Team:UI Indonesia/Engineering

ENGINEERING



Engineering is a cycle that never ends as we always learn to achieve better results. We start from learning the literature and design our system, model the vital component and design experiments accordingly. Due to the COVID-19 pandemic we have no access to the laboratory, hence we try to as clearly as possible explain our experiment design. Experiment is not the end, we also evaluate what outcomes that we may get and how to deal with the results for further improvement. Here we detail our System Design and Experimental Design.


In this section we are explaining our system design, the reason behind our constructed design. There are three systems, two of which are engineered: biofilm dispersion and H. pylori eradication, the other is naturally presented: chemotaxis.



We do not modify the natural chemotaxis pathway to AI-2 of our ECN. AI-2 is the only known quorum sensing signal secreted by H. pylori and also a chemoattractant signal for E. coli. Utilizing these features, our ECN hopefully will chase the H. pylori. Furthermore our product is for therapeutic purposes, so aggressively searching and chasing H. pylori through modification of chemotaxis is not necessary.


Bacteria are equipped by the ability to do chemotaxis, moving toward a specified chemoattractant within the frame of concentration gradient. In chemotaxis there are 3 main components: chemoattractant, chemoreceptor, and efector. Specific chemoattractant has a specific chemoreceptor. However, the effector is similar: flagella. Bacteria move with the assistance of flagella as the locomotion organelle.1

Bacteria, E. coli in particular, have two modes of locomotion. Running is biased movement towards a gradient by counterclockwise (CCW) rotation of flagellar motors. Clockwise (CW) rotation of one or many flagella facilitates tumbling or random direction movements of the organism. Direction is determined by temporal comparison of chemicals and modification of flagellar motor rotation to move towards or away from the gradient. When the organism is running down a chemorepellent gradient, tumbling will activate to redirect randomly until a direction suitably opposite of the gradient has been achieved.2

Environmental signals are detected by transmembrane chemoreceptors acting by Che signalling proteins. There are five chemotactic chemoreceptors in E. coli, Tsr, Tar, Tap, Trg, and Aer.3 CheA autophosphorylates to become the phosphate donor for CheY and CheB autokinase activities. Phosphorylated CheY (P-CheY) facilitates counterclockwise rotation of motors while phosphorylated CheB (P-CheB) has methylesterase activity for the chemoreceptors, except Aer. Phosphorylation of these two proteins are short-lived as autohydrolysis and CheZ for CheY dephosphorylates them. CheR has methyltransferase activity for chemoreceptor-CheW-CheA signalling complexes.

CheA autophosphorylation is also coupled to the chemoreceptors by CheW to form a signalling complex whose state determines the effect it has on flagellar motors. Methylation (by proxy CheR) and chemorepellents switch the kinase activity of this complex on while chemoattractants and demethylation (by proxy CheB) switch the kinase activity off. The OFF and ON state of this complex drives CCW and CW rotation of motors respectively. Therefore, the states of the signalling complexes determine the direction of movement.2

This ability also plays a role in the quorum sensing (QS), a cell to cell communication process within the population of bacteria (intra or interspecies) that affect numerous gene regulation which further alter the growth, metabolism, secretion system, biofilm formation, etc. The QS fluctuates depending on the organisms’ density which also reflects the QS signal molecules concentration. So, whenever the molecules reach a certain threshold, specific genes are expressed as a response. Numerous studies show convincing evidence of the presence of chemotaxis toward QS signals.2,4,5


In E. coli, the only known quorum sensing signal is autoinducer 2 (AI-2), a self produced molecule derived from 4,5-dihydroxy-2,3-pentanedione (DPD) cyclication. DPD is made from S-ribosylhomocysteine by the enzyme LuxS.5 The Lsr system, which is important for AI-2 quorum sensing in E. coli, is composed of eight genes (lsrKRACDBFG) encoding an ATP-binding cassette (ABC) transporter. A general schematic of the relevant components is as such:6
  • LsrA, LsrC, LsrD, and LsrB comprises the Lsr transport protein; LsrB is the substrate binding protein that internalises extracellular AI-2
  • LsrK phosphorylates AI-2 and is required to regulate AI-2 uptake
  • LsrR is a repressor of the lsr operon that is inhibited by phosphorylated AI-2 (P-AI-2). This step activates the expression of the rest of the Lsr quorum sensing system
  • LsrG isomerises P-AI-2 into phosphorylated 3-hydroxy-2,4-pentadione-5 (P-HPD)
  • LsrF is a thiolase to transfer an acetyl group from P-HPD onto another substrate

Figure 2 provides a visual representation of the above described. It is known that AI-2 is mediated by Tsr, but Tsr does not directly bind with AI-2. It is hypothesized that AI-2 bound LsrB assumes a conformation that allows it to interact with Tsr directly in the periplasm.5

Figure 2. Lsr system components interact with each other to produce chemotaxis. DPD: 4,5-dihydroxy-2,3-pentanedione.7


The H. pylori and AI-2 relationship is unique compared to other bacteria that perceive the molecules as quorum sensing signals. In H. pylori, AI-2 which is also produced in a LuxS-dependent manner is perceived as chemorepellent.8,9 When AI-2 is detected in a higher concentration, less bacteria is found. In the biofilm, this phenomenon is also observed. High AI-2 is correlated with a more sparse distribution of bacteria in the biofilm. Furthermore, a significant high amount of AI-2 promotes biofilm dispersion in H. pylori.

There is a negative correlation between AI-2 concentration and H. pylori density, which may imply that in H. pylori biofilm, there is a lower amount of AI-2. This finding is also supported by the higher biofilm amount in LuxS mutated strain. However the results were also conflicting since one of the most potent biofilm forming strains H. pylori, TK1402, was used as the control. Furthermore, the mutated LuxS strain (lower AI-2 production) is an extreme condition, while there are various strains of H. pylori with different potential of biofilm formation which are yet to be meticulously studied.8–10


Actually, we are not modifying our bacteria for chemotaxis purposes. Instead, we are taking advantage of the native nature of E. coli (and E. coli Nissle 1917 in our project) to chase AI-2. Fortunately, the only known quorum sensing signal in H. pylori is also AI-2. Although no straight forward studies evaluate the chemotaxis of E. coli towards H. pylori, based on the available evidence regarding the E. coli chemotaxis to AI-2 the chemotaxis system is plausible.

We are aware that some limitations may arise when using the AI-2 as our E. coli’s chemoattractant. However until the moment of writing no updated studies have found any secreted molecules that are specific for H. pylori. We are also aware that H. pylori excretes ammonia as it converts urea to ammonia and carbon dioxide, and detecting ammonia is possible. However, we are using the ammonia to induce another system that is more critical to be inducible. Making the bacteria swim and secreting the biofilm disperse agent, Proteinase-K, and antimicrobial peptide, PGLa-AM1, is exhaustive and ineffective.

As this is a therapeutic approach, we are also considering the reality that most patients, especially in Indonesia, come to a health care provider when a debilitating symptom has occurred. Which means the disease has progressed. In the context of H. pylori, the infection is abundant. High levels of AI-2 in clinical context may be perceived as a widespread biofilm formation and advanced disease with a critical amount of bacteria. Our project is subjected to cure, so it is administered after the diagnosis (which means the H. pylori definitely exists).


