Project Design

Background knowledge of our target disease

Alzheimer’s disease (AD) is a debilitating neurodegenerative disease characterised by the formation of highly stable amyloid β-plaques within the central nervous system that causes progressive loss of memory. By the year 2050, it is estimated that roughly 131 million people will be diagnosed with Alzheimer’s disease (Tiwari, Atluri, Kaushik, Yndart, & Nair, 2019). The origins of Alzheimer’s disease are difficult to understand due to a lack of reliable markers to determine a patient’s likelihood of developing the disease. Genome-wide association studies on AD have shown limited success in identifying causal genes associated with AD making it difficult to attribute a completely genetic origin to the disease (Wightman et al., 2021). Other factors such as diet, senescence and auto-immune history are also major contributors to the pathology of AD (Yusufov, Weyandt, & Piryatinsky, 2017). For the first step of our project, it was important for us to take this complicated disease and identify a specific target that we can effectively treat. Through researching literature and discussions with experts in polyphenols, AD, and our targeted pathway (NLRP3/caspase-1), we understand to a much better extent, the pathology and effects of Alzheimer’s disease. More detailed information on AD and pterostilbene in relation to AD can be found on our AD pathophysiology page.

Our first step was identifying commonly used categorizations of AD to target one specific pathology of the disease. AD has traditionally been divided into two types, early-onset and late-onset Alzheimer’s disease (Koedam et al., 2010). Early-onset AD is categorised as AD that develops <65 years of age while late-onset AD is characterised by development at >65 years of age. This distinction is largely arbitrary however as most individuals within the early-onset group who develop Alzheimer’s are of the 45-64 year category with a prevalence of 24.2/100 000 individuals in the USA (Reitz, Rogaeva, & Beecham, 2020). Therefore, our team decided to investigate a better division of AD to help us identify one pathology of AD we can effectively target.

While age may not be a defining characteristic of AD, another notable distinction is that of the intensity of symptoms at different stages. AD typically progresses slowly in 3 stages: early, middle, late (sometimes referred to as mild, moderate, severe). Since AD affects patients in different ways, each person may experience symptoms, or progress through the stages differently. On average, a person with AD lives 4-8 years after diagnosis, but they can live as long as 20 years, depending on other factors (Kumar et al., 2022).

The two pathologic hallmarks of AD are extracellular beta-amyloid deposits (in senile plaques) and intracellular neurofibrillary tangles (paired helical filaments) (Murphy & LeVine, 2010). A sustained immune response and inflammation have also been observed in the brain of patients with AD and some experts have proposed that inflammation is the 3rd core pathologic feature of AD (Kinney et al., 2018).

As importantly, we also now know the value our project would potentially bring as a therapeutic. From there, our project direction is validated and our design built to fit our aim of stagnating mild to moderate AD using therapeutically-effective concentrations of pterostilbene synthesised by genetically engineered E. coli.

Producing pterostilbene to treat Alzheimer's disease

The Genes involved

After establishing pterostilbene’s capacity to treat AD and its interaction with the NLRP3 inflammasome, we looked into methods of producing pterostilbene with a view to investigating its activity in vitro. We identified a synthetic biology approach as the best method of producing pterostilbene as a part of our proof of concept due to the economic pressure the extraction of pterostilbene would have on the environment through traditional production methods.

To produce pterostilbene from E. coli, our team first needed to identify the pathway commonly used in biosynthesis to produce polyphenols. In most papers detailing the biosynthesis of pterostilbene, four enzymes are required for pterostilbene biosynthesis (Heo, Kang, & Hong, 2017; Kang et al., 2014; Yan, Liang, Niu, Shen, & Liu, 2021):

1. A tyrosine ammonia-lyase enzyme that catalyses the deamination of tyrosine into 4-coumaric acid.
2. A 4-coumarate:CoA ligase enzyme that catalyses the ligation of 4-coumaric acid with Coenzyme A via a phosphate group obtained by ATP hydrolysis.
3. A stilbene synthase enzyme that produces resveratrol from a 4-coumaroyl-CoA and three malonyl-CoA molecules.
4. A resveratrol O-methyltransferase enzyme or equivalent (caffeic acid O-methyltransferase was used by Heo et al., 2017) to add one methyl group to resveratrol to produce pinostilbene and another methyl group to produce pterostilbene.

