Team:Thessaly/Design
DESIGN.
We created ODYSSEE in order to provide healthcare workforce across the world with a modular platform for the diagnosis of Tuberculosis, to be used in field applications. Our design is based on a cell-free system combining an isothermal amplification method of a specific DNA fragment with a one-step in vitro transcription/translation reaction. This allows for an enzymatic assay to take place and, consequently, for the detection result to be seen with a naked eye. ODYSSEE’s nature is universal, being able to detect many other pathogens, such as Hepatitis B Virus (HBV) which we tried to detect. On this page, you can discover the design behind ODYSSEE.
Our project was designed carefully, so that it would be linked firmly with the needs of the people it aims to affect. As mentioned in the project description, our goal is to replace the existing conventional diagnostics for Tuberculosis with a modular, universal diagnostic platform. Our test’s design is based on a cell-free system and by using an engineering approach, we managed to break it down into modules in order to work on them in parallel. Eventually, we aim at integrating all these modules into a final product, ready to be applied in field.
Biological sample and targeted biomarker
After interacting with different stakeholders about the use of different biological samples for our test, we concluded that urine was our best choice. As we were focusing on urine, in order to find a suitable biomarker, we discovered that we had limited choices and only a few of them were highly specific [1,2]. Mycobacterium tuberculosis (MTB) is not the only bacterium species that can cause Tuberculosis (TB); a lot of microorganisms from the Mycobacterium tuberculosis complex (MTBC) can do so as well [3]. A conserved gene commonly used for the detection of MTB, is IS6110. IS6110 belongs to the family of IS3 category and appears in multiple copies (up to 25 per genome) in MTB strains [4,5]. This specific and conserved sequence within the MTB genome can be found in the urine of patients with active TB, in the form of fragments approximately 150 bp long [6,7]. In order to detect these fragments, without the need for specialized personnel and equipment, we want our test to be able to function in relatively normal temperatures, not lower than 25 degrees and not higher than 42 degrees Celsius, while its readout must be visible and easy to interpret, without the need for expensive equipment.
A vague description of the steps in our design
The workflow of our diagnostic test designed to detect the specific and conserved IS6110, which is found in urine samples of TB patients in the form of fragments, consists of 3 steps:
1. Isothermal DNA amplification of the MTB DNA fragment, with the incorporation of two universal sequences, into 5’ and 3’ end respectively.
2. A one-step in vitro transcription/translation of the DNA sequence and a toehold switch regulated mRNA, that encodes for a reporter protein.
3. Colorimetric readout of the signal through a rapid enzymatic assay, changing the color of the reaction from yellow to red.
We aim to outline all the steps of our diagnostic platform and the purpose of their use below.
Figure 1. The workflow of our designed diagnostic test.
Isothermal DNA Amplification
The first step that we had to incorporate into our test, in order to increase the sensitivity resulting in a robust signal, was the amplification step. The main barrier of amplification methods – e.g. PCR – reaching everyday use in Low Resource Settings (LRS), is the need for specialized equipment (thermocycler), in order to rapidly regulate the temperature. To circumvent this major issue, we turned our interest towards isothermal amplification methods and especially the Recombinase Polymerase Amplification method (RPA), which not only works in low temperatures (37-42°C) but can amplify a DNA sequence (100-200 bp) in as low as 5 minutes [8,9].
An amplified sequence can be visualized directly, but the sensitivity of the assay is rather low [10]. In order to increase the sensitivity of the detection, we engineered the produced DNA sequence using RPA, to incorporate two universal sequences.