  1. Wadhams GH, Armitage JP. Making sense of it all: bacterial chemotaxis. Nat Rev Mol Cell Biol. 2004 Dec;5(12):1024–37.
  2. Laganenka L, Colin R, Sourjik V. Chemotaxis towards autoinducer 2 mediates autoaggregation in Escherichia coli. Nat Commun. 2016 Dec 23;7(1):12984.
  3. Baker MD, Wolanin PM, Stock JB. Signal transduction in bacterial chemotaxis. BioEssays News Rev Mol Cell Dev Biol. 2006 Jan;28(1):9–22.
  4. Holm A, Vikström E. Quorum sensing communication between bacteria and human cells: signals, targets, and functions. Front Plant Sci. 2014;5:309.
  5. Hegde M, Englert DL, Schrock S, Cohn WB, Vogt C, Wood TK, et al. Chemotaxis to the Quorum-Sensing Signal AI-2 Requires the Tsr Chemoreceptor and the Periplasmic LsrB AI-2-Binding Protein. J Bacteriol. 2011 Feb 1;193(3):768–73.
  6. Zuo J, Yin H, Hu J, Miao J, Chen Z, Qi K, et al. Lsr operon is associated with AI-2 transfer and pathogenicity in avian pathogenic Escherichia coli. Vet Res. 2019 Dec 12;50(1):109.
  7. Li J, Attila C, Wang L, Wood TK, Valdes JJ, Bentley WE. Quorum Sensing in Escherichia coli Is Signaled by AI-2/LsrR: Effects on Small RNA and Biofilm Architecture. J Bacteriol. 2007 Aug;189(16):6011–20.
  8. Rader BA, Wreden C, Hicks KG, Sweeney EG, Ottemann KM, Guillemin K. Helicobacter pylori perceives the quorum-sensing molecule AI-2 as a chemorepellent via the chemoreceptor TlpB. Microbiology. 2011 Sep 1;157(9):2445–55.
  9. Anderson JK, Huang JY, Wreden C, Sweeney EG, Goers J, Remington SJ, et al. Chemorepulsion from the Quorum Signal Autoinducer-2 Promotes Helicobacter pylori Biofilm Dispersal. mBio. 6(4):e00379-15.
  10. Cole SP, Harwood J, Lee R, She R, Guiney DG. Characterization of Monospecies Biofilm Formation by Helicobacter pylori. J Bacteriol. 2004 May;186(10):3124–32.



We are aiming to disperse the biofilm of H. pylori which is mainly composed of protein using Proteinase-K. The synthesis of Proteinase-K is under the control of arabinose-dependent pBAD promoter. The secretion is done through autolysis by E7 lysis protein which is triggered by ammonium sensing via glnAp2 promoter.


Bacteria often form biofilms to protect themselves from hostile environmental influences. Biofilms are made of immobile colonized microbes that are “submerged” by extracellular matrix (ECM): exopolysaccharides (EPS), proteins, extracellular DNA (e-DNA).1 The biofilm formation processes consist of four major steps: (1) bacteria attachment/adhesion to surface (from reversible to irreversible), (2) microcolony formation, (3) matrix formation, and (4) biofilm maturation. For the last step of the biofilm life cycle, the dispersion happens.1–3 Biofilm formation is closely related to bacterial quorum sensing (QS) via the QS signal as more AI-2 is produced with the increment of cell density. While many bacterias’ biofilms are mainly composed of EPS, proteins play a central role in the biofilms formation of H. pylori. The research in H. pylori’s biofilm itself is relatively new.4,5


Biofilm contributes to around 80% of chronic or recurrent bacterial infection.1 However, it is important to note that different populations of bacteria are found in the biofilm compared to planktonic ones genotypically or phenotypically in several aspects: metabolism, growth, resistance, etc. H. pylori infection is one of many chronic infections as almost 50% people have this bacteria throughout their life.6 And fortunately, it has the ability to form biofilm in the environment, on human gastric mucosa, as well as on in vitro abiotic surfaces.7 In Indonesia itself, Fauzia et al showed 93% of the H. pylori strain is biofilm forming. Biofilms offer increased resistance (even up to 1000 times !) to antimicrobial agents in addition to tolerating the host immune system and other environmental stresses.1 Studies in Indonesia showed up to 46.7% of metronidazole, 5.2% of amoxicillin, 9.1% clarithromycin, and 31.2% of levofloxacin resistance genotypically while phenotypically 28.6% of metronidazole, 9.5% of amoxicillin and clarithromycin, and 47.6% of levofloxacin resistance.8,9 However, since Indonesia is a plural and diverse country, the resistance pattern across different socio demographic populations should be noted. Dispersing the biofilm will expose the “hiding” bacteria to antimicrobial, in our system is antimicrobial peptide (AMP), and effectively kill the H. pylori.


Proteinase K is a 278 amino acids long serine protease that is naturally synthesized by the fungi Tritirachium album. It is known for its potent, wide pH optimum, and low peptide bond specificity.10 The active site is composed of catalytic triad: Asp39, His 69, and Ser224. While substrate recognition sites are amino acids number 99-104 and 132-136. Classically, its optimum activity is observed in the pH of 7-12 (but can reach as low as 3) and in the presence of Ca2+. The enzymatic activity is different depending on environmental conditions. However, its specific activity of hydrolyzes 1 µmole Ac-Tyr-Oet per min at 30oC and pH 9.3 is approximately 300 U/mg.11–13 In an optimal condition and concentration at least 1 mg/mL, Proteinase K can maintain its activity up to 48 hours without any observed reduction. In a much lower concentration of 0.01 mg/dL, it undergoes autolysis.14 There is limited information regarding its activity, half life, and degradation rate in other specified conditions.


As previously stated, H. pylori biofilm is mainly composed of protein which makes it susceptible to protease. Considering the versatility of Proteinase-K from its functional parameters such as optimum pH (as low as 3) and temperature (37oC)11, it is possible to use this enzyme in the human stomach environment.11 Windham et al showed that proteinase K is able to disperse H. pylori biofilm dramatically with only 25 µg/ml of concentration even in the first 24 hours. The other enzyme, DNase, and sodium periodate do not has significant effect on H. pylori biofilms.5 We are also aware that gastric pH in normal condition is extremely acidic with pH can reach as low as 1 during fasting. However, some studies show that in H. pylori infected stomach, there is a pH increment and hypochlorhydria.16,17 Furthermore to overcome the issue of extremely low gastric pH, post-prandial consumption is prefered.


We construct our biofilm dispersion system based on several components: induction and secretion process.


Induction
The synthesis of proteinase-K is done in an inducible manner. Constitutive synthesis is counterproductive as we are projecting our product to be delivered in the form of dairy products and a continuous production will burden the system and pose significant problems in packaging and storage. Some things to be considered in the induction system are listed below:
  1. the Proteinase-K is needed to be produced and worked before the AMP secretion
  2. the Proteinase-K production is inducible and not autoinduced
  3. the Proteinase-K has to achieve a targeted optimal concentration for biofilm dispersion before being administered

Acknowledge the aforementioned list regarding the induction system, we choose to use arabinose-induced pBAD promoter production. This is not based merely on the commonly used inducible promoter. We are projecting our product to be available in dairy product drinks, so it is important that our inducer is safe and well tolerated. Arabinose has no observed negative health effects and is even known to improve glycemic control.18,19 Further packaging design will be discussed in the Implementation section.

Secretion
After the synthesis, the next step is secretion. Due to the periplasmic subcellular location of Proteinase-K20, we design the secretion system via autolysis of our E. coli so we are sure that the Proteinase-K is secreted completely. The E7 lysis protein is chosen not only because of its popularity among iGEM-ers, but the latest study shows the effectiveness of E7 lysis protein in ECN for secreting protease.21 Since we have designed that the Proteinase-K optimal concentration has been achieved, it can be released immediately as it reach the gaster which is occupied by H. pylori. The urease activity of H. pylori converts urea to ammonia. So we find a significantly higher amount of ammonia and sense it via ammonia sensing promoter glnAp2.

Here in Table 1. We are listing the summary of our considerations in constructing the Proteinase-K system.

