Imperial/Cell by Date/Testing

From 2007.igem.org

< Imperial(Difference between revisions)
m (Validation and Conclusion)
m (<font color=blue>Investigating properties of system under isothermal scenarios</font>)
 
(60 intermediate revisions not shown)
Line 1: Line 1:
{{Template:IC07navmenu}}
{{Template:IC07navmenu}}
 +
<br clear="all">
 +
<html>
 +
<link rel="stylesheet" href="/igem07/index.php?title=User:Dirkvs/Stylesheets/IC07persist.css&action=raw&ctype=text/css" type="text/css" />
 +
 +
<div id="tabs">
 +
  <ul>
 +
    <li><a href="https://2007.igem.org/Imperial/Cell_by_Date/Introduction" title=""><span>Introduction</span></a></li>
 +
    <li><a href="https://2007.igem.org/Imperial/Cell_by_Date/Specification" title=""><span>Specifications</span></a></li>
 +
    <li><a href="https://2007.igem.org/Imperial/Cell_by_Date/Design" title=""><span>Design</span></a></li>
 +
    <li><a href="https://2007.igem.org/Imperial/Cell_by_Date/Modelling" title=""><span>Modelling</span></a></li>
 +
    <li><a href="https://2007.igem.org/Imperial/Cell_by_Date/Implementation" title=""><span>Implementation</span></a></li>
 +
    <li><a class="current" href="https://2007.igem.org/Imperial/Cell_by_Date/Testing" title=""><span>Testing</span></a></li>
 +
    <li><a href="https://2007.igem.org/Imperial/Cell_by_Date/Conclusion" title=""><span>Conclusion</span></a></li>
 +
  </ul>
 +
</div>
 +
<hr />
 +
<br clear="all">
 +
</html>
__NOTOC__
__NOTOC__
 +
= Cell by Date: Testing =
-
==Testing and Discussion==
+
==Summary==
 +
Key results of our experiments are :
-
===Investigation properties of system under isothermal scenarios===
+
1.'''Operating Range''' : Our System Operates between '''8-37 Degrees''' <br>
 +
2.'''Activation Energy''' : The activation energy of our system is '''1.5kJ/mol''' <br>
 +
3.'''Optimum DNA Concentration''' : The optimum DNA concentration for our system is '''2-4ug DNA''' <br>
-
[[Image:CBD Isothermal.png|Isothermal Data]]
+
==Aims==
-
[[Image:CBD Arrhenius Plot.PNG|Arrhenius Plot]] <br clear = "all" >
+
-
Previous work has been carried out to model the spoilage of ground beef by living organisms. 
+
Having determined that our system worked at a basic level in our [https://2007.igem.org/Imperial/Cell_by_Date/Implementation implementation phase] we set out to ascertain a clearer picture of the performance of our system to ultimately reveal whether our design had realised our specifications.
-
Above in figure 1 is is a model developed by Koutsoumanis in 2006 which colesly fit the behaviour of the spoilge organism we are interested in, Pseudomonas, under dynamic temperature conditions.  One of the Key conclusions drawn from this model is the almost instantaneous repsone time of the Pseudomonas' growth parameter.  The result of this is that our system needs to have a quick respone time to correclty report the temperature history of the beef.
+
-
Koutsoumanis' and also one Giannuzzi developed in 1998 are both based on the Gompertz model.  This model allows some insight into the mechanisms of ground beef spoilage.  In particular through manipulation of the Gompertz Parameters and assuming a Arrhenius type relationship between Pseudomonas' growth parameter and temperature we can infer the Activation energy of the spoilage reaction.  This is shown in figure 2 in which a stongly linear behaviour allows us to continue with our Arrhenius assumption and extract that Activatoin energy of the spoilge reaction.  As given in our specifications the Activation energy for ground meat seem to be around 30kJ/mol.  The result of this as per Taoukis' work is that our system needs to have a similar activation energy.  I hope to determine our system activation energy in the same way Giannuzzi determined Pseudomonas'.
+
==Results==
-
===Investigation System properties under Dynamic Temperature Conditions===
+
===<font color=blue>Investigating properties of system under isothermal scenarios</font>===
-
[[Image:CBD Steps.png]][[Image:CBD Steps.png]]<br clear ="all">
+
To estimate the properties of our system we looked at synthesis of GFPm with time for experiments where temperature is carefully kept constant. (Figure 1).<br>
 +
'''General Behaviour:''' <br>
 +
*In all experiment we can observe a linear synthesis of GFP corresponding to a steady rate of GFP synthesis.  
 +
*This steady rate is thought to be because of the degradation of GFP is negligible in our ''in vitro'' chassis, see the [[Imperial/Wet_Lab/Results/Res1.4| Degradation of GFP testing]]
 +
<br>
 +
[[Image:CBD isothermal.PNG|thumb|centre|600px|Figr 1:Isothermal Data - Averages GFPmut3b molecules synthesised for pTet-GFPmut3b construct at different temperatures]] <br clear="all">
-