These four enzymes have a wide variety of orthologues expressed in many organisms. The following enzyme orthologues chosen were chosen based on literature identifying these enzymes as optimal for pterostilbene synthesis:

1. Rhodotorula glutinistyrosine ammonia-lyase (RgTAL)
2. Arabidopsis thaliana4-coumarate:CoA ligase (At4CL)
3. Vitis viniferastilbene synthase (VvSTS)
4. Vitis viniferaresveratrol O-methyltransferase (VvROMT)

The enzymes At4CL and VvSTS were already present in the iGEM registry under the codes BBa_K1825008 and BBa_K1033001 respectively. Notably compared to the NCBI GenBank sequence DQ459351.1 for VvSTS, the VvSTS sequence BBa_K1825008 has an additional methionine at amino acid position 50. Because no advantage of this extra methionine was documented on the parts page, the GenBank sequence was chosen as the primary source for the VvSTS sequence.

From these sequences our team set about identifying potential mutations that might increase the catalytic efficiency of these enzymes to maximise pterostilbene yield. Yan et al. (2021) identified a series of mutations that when applied to the enzymes RgTAL, At4Cl, VvSTS and VvROMT, increased the pterostilbene production titre by a factor of 13.7. Due to this significant increase in pterostilbene, our team decided to implement the following mutations into the wild type enzymes at positions specified by Yan et al. (2021) (Table 1). Further justifications of each enzyme can be found on our parts page.

Table 1.Depicting mutations added to genes used for the goal of producing pterostilbene.

Having identified enzymes that we wished to express in our system we needed to first understand what expressions system and assembly standard we would use. The first iteration of our design involved the four genes all expressed under a single promoter and separated by their own ribosomal binding sites. This design is a derivative of the iGEM Uppsala 2013 plasmid construction for producing resveratrol. In their system, the genes At4CL and VvSTS were placed under the control of a singular promoter yielding an mRNA that encoded for both genes, separated by a ribosomal binding site (Figure 2).

Parts Domestication and RFC standard

Prior to construction of plasmids it was important our team domesticated our parts according to a set workflow (see figure 3.).

The presence of a signal peptide in a protein can result in undesirable subcellular localisation or even secretion so it was important for our team to account for this by screening all four enzymes for potential signal sequences. Determining the presence of signal sequences involved a separate workflow using two programmes:

SignalP 5.0

Signal P 5.0 is a programme used for determining the presence of signal sequences at the N-terminus of a protein. It has options for testing for signal sequences from Eukarya, Gram-negative bacteria, Gram-positive bacteria and Archaea. Developed by the Technical university of Denmark, SignalP 5.0 tests for three different signal peptides (Almagro Armenteros et al., 2019)

- Sec/SPI: "standard" secretory signal peptides transported by the Sec translocon and cleaved by Signal Peptidase I (Lep)

-Sec/SPII: lipoprotein signal peptides transported by the Sec translocon and cleaved by Signal Peptidase II (Lsp)

- Tat/SPI: Tat signal peptides transported by the Tat translocon and cleaved by Signal Peptidase I (Lep).

All enzymes were tested for signal sequences in all four organisms groups with a probability score of >0.1 used as a cutoff below which a signal peptide was unlikely as suggested by our human contact Obrvan Marko (see table 2).

Having excluded signal peptides and optimised codon sequences to E.coli standards, our team investigated how to construct our plasmid. It soon became clear that traditional methods of assembly such as 3A assembly using RFC10 assembly standard would not be suitable for a multi-transcriptional unit system due to the increased complexity yielded by serial ligation. Instead, it was necessary to construct our plasmid in a hierarchical manner starting with basic parts followed by transcriptional units which are finally ligated into a multi-transcriptional unit plasmid.

For this our team decided to use a golden gate assembly standard using BsmBI and BsaI. The iGEM distribution kit made available a series of joint universal modular plasmids (JUMP) which were designed to allow for type IIS assembly using the restriction enzymes BsaI and BsmBI. Type IIS restriction enzymes have a cleavage site that is separate from the recognition site which means overhangs are non-specific and can be changed to allow for complex ordered ligations (Valenzuela-Ortega & French, 2021).

An issue we quickly identified however was that the plasmid sequences available on the iGEM registry differed from those provided by the supplementary material of the original paper (Valenzuela-Ortega & French, 2021). Additionally, sequencing data that is commonly given to teams to confirm sequence identity of the distribution kits was made unavailable. We got into contact with Marcos Valenzuela-Ortega and were kindly provided with protocols to follow to successfully perform Type IIS assembly on JUMP plasmids. Details of Type IIS assembly are provided in our guide which details how a team may be able to use type IIS assembly using JUMP plasmids in the construction of their own gene systems.