Table 1. Biofilm dispersion system component and considerations
Component Considerations Proposed system
Induction
  • produce before AMP
  • induced production, no autoinduction
  • targeted optimal concentration pre administration
 arabinose induction
Secretion
  • periplasmic (modelling prediction)
  • release immediately after reaching the stomach
  • ammonia induction
  • e7 lysis protein


We construct a system as follow
  1. The synthesis of Proteinase-K starts if our ECN senses arabinose by its pBAD promoter
    2. Processes
    • AraC is produced constitutively in control of constritutive promoter which then repressed the pBAD promoter
    • In the presence of arabinose, AraC is inactivated and no pBAD repression and Proteinase-K is produced
  2. The secretion of Proteinase-K via E7 lysis protein starts if our ECN sense high ammonium level
    3. Processes
    • In low ammonia level, activated NtrC induce the glnAp2 promoter to start TetR synthesis
    • TetR represses the pTet which results in repression of E7 synthesis
    • In high ammonia level, NtrC is inactivated, so no TetR is produced to repress pTet
    • Unrepressed pTet promotes transcription of E7 and lysis is occured

Figure 5. Biofilm dispersion system


  1. Sharma D, Misba L, Khan AU. Antibiotics versus biofilm: an emerging battleground in microbial communities. Antimicrob Resist Infect Control. 2019 May 16;8(1):76.
  2. Rabin N, Zheng Y, Opoku-Temeng C, Du Y, Bonsu E, Sintim HO. Biofilm formation mechanisms and targets for developing antibiofilm agents. Future Med Chem. 2015 Mar;7(4):493–512.
  3. Banerjee D, Shivapriya PM, Gautam PK, Misra K, Sahoo AK, Samanta SK. A Review on Basic Biology of Bacterial Biofilm Infections and Their Treatments by Nanotechnology-Based Approaches. Proc Natl Acad Sci India Sect B Biol Sci. 2020 Jun 1;90(2):243–59.
  4. Hathroubi S, Servetas SL, Windham I, Merrell DS, Ottemann KM. Helicobacter pylori Biofilm Formation and Its Potential Role in Pathogenesis. Microbiol Mol Biol Rev [Internet]. 2018 Jun [cited 2021 Aug 23];82(2). Available from: https://journals.asm.org/doi/10.1128/MMBR.00001-18
  5. Windham IH, Servetas SL, Whitmire JM, Pletzer D, Hancock REW, Merrell DS. Helicobacter pylori Biofilm Formation Is Differentially Affected by Common Culture Conditions, and Proteins Play a Central Role in the Biofilm Matrix. Dudley EG, editor. Appl Environ Microbiol [Internet]. 2018 Jul 15 [cited 2021 Aug 26];84(14). Available from: https://journals.asm.org/doi/10.1128/AEM.00391-18
  6. Hooi JKY, Lai WY, Ng WK, Suen MMY, Underwood FE, Tanyingoh D, et al. Global Prevalence of Helicobacter pylori Infection: Systematic Review and Meta-Analysis. Gastroenterology. 2017 Aug;153(2):420–9.
  7. Yonezawa H, Osaki T, Kamiya S. Biofilm Formation by Helicobacter pylori and Its Involvement for Antibiotic Resistance. BioMed Res Int. 2015;2015:914791.
  8. Miftahussurur M, Syam AF, Nusi IA, Makmun D, Waskito LA, Zein LH, et al. Surveillance of Helicobacter pylori Antibiotic Susceptibility in Indonesia: Different Resistance Types among Regions and with Novel Genetic Mutations. PLoS ONE. 2016 Dec 1;11(12):e0166199.
  9. Fauzia KA, Miftahussurur M, Syam AF, Waskito LA, Doohan D, Rezkitha YAA, et al. Biofilm Formation and Antibiotic Resistance Phenotype of Helicobacter pylori Clinical Isolates. Toxins. 2020 Jul 24;12(8):473.
  10. Butler GH, Kotani H, Kong L, Frick M, Evancho S, Stanbridge EJ, et al. Identification and characterization of proteinase K-resistant proteins in members of the class Mollicutes. Infect Immun. 1991 Mar;59(3):1037–42.
  11. BRENDA - Information on EC 3.4.21.64 - peptidase K [Internet]. [cited 2021 Sep 1]. Available from: https://www.brenda-enzymes.org/enzyme.php?ecno=3.4.21.64#SYNONYM
  12. Saenger W. Proteinase K. In: Handbook of Proteolytic Enzymes [Internet]. Elsevier; 2013 [cited 2021 Aug 29]. p. 3240–2. Available from: https://linkinghub.elsevier.com/retrieve/pii/B9780123822192007146
  13. Bajorath J, Hinrichs W, Saenger W. The enzymatic activity of proteinase K is controlled by calcium. Eur J Biochem. 1988;176(2):441–7
  14. Bajorath J, Saenger W, Pal GP. Autolysis and inhibition of proteinase K, a subtilisin-related serine proteinase isolated from the fungus Tritirachium album Limber. Biochim Biophys Acta BBA - Protein Struct Mol Enzymol. 1988 Jan;954:176–82.
  15. Waterhouse A, Bertoni M, Bienert S, Studer G, Tauriello G, Gumienny R, et al. SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic Acids Res. 2018 Jul 2;46(W1):W296–303.
  16. Seo J-H, Park HK, Park JS, Yeom JS, Lim J-Y, Park C-H, et al. Association between Gastric pH and Helicobacter pylori Infection in Children. Pediatr Gastroenterol Hepatol Nutr. 2015 Dec;18(4):246–52.
  17. Furuta T, Baba S, Takashima M, Futami H, Arai H, Kajimura M, et al. Effect of Helicobacter pylori infection on gastric juice pH. Scand J Gastroenterol. 1998 Apr;33(4):357–63.
  18. Hao L, Lu X, Sun M, Li K, Shen L, Wu T. Protective effects of L-arabinose in high-carbohydrate, high-fat diet-induced metabolic syndrome in rats. Food Nutr Res. 2015 Dec 10;59:10.3402/fnr.v59.28886.
  19. Krog-Mikkelsen I, Hels O, Tetens I, Holst JJ, Andersen JR, Bukhave K. The effects of l-arabinose on intestinal sucrase activity: dose-response studies in vitro and in humans. Am J Clin Nutr. 2011 Aug 1;94(2):472–8.
  20. Gunkel FA, Gassen HG. Proteinase K from Tritirachium album Limber. Characterization of the chromosomal gene and expression of the cDNA in Escherichia coli. Eur J Biochem. 1989 Jan 15;179(1):185–94.
  21. Hwang IY, Koh E, Wong A, March JC, Bentley WE, Lee YS, et al. Engineered probiotic Escherichia coli can eliminate and prevent Pseudomonas aeruginosa gut infection in animal models. Nat Commun. 2017 Apr 11;8(1):15028.



We are designing a system to eradicate H. pylori by a proven and potent antimicrobial peptide (AMP) against H. pylori, PGLa-AM1. Since PGLa-AM1 is also toxic for our E. coli as the host, we construct the AMP in an inactivated form as fusion AMP (FAMP) by linking it to SUMO protein. In a time-controlled system, our SUMO-AMP is then cleavage by Ulp protease and the PGLa-AM1 is activated. Subsequently, the holin-antiholin system is also activated and lysis the E. coli to secrete our activated PGLa-AM1.