The first sequence consists of a T7 promoter, enabling the molecule to be transcribed into RNA in the presence of a T7 RNA Polymerase. In this way, we also enable the molecule to produce more copies of itself during transcription, increasing the sensitivity, too. The second sequence is a trigger DNA sequence that was carefully chosen so that it is a unique universal synthetic sequence, which is not found in nature and does not have any similarity to the human genome or to the human microbiome. This sequence is transcribed into RNA from a T7 RNA Polymerase in order to act as a trigger molecule in the following steps of our design. For these sequences to be incorporated into our design, we had to carefully design and test, first in silico and then in vitro, our primers of 30 bp length and incorporate 5’ overhangs in each primer. The T7 promoter sequence was incorporated into the forward primers as a 5’ overhanging sequence while the trigger sequence was incorporated into the reverse primers, also as a 5’ overhanging sequence. This engineering step - which only includes primer design – results in the amplification of a targeted sequence which also has new sequences in its ends that allow this device to orthogonally interact with another device in a downstream step, to produce a highly specific signal. In this way, we provide specific and sensitive DNA detection, with just a set of primers.
Figure 2. Recombinase Polymerase Amplification Method.
Toehold switch Regulation
After engineering the DNA biomarker whilst amplifying it, we started working on our detection system. We decided that the most viable and robust way of detection was through designing a toehold switch, which is orthogonally activated by the amplified biomarker and produces specific and sensitive signal.
A toehold switch is a precise RNA device, which forms a secondary hairpin structure and is used for the regulation of gene expression, during the translation stage. The sequence of the switch contains a strong ribosome binding site (RBS) and a start codon (AUG) followed by the coding sequence of a reporter gene. The hairpin conformation acts as a translational repressor, preventing ribosomes from binding to the RBS, thus inhibiting translation. In the presence of a single stranded trigger, complementary to the stem of the hairpin – in this case, our engineered biomarker – the switch unfolds and exposes both the RBS and start codon, enabling the initiation of translation [11-13].
Figure 3. The Toehold Switch mechanism.
In addition to the commercial toehold switch design, which is based on the target DNA sequence, we were trying to develop a universal tool which would be uncoupled from the MTB sequence, so it can be applied in the DNA detection of different pathogenic agents. For this reason, we designed a universal and orthogonal synthetic DNA trigger sequence, based on hyperthermophile bacteria and archaea, such as Geobacillus kaustophilus and Dictyoglomus turgidum. To achieve this, we combined the sequence of the original toehold switches from Keith Pardee’s paper with rRNA sequences from hyperthermophilic bacteria [12].
Figure 4. Our own designed trigger sequence derived from Geobacillus kaustophilus.
Creating a universal tool for DNA detection
In order to prove the universal nature of our design, we decided to detect one more disease with our tool. Aiming to mimic the conditions of our first developed detection scheme, we decided to detect DNA fragments that are also found in urine, derived from the Hepatitis B Virus (HBV). It is a small DNA fragment (280bp), that does not encode for any functional protein or toxin. It's a small part of a sequence that encodes for a DNA polymerase of the virus and in alternative reading frames it encodes for envelope proteins of the virus, or for a specific surface antigen [14-16].
The change that we applied to the previously designed system was the primer set, by doing it specific to the HBV genome. This is the only change that everyone who is going to use our detection system has to incorporate into the design so that it can be applied for every DNA sequence detection.
Visible Readout
Whilst working on the design of our detection system, we wanted to make sure that, no matter the device and the processes, the readout of the test is going to be easily readable, without the need for expensive equipment. Thus, we focused primarily on colorimetric assays and specifically on enzymatic reactions for the visualization of the result, so that it can be easily observed with a naked eye [17,18].
After a thorough research for the suitable readout, we came across the β-lactamase enzyme which hydrolyzes the β-lactam ring, a structure shared by the β-lactam class of antibiotics. It is a small 29kDa monomeric enzyme which provides an advantage for our system, since small proteins are produced faster than bigger ones, like β-galactosidase (75kDa). Additionally, β-lactamase can also be used to obtain colorimetric outputs by breaking down synthetic compounds such as nitrocefin, a cephalosporin with antibiotic activity. The result of nitrocefin hydrolysis is a color change from yellow (390nm) to red (490nm) [19]. In this way, the result would be seen with a naked eye, specially tailored for in field tests. Finally, we also tested a fluorescent assay, by using eGFP as a reporter gene in order to quantify the toehold’s activity and our system’s efficiency in both assays.