Due to the lack of sequencing data, our team decided to conduct electrophoresis digestion on the following plasmids to ensure that sequences were those identified either on addgene or on the parts registry.

Promoter Strength, Plasmid Copy number and In silico modeling

Due to the use of T7 promoters in many papers detailing pterostilbene production, our team decided to use T7 promoters and terminators in the earliest iteration of our project(Heo et al., 2017; Kang et al., 2014; Yu Jeong et al., 2015). Our team soon discovered a design issue regarding such a system, recombination. Because the promoters and terminators were identical, when the plasmid replicates there would be a high risk of recombination resulting in some genes potentially being excised out (Grengross et al.,2022). While Heo et al. (2017) was successful in expressing genes for pterostilbene production it was unclear as to how this risk was avoided. As a result, our team decided to look into another expression system, a synthetic operon. A synthetic operon is a method by which several genes can be expressed as one mRNA separated by RBS. Because this is the method by which E.coli expressed genes there is a reduced risk of recombination occurring.

Our team desired to create our plasmids with a high amount of rigour. As such we wanted to find a method to accurately determine which promoters and plasmid copy numbers to use. We acknowledged that improper selection of these factors could overwhelm our E.coli cell and as such we decided to investigate the factors in silico. This resulted in the kinetic model of E.coli production of pterostilbene. Using an early version of this model, we determined that the copy number ideal for pterostilbene production was 13 plasmids which was close to the 15-20 plasmids produced by the pBR322 origin.

Type II assembly

Because of the presence of four transcriptional units needed to be put into a singular plasmid, our design was well suited to Type IIS assembly. The main Type IIS assembly system supported by iGEM was the RFC1000 system. This system made use of BsaI and SapI to allow for insertion of transcriptional units consisting of a promoter, 5’Un-translated region, coding domain sequence as well as terminator. Upon looking through the registry kit however, we realized that the plasmids available were not of the RFC1000 system. We soon learned that this was a newer type of plasmid known as “joint universal modular plasmids” or JUMP. These plasmids were made by Marcos-Valenzuela Ortega and Christopher French to unify two different plasmid standards, the BioBrick assembly standard as well as the standard european vector architecture SEVA. Unlike RFC1000 assembly, this system makes use of the enzyme BsmBI instead of SapI. We constructed our plasmids accordingly, getting in touch with Marcos-Valenzuela Ortega to get insight into how to use these plasmids for specific usage. We also were kindly provided protocols by Marcos-Valenzuela Ortega which we made use of in our experiments. To allow future iGEM teams to understand JUMP assembly, our team also constructed a guide that is made available on our contributions page to allow future iGEM teams to more easily navigate these plasmids.

References

Almagro Armenteros, J. J., Tsirigos, K. D., Sønderby, C. K., Petersen, T. N., Winther, O., Brunak, S., . . . Nielsen, H. (2019). SignalP 5.0 improves signal peptide predictions using deep neural networks. Nature Biotechnology, 37(4), 420-423. doi:10.1038/s41587-019-0036-z

Elofsson, H. (2022). Part:BBa K1033003 - parts.igem.org. Retrieved 28 August 2022, from http://parts.igem.org/Part:BBa_K1033003

Heo, K. T., Kang, S.-Y., & Hong, Y.-S. (2017). De novo biosynthesis of pterostilbene in an Escherichia coli strain using a new resveratrol O-methyltransferase from Arabidopsis. Microbial Cell Factories, 16(1). doi:10.1186/s12934-017-0644-6

Kang, S.-Y., Lee, J. K., Choi, O., Kim, C. Y., Jang, J.-H., Hwang, B. Y., & Hong, Y.-S. (2014). Biosynthesis of methylated resveratrol analogs through the construction of an artificial biosynthetic pathway in E. coli. BMC Biotechnology, 14(1), 67. doi:10.1186/1472-6750-14-67

Kim, K. J., Kim, H. E., Lee, K. H., Han, W., Yi, M. J., Jeong, J., & Oh, B. H. (2004). Two-promoter vector is highly efficient for overproduction of protein complexes. Protein Sci, 13(6), 1698-1703. doi:10.1110/ps.04644504

Koedam, E. L., Lauffer, V., van der Vlies, A. E., van der Flier, W. M., Scheltens, P., & Pijnenburg, Y. A. (2010). Early-versus late-onset Alzheimer's disease: more than age alone. J Alzheimers Dis, 19(4), 1401-1408. doi:10.3233/jad-2010-1337