Aside from the antibiotic resistance related to biofilm, the intrinsic antibiotic resistance of H. pylori also raises some concerns. A meta-analysis of 66,142 isolates showed as much as ≥15% resistance to clarithromycin, metronidazol, and levofloxacin in all WHO regions. The exception is clarithromycin resistance in America (10%) and South-East Asia (10%), and levofloxacin resistance in Europe (11%). Furthermore, treatment failure especially in clarithromycin-based regiment increases up to 7 times higher in clarithromycin-resistant H. pylori.1 Considering the local resistance in Indonesia, there are 46.7%, 31.2%, 9.1%, 5.2%, and 2.6% resistance for metronidazole, levofloxacin, clarithromycin, amoxicillin, and tetracycline respectively.2


The AMP is a relatively new interest in the realm of antibiotic treatment, yet their presence basically are far long before the invention of antibiotics as most AMPs are naturally found in microorganisms (fungi, bacteria), insects, amphibians, to mammals as part of the immune system. The activity can be classified into 18 categories ranging from antibacterial to anti-tumor.3,4

Structurally, AMPs are short chain peptides (10-50s amino acids long) which have a positive net charge and have significant hydrophobic residues. Most AMPs are in α-helical and β-sheet conformation, while the rest are peptides with random-coil/extended structure. The random coil peptides lack secondary structure and have a high arginine, proline, histidine, glycine or tryptophan residues. 3,4

As antibacterial, AMPs work by targeting the bacterial membrane or non-membrane. As we will use the previous mechanism in our project it is important to notice the mechanism of membrane disruption. There are 2 main models: (1) pole (toroidal and barrel stave) model and (2) carpet model. Carpet model destroys cell membrane by acting like a detergent, while pore model forms a hole or channel in the cell membrane.3,4


Figure 6. Membrane disruption model of AMP mechanism of action4



PGLa-AM1 is composed of 22 amino acids (2.07 kDa), and was initially found expressed in Xenopus amieti skin secretions. This AMP acts through carpet model membrane disruption.5 and has shown very high minimum inhibitory concentrations for S. aureus (103 μg/mL) and E. coli (51 μg/mL).2 Additionally, derived proteins show significant antimicrobial activity even against multidrug resistant moieties of several species.3 Tested against our target pathogen, the MIC was 1 μg/mL, four times less than metronidazole, a standard antibiotic against gastric H. pylori infection. Further kinetics testing reinforced PGLa-AM1 as a promising antibacterial agent against H. pylori. As a cationic amphipathic α-helical peptide, PGLa-AM1 competitively replaces divalent cations in the cell membrane, which encourages membrane permeability. Adoption of amphipathic configurations by hydrophobic interactions will further increase membrane permeability. These increases then cause fatal damage to the target pathogen.5 The AMP works in time and concentration dependent manner. In vivo murine study showed 20%, 60%, and 100% clearance of H. pylori resides in the stomach over 7 days of 10, 20, and 50 mg/kg BW. Moreover, its insulin stimulating effect is another beneficial properties of PGLa-AM1.6
Figure 7. PGLa-AM1 3D structure (Expassy-Swiss Model)



We construct our H. pylori eradication system based on several components: induction, activation, and secretion process.


Induction
First of all, synthesis of an active AMP inside a bacteria will negatively affect the host (our E. coli) as PGLa-AM1 is also toxic to E. coli. Although the MIC for E. coli is extremely higher than for H. pylori (51.76 So there is a huge chance of our synthesized AMP causing the host to self-terminate. Unfortunately, if that happens we are unable to control the production rate or concentration and at the end we are unable to kill the H. pylori (instead just killing our E. coli). So, we are designing the AMP to be inactivated by a well known fusion system using SUMO peptide which incidentally also increases its solubility. The Peptidor project from iGEM TU-Delft 2013 inspired us a lot in developing this system.7

There are two main things to be considered for designing our induction system
  1. This system must work after the biofilm dispersion happens
  2. The production of FAMP is induced by the presence of H. pylori
  3. The production of FAMP is inducible and not autoinduced

To remind you, our Proteinase-K is being secreted as the E. coli detect ammonia - H. pylori is producing ammonia as a product of urea metabolism by urease -. It is not possible to use AI-2 (although it is widely used by several iGEM teams) as an inducer due to the fact that the E. coli is naturally secreting the same molecule (except if LuxS mutation is done).8 This will lead to a premature synthesis of FAMP because of the auto inducible production. Only ammonia can be used as a signal in this setting. We choose the ammonia inducible promoter (previously known as copper sensing promoter), pCopA.9

Activation
As we use ammonia sensing in the Proteinase-K secretion and FAMP production system, these will happen at the same time. To buy enough time for the Proteinase-K to disperse a significant amount of biofilm and to produce enough FAMP (targeted concentration), the timer cassette is set. To activate our FAMP: SUMO-PGLa-AM1, we are adapting the Peptidor system.7 The system involves double repressible promoters (pTet and pcI), aiming to have enough time for FAMP production. The last component of the activation system is to cleave the FAMP using Ulp-1 protease which acts specifically as SUMO protease.

Secretion
The last step is absolutely secreting the activated AMP. Still, we are benchmarking the Peptidor project as their mathematical model is quite well developed for adaptation. The holin-antiholin system is used for our kill switch mechanism, controlled under the pcI. The lysis of our bacteria is not only for secretion purposes, but also to make sure that our engineered bacteria, although served as a probiotic strain that is widely accepted as safe species, does not pose further risk of mutation and become pathogenic. Here in Table 2. we are listing the summary of our considerations in constructing the PGLa-AM1 system.

Table 2. H. pylori eradication system component and considerations
Component Considerations Proposed system
Induction
  • work after biofilm dispersion
  • induced by the existence of H. pylori (producing ammonia through urease)
 ammonia induction
Activation
  • inactive before being secreted
  • targeted concentration before activation
  • buying time for biofilm dispersion
  • Fusion AMP (SUMO-AMP)
  • Ulp-1 protease
  • Timer cassette
Secretion safety profile Lysis protein holin-antiholin system


We construct a system as follow
  1. The synthesis of SUMO-PGLa-AMI starts if our ECN senses the presence of ammonia
    2. Processes
    • Ammonium ions produced by H. pylori urease
    • NH4+ induces pCopA, expressing SUMO-PGLa-AM1 and TetR genes downstream
  2. The activation of PGLa-AM1 via cleavage of SUMO-PGLa-AM1 by the Ulp-1 protease that is controlled by a TetR-based timer cassette
    3. Processes
    • TetR represses pTet further downstream
    • cI expression, which was under pTet, is inhibited
    • After a time constant (see Modelling), extant cI to degrades
    • In turn, cI repression of pcI is inhibited
    • Ulp-1 is expressed and cleaves the SUMO from present SUMO-PGLa-AM1
    • PGLa-AM1 attains a biologically active structure without SUMOylated terminal
  3. The secretion of activated PGLa-AM1 via holin-antiholin lysis system
    4. Processes
    • pcI activates via cI (repressor) degradation
    • Holin and endolysin are expressed
    • Holin forms pores in the cell membrane
    • Extant antiholin, constitutively expressed, halts holin action to provide another time delay for PGLa-AM1 accumulation
    • Endolysin and other intracellular enzymes degrade the cell wall
    • Host is lysed and cell contents released into environment
    • Cell contents including activated PGLa-AM1 which also further enhances the kill switch function