Both for β-lactamase and eGFP, the activation pathway is the same. The universal trigger sequence binds to the complementary sequence of the toehold switch, leading to its linearization. The RBS and the staring codon are revealed and the translation of the downstream gene would be activated. In case of the eGFP UV light is needed to display the signal but in the other hand, β-lactamase needs only a drop of the nitrocefin substrate and the color-change can be visible with a naked eye in less than 30 minutes.
Figure 5. Expression of the β-lactamase enzyme regulated by a toehold switch and the hydrolysis of the chromogenic substrate, nitrocefin.
Positive Control
How could we make sure that a negative result is obtained due to the absence of MTB DNA in urine, and not because the test’s components are not functional? In order to answer that question, we are implementing a positive control on our diagnostic test, which will be performed in parallel with the actual result. This will ensure that the test’s components are functional and the result in the TB test is reliable.
Our selected Positive Control biomarker is a mitochondrial gene, COX3. The Cytochrome-c Oxidase 3 (COX3) gene encodes for an enzyme that takes part in the respiratory process of mitochondria. When cells die, fragments of mitochondrial DNA may escape from the matrix of the cytosol and end up in the systemic circulation. There are several studies which confirm that fragments of COX3 DNA are detectable in healthy individuals. We aim to presume upon this fact to detect these fragments as a positive control for our test, since they are always present in urine. The detection of COX3 could therefore be a reliable marker of our test’s reliability [20-21].
References
1. Goletti, D Elia, Mengui Lee, and Jannuan Wang. 2018. “INVITED REVIEW SERIES : TUBERCULOSIS UPDATES 2018 Update on Tuberculosis Biomarkers : From Correlates of Risk , to Correlates of Active Disease and of Cure from Disease,” 455–66. https://doi.org/10.1111/resp.13272.
2. Druszczynska, Magdalena, Sebastian Wawrocki, Rafal Szewczyk, and Wieslawa Rudnicka. 2017. “Mycobacteria-Derived Biomarkers for Tuberculosis Diagnosis,” no. December: 700–707. https://doi.org/10.4103/ijmr.IJMR.
3. Floyd, K., Raviglione, M., & Glaziou, P. (2018). Global Epidemiology of Tuberculosis. Seminars in Respiratory and Critical Care Medicine, 39(03), 271–285. doi:10.1055/s-0038-1651492
4. Thabet, S., & Souissi, N. (2016). Transposition mechanism, molecular characterization and evolution of IS6110, the specific evolutionary marker of Mycobacterium tuberculosis complex. Molecular Biology Reports, 44(1), 25–34.doi:10.1007/s11033-016-4084-x
5. Kabir S, M Khaja, M Jobayer, T Fannana, M Enamul, M Reaj, S Banu, and T Ahmed. 2018. International Journal of Infectious Diseases "Role of PCR Method Using IS6110 Primer in Detecting Mycobacterium Tuberculosis among the Clinically Diagnosed Childhood Tuberculosis Patients at an Urban Hospital in Dhaka”, International Journal of Infectious Diseases, 68: 108–14. https://doi.org/10.1016/j.ijid.2018.01.015.
6. Geldmacher C, Dheda K, Hoelscher M, Labugger I. Evaluation of a Urine-Based Rapid Molecular Diagnostic Test with Potential to be used at Point-of-Care for Pulmonary Tuberculosis: Cape Town Cohort. The Journal of Molecular Diagnostics;(2018). doi:10.1016/j.jmoldx.2017.11.005
7.Labugger I, Heyckendorf J, Dees S, Häussinger E, Herzmann C, Kohl T A, Richter E, Milla E R, Lange C. Detection of transrenal DNA for the diagnosis of pulmonary tuberculosis and treatment monitoring, Infection (2016). doi:10.1007/s15010-016-0955-2
8. Piepenburg O, Williams CH, Stemple DL, Armes NA (2006) DNA Detection Using Recombination Proteins. PLoS Biol 4(7): e204.
