Choice of DNA sequence
The two methods we have developed use cfDNA (cell-free DNA) as a biomarker, which means there is some wiggle room to choose which exact sequence is identified. In the current design, a sequence for cytochrome c oxidase subunit 1 which can be found in S.haematobium and S.japonicum has been used [1]. There are other possible sequences, so in determining the exact sequence there are three important parameters.Firstly, cfDNA concentration in the chosen type of sample is to be maximized. The creation mechanism of cfDNA from the parasite is not yet exactly known, but it is often spread by blood circulation, reaching a wide variety of possible sample types [1]. Searching for other sequences that can be used as biomarkers are important, but outside the scope of an iGEM project.
Secondly, a strong binding between the test systems and the cfDNA is wanted. Guide RNA (gRNA) is used in the cell-free system, but the binding strength of gRNA varies and it is difficult to predict [2]. Predictions can give some information and are an important first step before actual testing. In the cell-free system, the length between the dCas9 can be optimized. This was tested using modeling and yielded an optimum linker length of 16-19 nm, more than what is discussed in Model. Zinc-finger design follows a similar vein. Since zinc-fingers bind a specific sequence of 9 base pairs and generally can not be modified, there are more limitations in choice of target sequence [3].
Finally, off-site binding must be minimized. Off-site binding means false negatives for the test. There exists models that look for sequences similar to the binding sequence and estimate the binding strength of the off-site region. For gRNA, standardized methods exist [4] while zinc-fingers require a custom design to predict off-site binding. Later down the line, tests using human samples such as human saliva must be tried.
By taking these three parameters, a desirable cfDNA goal sequence can be found. The sequence we have used performs quite good on all three, but more tests are needed, especially for testing off-site binding.
Taking samples
There exist multiple possible liquids for test taking, including amongst other saliva, urin, serum, semen, stool [5] and nasal fluid. In choosing the right one, we have found three important parameters for choice of sample fluid. The first is the cfDNA of interest concentration. Secondly, the ease of collecting samples both in type of equipment and technique since the end goal is for more rural poorer communities. Finally, the ease of using the fluid in the following steps.The sequence for cytochrome c oxidase subunit 1 which the systems currently are designed for have been tested from blood serum, urine and saliva [1]. The easiest of these to obtain are saliva, which thus should be the focus on future exploration.
Cell-free system
The cell-free system can be suitably adopted as a lateral flow test [6]. The test is made up of two flow strips, the test and the control, to which both test liquids are added. Both strips contain X-gal, iGal, dCas9-nTEVp and dCas9-cTEVp at the conjugate release pad, which sits close to where the test liquid is added. One of the strips also contains the two gRNA sequences and this one works as the test strip. The other have the target DNA and work as a negative control. Near the end of the test flowstrip there would be TEVp, and preferably it would be anchored to the strip. This is a positive control since TEVp activates the cell-free system. The control flowstrip has instead gRNA sequences near the end which also start the signal chain since all parts are present. Thus, to have a valid test, the ends of both strips need to turn blue and the bulk of the control needs to stay white. The system is illustrated below.
Cell-based system
Implementation of the cell-based system must use living cells and thus freeze drying is a preferred technique to ease distribution and handling. In testing S. cerevisiae has survived but other species such as L. elongisporus work a lot better [7]. L.elongisporus is not suitable since it is a human blood pathogen but other species with similar qualities being small size and surviving high osmotic pressure should be tested [7]. After the freeze drying, the sample should be kept at 5 °C and be thawed before use.
We suggest using three wells. One for the actual test, one for negative control and one as positive control. Into all three wells, growth medium containing previously freeze dried yeast is added. The positive control contains the target DNA sequence. Alternatively, the mating factor Mf alpha can be used as a less complete control. The testing liquid is added to the test well. After waiting some time, the positive control should turn purple while the negative should not. The test is colored if it is a positive. The system is illustrated below.
Actual Implementation
The ideas presented above are all valid and relevant for the short term future of the project. For actually implementing a product like this, larger questions of economy and feasibility need to be taken care of. A diagnostic kit like the one we are developing, is usually paid by benevolent funds such as the Bill and Melinda Gates foundation, Ministries of Health, or non-profit organizations and not by the countries and communities where they are used according to Thierry Ramos and Sarah Nogaro at the FIND foundation. Limited funds means that it is critical that the test is not expensive. Russel Glanz at ITC diagnostics recommended at an interview 0.45 to 0.55 dollar per test at maximum with most lateral flow test being 0.25 to 0.30 dollars and more common ones being below 0.18 dollar. Our cell-based system probably compares quite well to this since it would essentially only require a bioreactor to grow S. cerevisiae.
ITC diagnostics also had other important criteria, a commercial test needs to meet. Test results are to be expected within 15 to 20 minutes, specificity should be near 100% and sensitivity should be above 80%. These criteria are hard to reach for every test and much work is needed to be able to achieve these. To get fast results, high enzyme and metabolic activity is desirable. Sensitivity and specificity is related to on-site and off-site binding strength, which is discussed above in choice of DNA sequence, together with how strongly the signal is amplified. Both systems have amplification steps, for the cell-free going from protein to metabolite gives a high increase in signal. The cell-free system uses cell to cell signaling, which makes the signal strength grow exponentially.
Overall, much work is left to reach a market product and many parameters need to be taken care of. But, if done the world would be changed to a better world.
References
[1] Weerakoon KG, McManus DP. Cell-free DNA as a diagnostic tool for human parasitic infections. Trends in Parasitology. 2016;32(5):378–91.[2] Corsi, G.I., Qu, K., Alkan, F. et al. CRISPR/Cas9 gRNA activity depends on free energy changes and on the target PAM context. Nat Commun 13, 3006 (2022). https://doi.org/10.1038/s41467-022-30515-0
[3] Acc. Chem. Res. 2006, 39, 1, 45–52 Publication Date:December 1, 2005 https://doi.org/10.1021/ar050158u
[4] Stemmer, M., Thumberger, T., del Sol Keyer, M., Wittbrodt, J. and Mateo, J.L. CCTop: an intuitive, flexible and reliable CRISPR/Cas9 target prediction tool. PLOS ONE (2015) doi: 10.1371/journal.pone.0124633
[5] da Silva AJ, Bornay-Llinares FJ, Moura INS, Slemenda SB, Tuttle TL, Pieniazek NJ. Fast and reliable extraction of protozoan parasite DNA from fecal specimens. Mol Diagn 1999;4:57-63.
[6] Ngom, B., Guo, Y., Wang, X. et al. Development and application of lateral flow test strip technology for detection of infectious agents and chemical contaminants: a review. Anal Bioanal Chem 397, 1113–1135 (2010). https://doi.org/10.1007/s00216-010-3661-4
[7] Miyamoto-Shinohara Y, Nozawa F, Sukenobe J, Imaizumi T. Survival of yeasts stored after freeze-drying or liquid-drying [Internet]. The Journal of General and Applied Microbiology. Applied Microbiology, Molecular and Cellular Biosciences Research Foundation; 2010 [cited 2022Oct9]. Available from: https://www.jstage.jst.go.jp/article/jgam/56/2/56_2_107/_article