Figure 8. H. pylori eradication system


  1. Savoldi A, Carrara E, Graham DY, Conti M, Tacconelli E. Prevalence of Antibiotic Resistance in Helicobacter pylori: A Systematic Review and Meta-analysis in World Health Organization Regions. Gastroenterology. 2018 Nov;155(5):1372-1382.e17.
  2. Miftahussurur M, Syam AF, Nusi IA, Makmun D, Waskito LA, Zein LH, et al. Surveillance of Helicobacter pylori Antibiotic Susceptibility in Indonesia: Different Resistance Types among Regions and with Novel Genetic Mutations. PLoS ONE. 2016 Dec 1;11(12):e0166199.
  3. Mahlapuu M, Håkansson J, Ringstad L, Björn C. Antimicrobial Peptides: An Emerging Category of Therapeutic Agents. Front Cell Infect Microbiol. 2016;6:194.
  4. Huan Y, Kong Q, Mou H, Yi H. Antimicrobial Peptides: Classification, Design, Application and Research Progress in Multiple Fields. Front Microbiol. 2020;11:2559.
  5. Zhang X, Jiang A, Wang G, Yu H, Qi B, Xiong Y, et al. Fusion expression of the PGLa-AM1 with native structure and evaluation of its anti-Helicobacter pylori activity. Appl Microbiol Biotechnol. 2017 Jul;101(14):5667–75.
  6. Neshani A, Zare H, Akbari Eidgahi MR, Hooshyar Chichaklu A, Movaqar A, Ghazvini K. Review of antimicrobial peptides with anti- Helicobacter pylori activity. Helicobacter. 2019 Feb;24(1):e12555.
  7. Team:TU-Delft - 2013.igem.org [Internet]. [cited 2021 Aug 26]. Available from: https://2013.igem.org/Team:TU-Delft
  8. Pereira CS, Thompson JA, Xavier KB. AI-2-mediated signalling in bacteria. FEMS Microbiol Rev. 2013 Mar 1;37(2):156–81.
  9. Shen P, Niu D, Permaul K, Tian K, Singh S, Wang Z. Exploitation of ammonia-inducible promoters for enzyme overexpression in Bacillus licheniformis. J Ind Microbiol Biotechnol [Internet]. 2021 Jun 1 [cited 2021 Sep 4];48(5–6). Available from: https://doi.org/10.1093/jimb/kuab037
  10. Part:BBa R0040 - parts.igem.org [Internet]. [cited 2021 Sep 6]. Available from: http://parts.igem.org/Part:BBa_R0040
  11. Part:BBa K1022123 - parts.igem.org [Internet]. [cited 2021 Sep 6]. Available from: http://parts.igem.org/wiki/index.php?title=Part:BBa_K1022123


DISCLAIMER!!! We are not doing the experiments due to lab restriction during the COVID-19 pandemic

However, in this section we are explaining the experiment design to test some functions in our systems. While conceptually, there are three systems: (1) chemotaxis, (2) biofilm dispersion by Proteinase-K, and (3) H. pylori eradication, we are designing the experiment only for the last two systems.


We are not doing experiments for chemotaxis function since it is postulated that E. coli naturally chase the AI-2 molecules and as mentioned earlier in the design section this approach is therapeutic which means patients have been diagnosed with H. pylori and high specificity is not necessary. However for further experiment, this function can be tested qualitatively through a simple approach using an agar plate system. Two wells are made in an agar medium (one for E. coli, one for AI-2 as the chemoattractant), separated 3 cm away, and connected via a channel. The AI-2 (200 μM) containing well is a positive control, for negative control only a 10% PBS buffer is used. For the treatment group, cultured H. pylori is inserted in a well. Since H. pylori and E. coli culture medium is different, splitting the plate for two agar plates can be done (H. pylori used columbia based agar, E. coli used LB agar). Our E. coli is transformed by unmodified pM2s2TsR plasmid which encodes RFP, so identification of chemotaxis can be done quantitatively. Observation is done every 30 minutes for 3 hours.


DISCLAIMER!!! We are not doing the experiments due to lab restriction during the COVID-19 pandemis. In this section we are explaining the procedure that will be applied to test the biofilm dispersion by Proteinase-K. The procedures are done as follows.


  • Plasmid assembly
    For experiment purposes, we split our parts into two separate inducible parts which will be further detailed in the next section. We are using Gibson assembly, thus our plasmid is arranged for Gibson's compatible. PCR is done before the Gibson assembly. The pM2s2TsR is used as the plasmid which contains red fluorescence protein (RFP) & tetracycline resistance genes. The inserted sequences are positioned replacing the RFP gene.
  • Transformation and cell competent selection
    E. coli culture, plasmid transformation by heat shock (or electroporation as alternative), and cell competent selection are done according to the protocol. Cells were grown in a LB agar, selected via double selection (color and tetracycline resistance). If insertion is successful, no color is observed. Confirmation of insertion will be done using Colony PCR. The used primers are shown below.
    Direction Sequence (5’-..-3’)( Tm (oC) Target Amplicon Length (bp)
    Forward TTCACACTGGCTCACCTTCG 60.25 Upstream junction (TT - pM2s2TsR) 137
    Reverse CCGTCTTCCGGGTACATACG 59.97


  • Aim: to assess how much Proteinase-K (concentration) produced in n-time after induction by arabinose
  • Experiment process
    Before testing the function of our ECN synthesized Proteinase-K against the H. pylori biofilm, identification of its presence is important. LB medium is used for expression. All experiments are performed in triplicate.
    1. Induction
      The synthesis of Prot-K is induced by 0.1 mM, 1 mM, and 10 mM of arabinose.
    2. Incubation
      Incubation time is set for (1) short period: 10, 20, and 30 minutes and (2) long period: 4, 8, 12, and 24 hours at 4 and 25oC.1
    3. Purification
      For purification purposes, 6x His-tag has been attached on the 3’-end of Proteinase-K coding sequences, then Ni-NTA purification is done.
    4. Detection and concentration count
      confirmation of the presence of Prot-K is done by SDS-PAGE. For concentration determination, concentration of purified Proteinase-K is quantified using a NanoDrop.
  • Analysis of outcomes
    The concentration of Proteinase-K in different environments are compared statistically using ANOVA and Post Hoc (normal distribution) or Kruskal-Wallis and Mann-Whitney (non-normal distribution) analysis. Three highest concentration groups from each temperature are selected for further analysis. Statistical analysis will be performed using SPSS 24.0.
  • Expected response after experiment
    1. concentration of produced Proteinase-K in specific incubation time after induction can be used as a parameter to adjust the number of bacteria to be administered so it achieves the optimal amount of Proteinase-K for dispersing biofilm which is known to be 25 μg/ml from literature.2
    2. data from numerous concentration of arabinose and its incubation time in correlation to the concentration of Prot-K produced is valuable to choose whether the product is more convenient to be consumed in a relatively short period of time after mixture (adding arabinose) i.e 10-30 minutes, or the patients need to mix it for the upcoming dose i.e 4-24 hours.
  • Parts and Plasmid construction
    • BBa_K3932006: araC/pBAD-Proteinase K
    • Plasmid: pM2s2TsR
    • NcoI restriction site (in mRFP1, located at 6,322 bp downstream)
    • Primer used for Gibson Assembly:
      Forward ccggttatgcagaaaaaaacTCACACTGGCTCACCTTC
      Reverse tcggtggaagcttcccaaccAAATAATAAAAAAGCCGGATTAATAATCTG


  • The H. pylori biofilm formation is prepared as the aforementioned protocol in liquid (quantitative) and agar medium (qualitative). All experiments are performed in triplicate. Treatment groups are prepared as follow:
    • Negative control: culture medium with H. pylori without Proteinase-K treatment
    • Positive control: culture medium with H. pylori added with Proteinase-K (Sigma Aldrich)
    • Treatment group: culture medium with H. pylori added with purified Proteinase-K treatment
    • Blank: culture medium without H. pylori
  • Identification of biofilm
    Biofilm identification is done using crystal violet staining. Observation of optical density is done on a wavelength of 590 nm (OD590) over 12, 24, and 48 hours, in 37oC using liquid medium. For agar medium, qualitative assessment is done. All experiments are performed in triplicate.