Reitz, C., Rogaeva, E., & Beecham, G. W. (2020). Late-onset vs nonmendelian early-onset Alzheimer disease. Neurology Genetics, 6(5), e512. doi:10.1212/nxg.0000000000000512

Tiwari, S., Atluri, V., Kaushik, A., Yndart, A., & Nair, M. (2019). <p>Alzheimer’s disease: pathogenesis, diagnostics, and therapeutics</p>. International Journal of Nanomedicine, Volume 14, 5541-5554 doi:10.2147/ijn.s200490

Valenzuela-Ortega, M., & French, C. (2021). Joint universal modular plasmids (JUMP): a flexible vector platform for synthetic biology. Synthetic Biology, 6(1). doi:10.1093/synbio/ysab003

Wightman, D. P., Jansen, I. E., Savage, J. E., Shadrin, A. A., Bahrami, S., Holland, D., . . . Posthuma, D. (2021). A genome-wide association study with 1,126,563 individuals identifies new risk loci for Alzheimer’s disease. Nature Genetics, 53(9), 1276-1282. doi:10.1038/s41588-021-00921-z

Yan, Z.-B., Liang, J.-L., Niu, F.-X., Shen, Y.-P., & Liu, J.-Z. (2021). Enhanced Production of Pterostilbene in Escherichia coli Through Directed Evolution and Host Strain Engineering. Frontiers in Microbiology, 12 doi:10.3389/fmicb.2021.710405

Yu Jeong, J., Su Gyeong, W., Chul Han, A., Hyung Jae, J., Young-Soo, H., Young-Min, K., . . . and Cha Young, K. (2015). Metabolic Engineering for Resveratrol Derivative Biosynthesis in <italic>Escherichia coli</italic>. Molecules and Cells, 38(4), 318-326 doi:10.14348/molcells.2015.2188

Yusufov, M., Weyandt, L. L., & Piryatinsky, I. (2017). Alzheimer's disease and diet: a systematic review. Int J Neurosci, 127(2), 161-175. doi:10.3109/00207454.2016.115557

Lim, C. G., Fowler, Z. L., Hueller, T., Schaffer, S. et Koffas, M. A. G.. (2011). High-Yield Resveratrol Production in Engineered Escherichia coli. Applied and environmental microbiology, 77(10), 3451‑3460. doi:10.1128/aem.02186-10

Wu, J., Zhou, P., Zhang, X. et Dong, M.. (2017). Efficient de novo synthesis of resveratrol by metabolically engineered Escherichia coli. Journal of industrial microbiology and biotechnology, 44(7), 1083‑1095. doi:10.1007/s10295-017-1937-9

Yang, Y., Lin, Y., Li, L., Linhardt, R. J., & Yan, Y. (2015). Regulating malonyl-CoA metabolism via synthetic antisense RNAs for enhanced biosynthesis of natural products. Metabolic engineering, 29, 217–226. https://doi.org/10.1016/j.ymben.2015.03.018

Sun, T., Li, S., Song, X., Pei, G., Diao, J., Cui, J., . . . Zhang, W. (2018). Re-direction of carbon flux to key precursor malonyl-CoA via artificial small RNAs in photosynthetic Synechocystis sp. PCC 6803. Biotechnology for Biofuels, 11(1), 26. doi:10.1186/s13068-018-1032-0

Liang, J.-l., Guo, L.-q., Lin, J.-f., He, Z.-q., Cai, F.-j., & Chen, J.-f. (2016). A novel process for obtaining pinosylvin using combinatorial bioengineering in Escherichia coli. World Journal of Microbiology and Biotechnology, 32(6), 102. doi:10.1007/s11274-016-2062-z

Yoo, S. M., Na, D. et Lee, S. Y.. (2013). Design and use of synthetic regulatory small RNAs to control gene expression in Escherichia coli. Nature protocols, 8(9), 1694‑1707. doi:10.1038/nprot.2013.105

Morgan, K. (2022). Plasmids 101: Origin of Replication. Retrieved 4 October 2022, from . https://blog.addgene.org/plasmid-101-origin-of-replication

Gerngross, D., Beerenwinkel, N. et Panke, S.. (2022). Systematic investigation of synthetic operon designs enables prediction and control of expression levels of multiple proteins. doi:10.1101/2022.06.10.495604