With some understanding of how Pseuodmonas spoils aerobically stored ground beef it is really important for us to gain some understanding of how our sytem behaves in similar scenarios to the models developed by Koutsoumanis and Giannuzi in order to extract the same parameters.<br><br>For Isothermal conditions we have been able to devlope some simple models of the gene expression of our sytem with a strong dependance on energy depletion as will feel this is the major limiting factor of our system.  As seen above in figures 3 and 4 with energy depletion as our major limiting factor we expect the concentration of our reporter gene to increase to a peak and then as energy runs out expressoin is curbed and decay of our reporter takes over giving an exponential like decay.
+
'''Operating Range:'''<br> 
 +
*Our system seems to 'turn off' at around 4 degrees C, having negligible rates of GFP synthesis.<br>  
 +
*At 8 degrees minimal expression occurs, when the temperature is increased to 37<sup>o</sup>C the GFP expression increases substantial expression.<br>
-
==Validation and Conclusion==
+
'''Synthesis rate vs Temperature'''<br>
 +
*Over the range of temperatures considered here the GFP synthesis and hence the rate was shown to increases with temperature.<br>
 +
*To investigate the relation between synthesis rate and temperature we extracted the rate of synthesis of GFP for each experiment and plotted it against 1/T (Figure 2).<br>
 +
*A strong linear correlation in the log plot supports an arrhenius type dependence on temperature with an activation energy of 1.5kJ/mol.<br>
 +
[[Image:CBD Arrhenius Plot.PNG|centre|thumb|600px|Fig 2: Arrhenius Plot]] <br clear = "all" >
 +
 +
For more detail on this analysis click [https://2007.igem.org/Imperial/Wet_Lab/Results/CBD2.2#Results here]
 +
 +
 +
'''Other practical considerations:''' <br>
 +
*Unfortunately we have not been able to use a visible reporter and so have no idea whether the construct can in fact produce a visible signal. However we are using a stronger visual reporter such as DsRed express the system could potentially produce a strong visual signal <br>
 +
*In terms of lifespan we have had major problems with evaporation meaning that the lifespan of our system is limited to a few days as discussed in our packaging experiment below. In addition work we worked on vesicles to potentially increase the stability of our system, click here for some more information on [http://partsregistry.org/Chassis/Cell-Free_Systems/Vesicle Vesicles] <br>
 +
 +
===<font color=blue>Investigating properties of system under dynamic temperature scenarios</font>===
 +
 +
After carrying out the experiments above we saw great potential to expand on our work on rate of synthesis and temperature. As per our  [https://2007.igem.org/Imperial/Cell_by_Date/Specification specifications]  show we are also wanted to expand on the idea of looking at dynamic temperature scenarios which have potential use in monitoring the maintenance of a cold chain. The cold chain is key for transport of many key temperature sensitive products, of particular concern are the transport of vaccines in third world countries. <br><br>
 +
 +
Below in figure 3 are the results from one of our experiments in which we subjected our system to a number of steps.
 +
 +
[[Image:IC 2007 CBD steps.PNG|600px|center|thumb|Fig 3:Molecules of GFPmut3b produced on average over time for two samples at constant temperatures of 4&degC and 20&deg;C and two with a temperature step (4-20&deg;C and 20-4&deg;C)]] <br clear = "all">
 +
 +
'''General Behaviour''' :<br><br>
 +
*Step Down from 20 to 4 Degrees : Rate of fluoresence decreases in comparison to the 20 degree isothermal curve.  This is a positive result as it correlates with Koutsoumanis' meat model in our modelling section. If temperature increases the rate of synthesis of GFP increases.
 +
 +
*Step Up from 4 to 20 Degrees : Rate of fluoresence increases rapidly in comparison to the 20 degree isothermal curve.  This is a positive result as it correlates with Koutsoumanis' meat model in our modelling section in which the response time of spoilage bacteria in beef is very small.
 +
 +
===<font color=blue>Other Experiments</font>===
 +
1. DNA Concentration : We conducted this experiment to tune our system, by varying the concentration of DNA in our cell extract,to realise the optimum performance of our pTet-GFPmut3b construct.  We found that the optimum concentration of DNA is between 2-4µg on the basis of material cost and peak expression rate. For more information please click [https://2007.igem.org/Imperial/Wet_Lab/Results/Res1.8 here]
 +
 +
2. Packaging : We conducted this experiment to determine the effect of different packaging methods on the evaporation rate of our system.  We found that out none of the packaging methods investigated decreased the rate of evaopration such that our system could survive for a week, as requested by our specifications. For more information please click [https://2007.igem.org/Imperial/Wet_Lab/Results/CBD2.1 here]
-
{|border="1" width="80%" align="center"
 