9. I.M. Lobato, C.K. O’Sullivan, Recombinase Polymerase Amplification: Basics, applications and recent advances, Trends in Analytical Chemistry (2017)
10. Daher, R. K., Stewart, G., Boissinot, M., & Bergeron, M. G. (2016). Recombinase Polymerase Amplification for Diagnostic Applications. Clinical Chemistry, 62(7), 947–958. doi:10.1373/clinchem.2015.245829
11. Green, Alexander A, Pamela A Silver, James J Collins, and Peng Yin. 2014. “Resource Toehold Switches : De-Novo-Designed Regulators of Gene Expression.” Cell 159 (4): 925–39. https://doi.org/10.1016/j.cell.2014.10.002.
12. Pardee Keith, Alexander A Green, Melissa K Takahashi, David H O Connor, Lee Gehrke, James J Collins (2016). “Rapid , Low-Cost Detection of Zika Virus Using Programmable Biomolecular Components Resource Rapid , Low-Cost Detection of Zika Virus Using Programmable Biomolecular Components.” Cell, 1–12. https://doi.org/10.1016/j.cell.2016.04.059.
13. Pardee, Keith, Alexander A Green, Tom Ferrante, D Ewen Cameron, Ajay Daleykeyser, and Peng Yin (2014). “Resource Paper-Based Synthetic Gene Networks.” Cell 159 (4): 940–54. https://doi.org/10.1016/j.cell.2014.10.004.
14. Alves, Danielle, Gomes Zauli, Carla Lisandre, Paula De Menezes, Cristiane Lommez, De Oliveira, Elvis Cristian, Cueva Mateo, Alessandro Clayton, and De Souza Ferreira. 2016. “Genetics and Molecular Microbiology In-House Quantitative Real-Time PCR for the Diagnosis of Hepatitis B Virus and Hepatitis C Virus Infections.” Brazilian Journal of Microbiology 47 (4): 987–92. https://doi.org/10.1016/j.bjm.2016.07.008.
15.Jain, Surbhi, Ying-hsiu Su, Yih-ping Su, Sierra Mccloud, Ruixia Xue, Tai-jung Lee, Shu-chuan Lin, et al (2018) “Characterization of the Hepatitis B Virus DNA Detected in Urine of Chronic Hepatitis B Patients,” 19–21.
16. Karayiannis, P, D M Novick, A S F Lok, and H C Thomas. 1985. “Hepatitis B Virus DNA in Saliva , Urine , and Seminal Fluid of Carriers of Hepatitis B e Antigen” 290 (June): 1853–55.
17. Boehle, Katherine E., et al. “Paper-Based Enzyme Competition Assay for Detecting Falsified β-Lactam Antibiotics.” ACS Sensors, 3 (7) 2018, pp.1299–1307., doi:10.1021/acssensors.8b00163.
18. Qureshi, Sohail A. “β-Lactamase: an Ideal Reporter System for Monitoring Gene Expression in Live Eukaryotic Cells.” BioTechniques, vol. 42, no. 1, 2007, pp. 91–96., doi:10.2144/000112292.
19. Remy, I., Ghaddar, G., & Michnick, S. W. (2007). Using the β-lactamase protein-fragment complementation assay to probe dynamic protein–protein interactions. Nature Protocols, 2(9), 2302–2306. doi:10.1038/nprot.2007.356
20. Eirin, Alfonso, Ahmed Saad, Hui Tang, Sandra M Herrmann, John R Woollard, Amir Lerman, Stephen C Textor, and Lilach O Lerman. 2017. “URINARY MITOCHONDRIAL-DNA COPY NUMBER IDENTIFIES CHRONIC RENAL INJURY IN HYPERTENSIVE PATIENTS” 68 (2): 401–10.
21. Hu, Qiongyuan, Jianan Ren, Jie Wu, Guanwei Li, Xiuwen Wu, Song Liu, Gefei Wang, et al. 2017. “URINARY MITOCHONDRIAL DNA LEVELS IDENTIFY ACUTE KIDNEY INJURY IN SURGICAL CRITICAL ILLNESS PATIENTS” 48 (1): 11–17. https://doi.org/10.1097/SHK.0000000000000830.