  • Aim: to assess how much the biofilm is dispersed after exposure to determined Proteinase-K concentration in specific time
  • Experiment process
    All experiments are performed in triplicate.
    1. Proteinase-K addition
      H. pylori biofilm stained by crystal violet is added by Proteinase-K in the form of
      • purified Proteinase-K from ECN expression (three highest concentration groups in each temperature)
      • Proteinase-K 25 μg/ml (Sigma) as positive control
    2. Activity assessment
      Observation is done in OD590 over 2, 4, 8, 24 and 48 hours in 37oC to assess the dispersion rate of biofilm.
    3. Activity after low pH exposure assessment
      The treatment group that gives the most significant OD reduction in the first 4 hours (because based on the modelling, our AMPs are secreted after 4 hours) is chosen for further activity assessment in different pH levels. The Proteinase-K undergoes 30 minutes of low pH exposure (1, 2, 3, and 4) using adjusted 2 mL HCl (this represents exposure of Prot-K in stomach lumen prior to actively working in the mucosa). Proteinase-K is separated from the HCl solution and retested for its activity.
  • Analysis of outcomes
    The outcome of OD will be plotted, compared between observation periods and incubation time of Prot-K expression, unpaired. ANOVA and Post Hoc (normal distribution) or Kruskal-Wallis and Mann-Whitney (non-normal distribution) analysis will be done. Statistical analysis will be performed using SPSS 24.0.
  • Expected response after experiment
    1. Residual activity is known in the environment exposed to low extreme pH and temperature of 37oC which are the physiological condition of human stomach
    2. Residual activity of Prot-K will affect the degradation of antimicrobial peptide (AMP) PGLa-AM1, which affect the bactericidal activity of AMP


  • Aim: to assess the time needed to lysis ECN via E7 expression after ammonium induction.
  • Experiment process
    As the E7 expression in ECN has been proved to effectively result in autolysis, here we assess how much time is needed to lysis the ECN via E7 expression that is under the control of ammonia sensing promoter glnAp2. All experiments are performed in triplicate. Treatment groups are prepared as follow
    • Negative control: LB culture medium without E7 lysis transformed ECN and without lysis protocol.
    • Positive control: LB culture medium without E7 lysis transformed ECN and with lysis protocol.
    • Treatment group: LB culture medium with E7 lysis transformed ECN and without lysis protocol induced by ammonium hydroxide 0.005% (w/v)3 0.01 fold, 0.1 fold, 10 fold, and 100 fold.
    • Blank: culture medium without ECN
    1. Induction
      The synthesis of E7 is induced by ammonium hydroxide 0.005% (w/v), 0.01 fold, 0.1 fold, 10 fold, and 100 fold concentration
    2. Incubation
      Incubation time is set for 30 minutes, 1, 2, 3, and 4 hours at 37oC
    3. Lysis Assessment
      Serial documentation of OD600 is done after respective incubation time. Standard curve is made using iGEM calibration protocol Conversion of OD600 to CFU/ml. OD600 then is converted to CFU/mL.
  • Analysis of Outcomes
    The outcome of CFU will be plotted, compared between incubation periods and induction concentration, unpaired. ANOVA and Post Hoc (normal distribution) or Kruskal-Wallis and Mann-Whitney (non-normal distribution) analysis will be done. Statistical analysis will be performed using SPSS 24.0.
  • Expected response after experiment:
    • optimal concentration of ammonium that triggers the highest lysis activity is an important information to determine how long the system will work and how fast is the rate of Prot-K secretion. Since arabinose may be “washed” from the stomach while Prot-K is relatively more stable and resilient in the human stomach, considering to synthesis enough amount of Prot-K since the beginning by adjusting the number of bacteria, amount of arabinose for induction, and in a relatively short time of incubation (consider the consumer convenient) before administration is important.
    • The concentration of ammonium surrounding the H. pylori is unknown. A study examined the urease activity from H. pylori. In an acidic environment, optimally at pH of 3, as much as 109 CFU/mL H. pylori are viable, the urease activity is documented as 204 μmol NH3/min/mg protein. If converted to w/v it is approximately 0.3% of ammonium.4
    • Considering the average amount of H.pylori in active infection reaches up to 106 CFU (in a certain biopsy location)5, the amount of ammonium produced by the urease is lower than 0.005% (baseline experiment concentration). But, several data shows that a lower concentration (0.02 fold of 0.005%) just needs an extra 30-60 minutes for product amount to rise significantly.
    • However, if even the highest concentration of ammonium (0.5% w/v) does not exhibit a satisfied response and produce enough E7, neither enable the release of Prot-K in desirable rate, addition of lysis protein or transcription enhancer may be needed. From another point of view, the low secretion rate of Prot-K may be beneficial as a model of extended release drugs.
    • Parts and Plasmids Construction
    • BBa_K3932007: glnAp2-TetR-pTet-E7
    • Plasmid pM2s2TsR
    • NcoI restriction site (in mRFP1, located at 6,322 bp downstream)
    • Primer used for Gibson Assembly:
      Forward ccggttatgcagaaaaaaacTAGATGCCTCCACACCGC
      Reverse tcggtggaagcttcccaaccAAATAATAAAAAAGCCGGATTAATAATCTGGC


DISCLAIMER!!! We are not doing the experiments due to lab restriction during the COVID-19 pandemis. In this section we are explaining the procedure that will be applied to test the H. pylori eradication by PGLa-AM1. The procedures are done as follows.


  • Plasmid Assembly
    For experiment purposes, we split our parts into three separate inducible parts which will be further detailed in the next section. We are using Gibson assembly, thus our plasmid is arranged for Gibson's compatible. PCR is done before the Gibson assembly. The pM2s2TsR is used as the plasmid RFP & Tetracycline Resistance Genes. The inserted sequences are positioned replacing the RFP gene.
  • Transformation and cell competent selection
    E. coli culture, plasmid transformation by heat shock (or electroporation as alternative), and cell competent selection are done according to the protocol. Cells were grown in a LB agar, selected via double selection (color and tetracycline resistance). If insertion is successful, no color is observed. Confirmation of insertion will be done using Colony PCR. The used primers are shown below.
    Direction Sequence (5’-..-3’)( Tm (oC) Target Amplicon Length (bp)
    Forward TTCACACTGGCTCACCTTCG 60.25 Upstream junction (TT - pM2s2TsR) 137
    Reverse CCGTCTTCCGGGTACATACG 59.97


  • Aim: to assess how much SUMO-AMP produced in n-time after induction by ammonium in different concentrations.
  • Experiment Process
    1. Induction
      The synthesis of SUMO-AMP is induced by ammonium hydroxide 0.005% (w/v), 0.01 fold, 0.1 fold, 10 fold, and 100 fold concentration
    2. Induction
      Incubation time is set for 2, 4, 6, and 8 hours at 37oC using a shaker incubator.
    3. Purification
      A 6x his-tag is attached to the 5’-end of SUMO-AMP structure, thus Ni-NTA purification is done.
    4. Detection and concentration count
      The solution is put at the SDS-PAGE for further characterization. Concentration of purified SUMO-AMP is measured using a NanoDrop.
  • Analysis of outcomes
    The concentration of SUMO-AMP in different environments are compared statistically using ANOVA and Post Hoc (normal distribution) or Kruskal-Wallis and Mann-Whitney (non-normal distribution) analysis. Three highest concentration groups are selected for further analysis. Statistical analysis will be performed using SPSS 24.0.
  • Expected response after experiment
    1. concentration of produced SUMO-AMP in a specific timestamp after induction in response to several different concentrations of ammonium can be used as a parameter to adjust the number of bacteria to be administered so it achieves the optimal amount of SUMO-AMP as the precursor of active AMP (PGLa-AM1).
      • PGLa-AM1 MIC from in vitro study is 1μg/ml
      • PGLa-AM1 optimal dose from in vivo (murine) study is 60 mg/kg
    2. The optimal response corresponding to the inducer (ammonium) concentration is valuable information to further reevaluate the inducer system. This also correlates with how much ammonia that is available to be sensed from H. pylori production. If higher concentrations are needed, an enhancer of transcription may be necessary. We discuss more detail regarding the ammonium concentrations in the real world (stomach mileu) below.

