Team:Groningen/Template/MODULE/project/MBD

From 2014.igem.org

(Difference between revisions)

Latest revision as of 02:59, 18 October 2014

Project > Model-based design

Why modeling?

Staphylococcus aureus and Pseudomonas aeruginosa are the two pathogens that cause infection in the burn wounds. The aim of our project was to design a smart bandage that can produce Infection Prevention Molecules (IPMs) only in the presence of these two pathogens. The bandage has freeze-dried Lactococcus lactis cells which are activated by hydrating the gel. The hydrated gel contains all the nutrient supplements that are required for the L. lactis to grow. The quorum molecules produced by the pathogens in the wound diffuse through the hydrogel and reach the L. lactis . This triggers the production of IPMs. The IPMs which are produced by the genetically engineered L. lactis in the hydrogel must diffuse to the site of infection in the burn wounds. Understanding the diffusion of IPMs is important for us to design a smart bandage which is quite realistic. Based on the bandage requirements there were different possible designs. Modeling those designs helped us to choose the best design and study the diffusion characteristics in our bandage. The best design was further analyzed to obtain information regarding the life time of the bandage after it is being activated. The bandage model gives an overall idea about production rate of IPMs, diffusion of the IPMs, inimum number of bacteria that should be available in the bandage and nutrient consumption.

Bandage specifications & requirements

Specifications

1. Top layer: Transparent polymethylpentene membrane permeable to gases, flexible.
2. Middle layer: Polyacrylamide hydrogel containing the chemically defined media.
3. Bottom layer: This layer acts as barrier between the wound and the ''L. lactis'' cells. They are inactive since the gel is normally in dehydrated form. This layer consists of a cellulose nitrate membrane of pore size 0.2 micrometers, permeable to protein, being hydrophilic and flexible.

Requirements

Most of the requirements were found by placing ourselves in the minds of end-users and derived from interviews with people working at the Martini hospital.

1. Start-up: breaking of the water pockets and hydrating gel/activating L. lactis.
2. Operation: growth of L. lactis, detecting quorum molecules and secreting Nisin, AiiA, DspB.
3. Shut-down: nutrient depletion, cell destruction.

Requirement	Value	Note
Activation time	3 h	Time required for the gel to hydrate after water pockets have been broken to be fully hydrated.
Therapeutic action time	4h	After being applied to the wound. Amount of time for reaching the inhibitory concentrations of nisin, DspB and AiiA to pass the bottom membrane and into the wound.
Inhibitory nisin concentration	2-16 ng/µL	This concentration can be used for both MRSA and VRE. Source: Nisin, alone and combined with peptidoglycan-modulating antibiotics: activity ……. Brumfitt, Salton, Hamilton-Miller.
Inhibitory AiiA concentration	15 ng/µL	How much is needed to inhibit growth of PA/SA?
Inhibitory DspB concentration	15 ng/µL	How much is needed to inhibit growth of PA/SA?
Operating temperature	20-40 °C
Lifetime	3 days

Equations

Results

Based on the simulation data we compared our design outcome with the design requirements.

Performance table

Design	Tnisin	TAiiA	TDspB	Ease of manufacturing	Rating
1	0.4	0.5	0.5	Too easy	1
2	0.4	0.5	0.5	Moderate	3
3	0.3	0.3	0.3	Moderate	2
4	0.2	0.3	0.3	Moderate	1
5	1.9	2.2	2.4	Difficult	5
6	1.9	2.3	2.4	Difficult	4

Conclusion

Out of the 6 designs we had for our bandage, design 1 and design 4 was providing the expected data at a shorter duration. Design 5 was the worst design for the bandage. So we decided to make our bandage based on design 1.

Figure 1: The best design for our purposes.

Modeling experiments

The hydrogel in the bandage should have nutrient source for the bacteria to grow. Using rich media like M17 in the bandage for growing Lactococcus lactis is not a good idea. Rich media might support the growth of other bacteria’s present on the wound which might cause adverse problems. To avoid this kind of complications we decided to use chemically defined media. Chemically defined media is a buffered media containing all the aminoacid, vitamins and other metal suppliments.

