chymotrypsin can give us unexpected results as these three chemicals can block invasion completely. Because of lack of standardization, there is a chance of FACS not giving promising results.Usage of sorbitol in synchronisation of ring stage in infected erythrocytes can affect reinvasion.

One mode of decreasing assay failure is to ensure equipment errors are removed or at least minimised. This may be done by optimizing the equipment employed before running the assay. Analyzing the possible error sources that can hinder our assay process, we came up with some precautions and troubleshooting techniques that can be adapted to modify these assays.

  1. Solid-phase protein interaction assay

    Depending on what concentration we get in the BCA assay, we will increase or decrease to attain the concentration we need.

  2. Half maximal inhibitory concentration (IC50) assay

    To avoid handling errors,this experiment is done in triplets and their average is taken. RPMI and DMSO control are done to check whether the drug solvent is responsible for any inhibition effect or not. DMSO is the drug solvent. If the fluorometer readings of DMSO control and RPMI control are similar then that means that the solvent is not contributing to the inhibition effect.A blank control with no parasites can be performed to deduct the background noise from all readings.To check if the proposed drug acts as an inhibitor we first select a broad drug concentration range and if the IC50 value is obtained then the experiment is performed again with a narrow drug concentration range to get a more accurate value.

  3. RBC invasion assay

    Treatment of ring stage infected O+ erythrocytes with a solution of α-2-3,6,8– Vibrio cholera neuraminidase, trypsin and chymotrypsin can be done using lower concentrations and the relative effectiveness of different peptide inhibitors can be studied.If FACs fail to give promising results, fluorometry based analysis can be performed using a spectrophotometer, which will be speedier as well. To avoid sorbitol treatment as much as possible, fresh parasite cultures can be revived to grow in a synchronized manner for the initial cycles.

We developed each inhibitor sequence after an intensive computational analysis of each interaction. Although, in the worst scenario, there might come a time when these inhibitors fail. We targeted crucial interactions between the host and parasite protein that would have the greatest possibility of successfully preventing entry and infection of the Plasmodium parasite.

Directed Evolution(DE), a method used in protein engineering that mimics natural selection, can be used as our last expedient if our inhibitors fail to provide results. DE generates random mutations in the gene of interest and can be used as an alternative to rationally designing modified proteins. Site-directed mutagenesis(SDM) is an invitro method used in DE for creating a specific mutation in a known sequence. In our project, we can use this method to select or screen for mutations with the desired property of inhibiting our studied interactions. The simplest and most broadly applicable SDM protocol is the QuickChange Site-Directed Mutagenesis System (QCM) developed by Stratagene, which we plan to use in our project.

Experiments Page

References

1: Fairhurst, R. M., & Dondorp, A. M. (2016). Artemisinin-Resistant Plasmodium falciparum Malaria. Microbiology Spectrum, 4(3).
DOI: 10.1128/microbiolspec.ei10-0013-2016

2: Gould A, Camarero JA. Cyclotides: Overview and Biotechnological Applications. Chembiochem. 2017;18(14):1350-1363.
DOI: 10.1002/cbic.201700153

3: Gould A, Ji Y, Aboye TL, Camarero JA. Cyclotides, a Novel Ultrastable Polypeptide Scaffold for Drug Discovery. Curr Pharm Des. 2011;17(38):4294-4307.
DOI: 10.2174/138161211798999438

4: Dondorp, A. M., Nosten, F., Yi, P., Das, D., Phyo, A. P., Tarning, J., Lwin, K. M., Ariey, F., Hanpithakpong, W., Lee, S. J., Ringwald, P., Silamut, K., Imwong, M., Chotivanich, K., Lim, P., Herdman, T., An, S. S., Yeung, S., Singhasivanon, P., … White, N. J. (2009). Artemisinin Resistance in Plasmodium falciparum Malaria. New England Journal of Medicine, 5, 455–467.
DOI: 10.1056/nejmoa0808859

5: WHO | Malaria. (n.d.). WHO | World Health Organization. Retrieved October 20, 2020, from www.who.int/ith/diseases/malaria/en/

6: Gould, A., Ji, Y., L. Aboye, T., & A. Camarero, J. (2011). Cyclotides, a Novel Ultrastable Polypeptide Scaffold for Drug Discovery. Current Pharmaceutical Design, 38, 4294–4307.
DOI: 10.2174/138161211798999438

7: Bank, P. D. (n.d.). RCSB PDB: Homepage. Retrieved October 11, 2020, from www.rcsb.org

8: igemsoftware2020. (n.d.). GitHub - igemsoftware2020/IISER-Pune-India: Inhibitor peptides against falciparum Malaria and a Deep learning Web API for Malaria Diagnosis. GitHub. Retrieved October 20, 2020, from github.com/igemsoftware2020/IISER-Pune-India

9: Schymkowitz, J., Borg, J., Stricher, F., Nys, R., Rousseau, F., & Serrano, L. (2005). The FoldX web server: an online force field. Nucleic Acids Research, Web Server, W382–W388.
DOI: 10.1093/nar/gki387

10: Indian Biological Engineering Competition (iBEC) PRE- iGEM Competition | Department of Biotechnology. (n.d.). Home | Department of Biotechnology. Retrieved October 20, 2020, from dbtindia.gov.in/schemes-programmes/building-capacities/awards/indian-biological-engineering-competition-ibec-pre

11: Webb, B., & Sali, A. (2014). Comparative Protein Structure Modeling Using MODELLER. Current Protocols in Bioinformatics, 1, 5.6.1-5.6.32.
DOI: 10.1002/0471250953.bi0506s47


Team IISER Pune India