    Parts and Plasmid Construction
    • BBa_K3932008: pCopA-His6-SUMO-PGLaAM1
    • Plasmid pM2s2TsR
    • NcoI restriction site (in mRFP1, located at 6,322 bp downstream)
    • Primer used for Gibson Assembly:
      • Forward ccggttatgcagaaaaaaacCCTTTTTATAGATGCGGG
      Reverse tcggtggaagcttcccaaccAAATAATAAAAAAGCCGGATTAATAATC


  • Aim: (1) To determine how much Ulp1 is expressed in a given time after induction by ammonia, (2) to determine the time delay of timer cassette.
  • Experiment process
    1. Induction
      The synthesis of Ulp-1 is induced by ammonium hydroxide 0.005% (w/v), 0.01 fold, 0.1 fold, 10 fold, and 100 fold concentration.
    2. Incubation
      Incubation time is set for 1, 2, 3, and 4 hours at 37 oC using a shaker incubator.
    3. Purification
      For purification purposes 6x his-tag is attached to Ulp-1 protein, thus Ni-NTA purification is done.
    4. Detection and concentration count
      The solution is put at the Tricine SDS-PAGE for further characterization. Concentration of purified SUMO-AMP is measured using a NanoDrop.
  • Analysis of outcomes
    The concentration of Ulp-1 in different environments are compared statistically using ANOVA and Post Hoc (normal distribution) or Kruskal-Wallis and Mann-Whitney (non-normal distribution) analysis. Three highest concentration groups are selected for further analysis. Statistical analysis will be performed using SPSS 24.0.
  • Expected response after treatment:
    • The difference in incubation time would provide insights in Ulp-1 expressions over time. The different amount of expressed Ulp-1 is used to determine how many SUMO-AMP each can cleave.
      • 1/1000 ratio of Ulp1/SUMO can cleave SUMO tag 95% after 12 hours.
    • The amount of Ulp1 received is valuable as a parameter to measure the time needed to express sufficient Ulp-1 concentration. The time can be adjusted to express Ulp necessary for a sum of SUMO-AMP expressed at a specific time. The amount of expressed Ulp1 can be adjusted by its incubation time if necessary. This also will lead to adjustment of the timer cassette that can be shortened to produce enough Ulp-1 for SUMO-AMP cleavage.

Parts and Plasmid construction
  • BBa_K3932009: pCopA-TetR-pTet-cI-pcI-Ulp1
  • Plasmid pM2s2TsR
  • NcoI restriction site (in mRFP1, located at 6,322 bp downstream)
  • Primer used for Gibson Assembly:
    • Forward ccggttatgcagaaaaaaacCCTTTTTATAGATGCGGG
    Reverse tcggtggaagcttcccaaccAAATAATAAAAAAGCCGGATTAATAATC


  • Aim: to assess how much SUMO-AMPs are cleaved and PGLa-AM1s are activated in a correlated time.
  • We mix the expressed Ulp-1 and the expressed SUMO-AMP complex in a buffer (50 mM Tris/HCl, pH 8.0, 0.1M NaCl, 10mM DTT) at 25 oC, and 37 oC for 4, 6, 8, 12, and 24 hours. We calculate PGLa-AM1 concentration with NanoDrop. PGLa-AM1 is characterized and analyzed using Tricine SDS-PAGE with PageBlue staining.
  • Analysis of outcomes
    The concentration of PGLa-AM1 from paired SUMO-AMP and Ulp-1 (e.g both from 1 hour induction) are compared statistically using ANOVA and Post Hoc (normal distribution) or Kruskal-Wallis and Mann-Whitney (non-normal distribution) analysis. Three highest concentration groups are selected for further H. pylori eradication tests. Statistical analysis will be performed using SPSS 24.0.
  • Expected results after treatment:
    • The amount of successfully cleaved PGLa-AM1 is needed to measure the necessary expression of SUMO-AMP and corresponding Ulp1 over different time and temperature.
    • This test is also necessary to determine whether or not the amount of PGLa-AM1 produced is enough to kill a given amount of H. pylori at a given time of Ulp1 and SUMO-AMP expression. The amount of Ulp1 expression may be increased to cleave all SUMO-AMP complexes faster.


  • Aim: To measure the time it took for induced E. coli cells to lyse over time after the induction by ammonium.
  • Experiment process
      Here we assess how much time is needed to lysis the ECN via the holin-antiholin-endolysin (kill switch system) expression under the control of ammonia sensing promoter glnAp2. All experiments are performed in triplicate. Treatment groups are prepared as follow
    • Negative control: LB culture medium without kill switch parts being transformed ECN and without lysis protocol.
    • Positive control: LB culture medium without kill switch parts being transformed ECN and with lysis protocol.
    • Treatment group: LB culture medium with kill switch parts being transformed ECN and without lysis protocol induced by ammonium hydroxide 0.005% (w/v)3 0.01 fold, 0.1 fold, 10 fold, and 100 fold.
    • Blank: LB culture medium without ECN.
    1. Induction
      The kill switch system is activated by the induction using ammonium hydroxide 0.005% (w/v), 0.01 fold, 0.1 fold, 10 fold, and 100 fold concentration.
    2. Incubation
      Incubation time is set for 2, 4, 6, and 8 hours at 37 oC using a shaker incubator.
    3. Activity assessment
      Serial documentation of OD600 is done after respective incubation time. Conversion of OD600 to CFU/ml is done using iGEM calibration protocol.
  • Analysis of outcomes
    The OD in different environments are compared statistically using ANOVA and Post Hoc (normal distribution) or Kruskal-Wallis and Mann-Whitney (non-normal distribution) analysis. Statistical analysis will be performed using SPSS 24.0.
  • Expected results after treatment:
    • This test is necessary to know the time it takes for the E. coli to lyse itself. With this test, one can know if the necessary proteins (PGLa-AM1) are expressed enough before the lysis takes place. If the optimal amount of PGLa-AM1 is not achieved, synthesis intensification is needed using an enhancer or more bacteria is needed.
    • The kill switch promotor can be induced less or more, to adjust the AMP secretion rate.
Parts and Plasmid construction
  • BBa_K3932010: pCopA-TetR-pTet-cI-pcI-kill switch
  • Plasmid pM2s2TsR
  • NcoI restriction site (in mRFP1, located at 6,322 bp downstream)
  • Primer used for Gibson Assembly:
    • Forward ccggttatgcagaaaaaaacCCTTTTTATAGATGCGGGAGG
    Reverse tcggtggaagcttcccaaccGAGAGCGTTCACCGACAAAC


  • Growth is preferable to a system atmosphere of 10% CO2, 5% O2, and 85% N2 (Oxoid Campygen CN0025). We use H. pylori of strain ATCC 49503. We use two types of medium depending on the analysis:
    • Solid medium is prepared using columbian blood-based agar.
    • Liquid medium is prepared using brucella broth.
    • For non biofilm forming culture, 2 mg/mL N-acetylcysteine is added.
  • Expected results after treatment:
    • This protocol should give the idea of the best medium to culture with different analysis. Liquid medium, for example, is better used when we want to count cells via absorbance.
    • The results of this experiment should also provide us data on the viability of the chosen mediums to compare if a necessary junction appears to change mediums for a more optimised system.
    • This protocol also gives us the exact number of starting viable cells before we count them again after addition of PGLa-AM1.