1. Glucose Concentration optimization for Nisin production

To increase the lifetime of the bandage we decided to increase the amount of carbon source. Lactococcus lactis is lactic acid producing bacteria, increasing the amount of carbon source in the media results in faster production of lactic acid. Lactic acid present in the media represses the growth of bacteria. The presence of phosphate buffer in the chemically defined media solves this problem to some extent. It has been reported that Nisin is produced only in exponential growth phase. In order to evaluate the glucose concentration at which Nisin production is higher we grew Nisin producing strain in CDM media and every two hours sample was collected to perform Nisin activity test.

Figure 5: Diffusion experiment wilt L. lactis over time.

Based on the nisin activity assay we got similar diameter of halo for 20 g/l and 40 g/l of glucose concentration. We analyzed the supernatants using HPLC to find the exact concentrations of nisin. HPLC analysis showed that at 20 g/l of glucose concentration we get higher nisin yield.

Figure 5: Growth curve for 20 g/l of glucose concentration.

initial conditions 1 matlab
initial conditions 2 matlab
initial conditions 3 matlab
initial conditions 4 matlab
initial conditions 5 matlab
initial conditions 6 matlab
two dimension model
growth curve L. lactis

Model-based bandage design

Our main goal for this project is to design a bandage prototype for burn wounds. Burn wounds are mainly infected with S. aureus and P. aeruginosa. The quorum molecules produced by these two pathogens should diffuse through the bandage and activate the production of Nisin, Aiia and DspB proteins. These three proteins should diffuse out of the bandage and act on the pathogens.

Figure 4: Scheme for modeling the bandage

To evaluate different bandage designs, we develop a multi-scale dynamic model of the bandage. The bandage is discretized into lattices where each lattice contains differential equations describing the growth of bacteria, production of nisin, production of Aiia, production of DspB and the detection of quorum molecules. Apart from the differential equations for the productions of the three IPM molecules we also consider the diffusion parameters. This makes the model more dynamic to study characteristics of our bandage.

Each state variable in each lattice is initialized according to the different bandage designs. Each lattice contains few bacteria which uses glucose as nutrient source and grows. Actively growing bacteria produce Nisin, Aiia and DspB only in response to the quorum molecules produced by both Staphylococcus aureus and Pseudomonas aeruginosa. In presence of quorum molecules in the lattice, the bacteria starts producing Nisin, Aiia and DspB. Nisin, Aiia and DspB produced in a lattice diffuses to nearby lattices until equilibrium is reached.

Studying the diffusion rates of Nisin, Aiia and DspB is important to estimate the time taken to reach the threshold concentrations. The threshold concentration is the minimum concentration of the proteins that is required to breakdown biofilm, kill S. aureus and quorum quench P. aeruginosa population. The diffusion constants for these three proteins were not available directly. For more info on the rate equations look here.

6 bandage designs

Studying the diffusion rates of Nisin, Aiia and DspB is important to estimate the time taken to reach the threshold concentrations. The threshold concentration is the minimum concentration of the proteins that is required to breakdown biofilm, kill Staphylococcus aureus and quorum quench Pseudomonas aeruginosa population. The diffusion constants for these three proteins were not available directly. They were calculated using number formulas. Diffusion rate inside the polyacrylamide gel is different from the diffusion in a solvent.

Figure-movie 1 : The bacteria are uniformly distributed all over the hydrogel in the bandage.

Figure-movie 2 : Instead of having uniform bacterial distribution, we have single layer of bacteria beneath the top membrane of the bandage.

Figure-movie 3 : Single layer of bacteria are placed in the bottom of the bandage.

Figure-movie 4 : Single layer of bacteria is placed in the middle of the gel.

Figure-movie 5 : Bacteria are localized to a single lattice in the top of the bandage.

Figure-movie 6 : Bacteria are localized to two separate lattices in the middle of the bandage.