  • Aims: to determine the bactericidal activity of produced PGLa-AM1 (concentration from the purification) to H. pylori
  • Experimental process
    Brucella broth with different concentrations of PGLa-AM1 (according to incubation time) is prepared in a microaerophilic environment at 37 oC. H. pylori cells are added to the various PGLa-AM1 concentrations and dilute them periodically at 0, 15, 30, 45, 60 minutes. The diluted cultures were incubated before measuring the number of viable CFU/mL.
  • Analysis of outcomes
    The OD in different PGLa-AM1 concentration treatment and dilution time are compared statistically using ANOVA and Post Hoc (normal distribution) or Kruskal-Wallis and Mann-Whitney (non-normal distribution) analysis. Statistical analysis will be performed using SPSS 24.0.
  • Expected response after treatment:
    • The number of viable H. pylori available over incubation time after mixing with PGLa-AM1 can give insights on the activity of different amounts of PGLa-AM1 to kill H. pylori. This parameter can be used to determine the necessary amount of PGLa-AM1 and the time needed to fully eradicate H. pylori. This is also important information about how many doses are needed to fully eradicate H. pylori since H. pylori keep reproducing while they got killed.
    • The number also correlates with the amount of E. coli cells, and its induction time needed to produce adequate amounts of PGLa-AM1. This can be achieved by increasing the number of E. coli cells, increasing the induction activity for protein expression, or lowering the kill-switch timer.
    • The resultant MIC gives us a baseline to adjust the PGLa-AM1 production by its expression and activation (in tandem with results from experiments described previously) as a starting point to tweak the final output of the system.
    • The bactericidal graph allows us to tune the final genes using results from previous experiments in order to produce the optimum amount of PGLa-AM1 by elucidating the relative effect over time the PGLa-AM1 production rate has on the target pathogen.
    • Similarly, it also allows us to optimize the system output to prevent toxicity to our host organisms (humans and delivery bacteria both) while still retaining satisfactory results.
    • Estimation of the relative effect per time will be correlated with production rate over time to produce a dose effect graph for humans for PGLa-AM1 which, while not directly useful to our project, is novel information for future projects within or without iGEM.


  • Aims: to determine the residual activity of produced PGLa-AM1 after exposure to Proteinase-K
  • Experimental process
    As Proteinase-K will cleavage our AMP, it is necessary to do an experiment of this cleavage and inactivation of PGLa-AM1by Proteinase-K. The optimal PGLa-AM1 concentration from the previous experiment and its parallel Proteinase-K production (from the same condition: ammonium concentration, temperature, and incubation time) are mixed and proceed to the PGLa-AM1 function test.
  • Expected response after treatment
    • The degree of which proteinase K affects PGLa-AM1 function given a certain time. This insight will allow us to assess the needed rates of production required to overcome proteinase K hydrolysis of our functional protein. This upper limit of proteinase K production and lower limit of PGLa-AM1 production will allow us maximise biofilm degradation whilst maintaining proper antibacterial function.
    • Conversely, we also need to determine the lower limit of proteinase K concentration to still provide effective biofilm degradation without significantly impairing PGLa-AM1 action. This would also allow us to determine the upper limit of PGLa-AM1 production before it can overcome proteinase K inhibition. This would allow us to precisely engineer a rate that produces exactly as much as needed as opposed to grossly overshooting the PGLa-AM1 production rate in an effort to blindly counter proteinase K action.



Through our literature searching and expert consultation, H. pylori culture is extremely challenging from its technical and methodological aspects as the bacteria itself is fastidious. In our institution we haven’t successfully done it until last year. The available studies that need culturing H. pylori in Indonesia are done in collaboration with Japan. To our knowledge, the research of H. pylori that can be done in Indonesia is still limited to genotype and molecular study without assessing its phenotype. However, this may also result because of biopsy and stomach isolate H. pylori strain. Here, we use the specialized standardized strain: ATCC 49503 which is known for its biofilm formation ability. Failure to generate biofilm can be a consideration to change the strain with a highly potent strain: TK402.

The formation of biofilm is not only affected by strain, but also medium used and its content. The microaerophilic condition is also important to be noted. When failure to grow the H. pylori culture, with or without biofilm formation the medium is needed to be reassessed. Several choices other than Columbia agar are Brucella broth, Trypticase soy agar/broth, brain-heart infusion (BHI), or campylobacter agar. Numerous supplements such as added with 5-10% sheep blood or 5-10% FBS may also be tested.


Upon transformation failure, as indicated by function or electrophoresis, four different scenarios were predicted and prepared for: 1) successful antibiotic selection with failed function tests, 2) complete antibiotic inhibition, 3) outgrowth failure, and 4) equipment failure. Scenario 1 involves a transformation failure, compromised antibiotic efficacy, and/or inappropriate organism response to antibiotic. Scenario 2 involves a fairly straightforward procedural review. Similarly, scenario 3 requires testing of the stock cells used. These variables are tested separately, in order of most suspicious, to determine points of failure. Once one or multiple points of failure are identified and corrected, experiments are redone. If experiments do not yield identification, then equipment is vetted for failure as the penultimate measure to complete protocol rewrite.


One of the most major possible obstacles is the expression of Proteinase-K as an inclusion body, its ability to non specifically digest membrane proteins of our engineered ECN, and thus affects the cell viability. When these happen, focus on the level of synthesis (limit the production of Prot-K) or fasten the autolysis through adding another lysis protein after the E7.

Failure in the proteinase K producing line may be the resultant of the host, the target or its testing environment. Transformation failure is discussed elsewhere whereas intrinsic failure of a successfully transformed system will first be assessed quantifiably and then corrected then. Problems with the testing targets or conditions will simply be verified and the targets replaced or experiment repeated upon clearance. Untestatable failures will get a consideration of faulty equipment before a case-by-case assessment is made upon it before protocol rewrite and/or reliance on secondary data.


We consider our function testing protocol to be separated into three different phases: initiation, production, and execution. We first test to see if our system is capable of producing the desired peptides under the desired conditions. If our initiation circuitry succeeds, only then we quantify the response produced. Finally, the system’s response is tested to see whether it is correct and significant. Failures are then also concentrated into these three categories. Initiation problems may rise from transformation problems (see 2. plasmid transformation) or errors within our models. Errors of production are similar to that of errors of initiation and our measures are similar. A weakened response in this stage will be mitigated by stronger and/or extended stimulation. Finally, inappropriate responses may arise from the same etiologies as the previous two or errors in testing medium. If our testing conditions cannot be met or are violated through the process, we will halt and correct them. Similar to our other protocol sets, equipment failure will be our penultimate review before a rewrite or difference on secondary data.




  1. The simulation of the ammonia-rich environment via the presence of H. pylori is needed to confirm the hypothesis that the amount of ammonia around H. pylori effectively induces the release of lysis protein.
  2. Although the Proteinase-K shows optimal biofilm dispersion in 25 μg/ml even after only 24 hours, this activity is observed in environmental conditions that differ compared to stomach mileu. We are unable to assess the residual activity of Proteinase-K in exposure of the most available enzyme in the stomach: pepsin. A model of the human gastric environment is needed to be constructed to test several variables that influence the Prot-K activity such pH and the presence of other proteases such as pepsin.
  3. Similarly, our experiments around PGLa-AM1 do not sufficiently replicate the human gastric environment and will require ideally the same model (as the two systems will be in practice) mentioned in 2.
  4. The experiment using clinical isolates of H. pylori is needed to further confirm the model using the “real” bacteria strain.
  5. In vivo study is also necessary to further test the appropriate dose for animals, before clinical trials.