@@ Line 6: / Line 6: @@
 <!-- TEST TEXT MODULE TEMPLATE ->
-  </html>{{:Team:Groningen/Template/MODULE/tmpl_textmodule|<b>Why modeling?</b>|Modeling is an important tool used for understanding the behavior of variables without testing it in real time. From designing small genetic circuits to space shuttles, modeling plays a pivotal role. In our case, modeling is the backbone of our project.
+  </html>{{:Team:Groningen/Template/MODULE/tmpl_textmodule|Why modeling?|''Staphylococcus aureus'' and ''Pseudomonas aeruginosa'' are the two pathogens that cause infection in the burn wounds. The aim of our project was to design a smart bandage that can produce Infection Prevention Molecules (IPMs) only in the presence of these two pathogens. The bandage has freeze-dried ''Lactococcus lactis'' cells which are activated by hydrating the gel. The hydrated gel contains all the nutrient supplements that are required for the ''L. lactis'' to grow. The quorum molecules produced by the pathogens in the wound diffuse through the hydrogel and reach the ''L. lactis'' . This triggers the production of IPMs. The IPMs which are produced by the genetically engineered ''L. lactis'' in the hydrogel must diffuse to the site of infection in the burn wounds. Understanding the diffusion of IPMs is important for us to design a smart bandage which is quite realistic. Based on the bandage requirements there were different possible designs. Modeling those designs helped us to choose the best design and study the diffusion characteristics in our bandage. The best design was further analyzed to obtain information regarding the life time of the bandage after it is being activated. The bandage model gives an overall idea about production rate of IPMs, diffusion of the IPMs, inimum number of bacteria that should be available in the bandage and nutrient consumption.
-The main focus of the modeling in this project lies in aiding the development of the actual prototype. The results from modeling should help the design of a bandage that detects pathogens in burnwounds and secretes molecules that either kill- or inhibit growth of these pathogens. Key here is the ability to produce nisin, DspB and AHLase upon sensing quorum molecules from the pathogens. In order to provide useful information to the ‘material’-people in our team, we made a first model that shows how nisin, DspB and AHLase are produced by L. lactis and diffuse through the bandage. By using this model, we can estimate which of six possible designs (LINK) is the best.
-Awesome! The ‘why modeling’-question has been answered, let’s show its actual use!
-Staphylococcus  aureus and Pseudomonas aeruginosa are the two pathogens that cause most infections in burn wounds. The aim of our project is to design a smart bandage that can produce Infection Prevention Molecules (IPMs) only in the presence of these two pathogens. This kind of bandage will reduce the problem of antibiotic resistance by reducing the amount of antibiotics applied to the wound. The IPMs are produced by our genetically modified Lactococcus lactis.  These bacteria are fixed in the hydrogel. It is key that the molecules produced by L. lactis reach the infected wound. Understanding the nuances that are associated with the diffusion of IPMs is therefore important for us.
-To start with actual modeling, we will first give an overview of the parameters involved in the modeling.
 }}<html>
+ </html>{{:Team:Groningen/Template/MODULE/project/MBD/specifications&equations}}<html>
+ </html>{{:Team:Groningen/Template/MODULE/project/MBD/results}}<html>
+ </html>{{:Team:Groningen/Template/MODULE/project/MBD/conclusion}}<html>
+ </html>{{:Team:Groningen/Template/MODULE/project/MBD/experiments}}<html>
 </div><div class="entry grid gridcell last">
 <!-- rightcolmn modules here -->
-<!-- TEST IMAGE MODULE TEMPLATE ->
+  </html>{{:Team:Groningen/Template/MODULE/project/MBD/bandage}}<html>
-  </html>{{:Team:Groningen/Template/MODULE/tmpl_imagemodule|<b>Bandage modeling</b>|https://static.igem.org/mediawiki/2014/c/cc/Modelling_scheme_1.art.png|
+</html>{{:Team:Groningen/Template/MODULE/project/MBD/diffusionrates}}<html>
-Our main goal for this project is to design a bandage prototype or burn wounds. Burn wounds are usually infected with <i>Staphylococcus aureus</i> and <i>Pseudomonas aeruginosa</i>. The quorum molecules produced by these two pathogens should diffuse through the bandage and activate the production of Nisin, Aiia and DspB proteins. These three proteins should diffuse out of the bandage and act on the pathogens. There were different possible bandage designs considering our bandage specifications. We’ve made six, to get a rough estimation of the best design.<br>
-You’ve probably seen our design on the front page of our wiki. If not, here is a link that describes the different components of the bandage in a more precise way (LINK TO 3D MODEL). We are mainly interested in how fast and how much our three Infection Preventing Molecules (IPM’s) are diffusing into the wound. That makes our to-obtain-results clear: <br>
-.	Amount of nisin that diffuses from the bandage into the wound over time.<br>
-.	Amount of DspB that diffuses from the bandage into the the wound over time.<br>
-.	Amount of AiiA that diffuses from the bandage into the the wound over time.<br>
-These amounts can be found by modeling. We made six designs with different distributions of our L. lactis throughout the gel. These designs and their modeled diffusion results are displayed below.
-}}<html>
 </div>
 </html>