-
|-
 
-
|width="20%" style="background:#ffffcc"|<center>'''Property'''</center>
 
-
|width="--"|<center>'''Value'''</center>
 
-
|<center>'''Achieved ?'''</center>
 
-
|-
 
-
|style="background:#ffffcc"|Inputs
 
-
|<center>Isothermal conditions between 0 & 40 C</center>
 
-
|<center> No </center>
 
-
|-
 
-
|style="background:#ffffcc"|
 
-
|<center>Dynamic conditions eg. steps & ramps</center>
 
-
||<center>hmm</center>
 
-
|-
 
-
|style="background:#ffffcc"|Outputs
 
-
|<center>System should give a visual signal to indicate level of thermal exposure where beef is off</center>
 
-
|<center>No we were unable to use a visual reporter</center>
 
-
|-
 
-
|style="background:#ffffcc"|Activation Energy
 
-
|<center>System needs to have an activation Energy 30 +/- 20 kJ/mol</center>
 
-
||<center>No activation energy of system was calculated to be 1.5kJ/mol</center>
 
-
|-
 
-
|style="background:#ffffcc"|Health Regulations
 
-
|<center>System must not be living replicating bacteria</center>
 
-
|<center>Yes through use of cell free chassis</center>
 
-
|-
 
-
|style="background:#ffffcc"|Response Time
 
-
|<center>System needs to have a response time under 1 hour</center>
 
-
||<center>hmm tbd</center>
 
-
|-
 
-
|style="background:#ffffcc"|Lifespan
 
-
|<center>System must have a shelf life of 7 days</center>
 
-
||<center>No with current packaging methods evaporation meant that our system 'died' after 2 days</center>
 
-
|}
 
-
<br clear="all">
 
-
Previous work has been carried out to model the spoilage of ground beef by living organisms. 
 
-
Above in figure 1 is is a model developed by Koutsoumanis in 2006 which colesly fit the behaviour of the spoilge organism we are interested in, Pseudomonas, under dynamic temperature conditions.  One of the Key conclusions drawn from this model is the almost instantaneous repsone time of the Pseudomonas' growth parameter.  The result of this is that our system needs to have a quick respone time to correclty report the temperature history of the beef.
 
-
Koutsoumanis' and also one Giannuzzi developed in 1998 are both based on the Gompertz model. This model allows some insight into the mechanisms of ground beef spoilageIn particular through manipulation of the Gompertz Parameters and assuming a Arrhenius type relationship between Pseudomonas' growth parameter and temperature we can infer the Activation energy of the spoilage reaction. This is shown in figure 2 in which a stongly linear behaviour allows us to continue with our Arrhenius assumption and extract that Activatoin energy of the spoilge reaction.  As given in our specifications the Activation energy for ground meat seem to be around 30kJ/mol. The result of this as per Taoukis' work is that our system needs to have a similar activation energy.  I hope to determine our system activation energy in the same way Giannuzzi determined Pseudomonas'.
+
<center> [https://2007.igem.org/Imperial/Cell_by_Date/Implementation << Implementation ] | Testing/Validation | [https://2007.igem.org/Imperial/Cell_by_Date/Conclusion Conclusions >> ]</center>

Latest revision as of 03:44, 27 October 2007




Cell by Date: Testing

Summary

Key results of our experiments are :

1.Operating Range : Our System Operates between 8-37 Degrees
2.Activation Energy : The activation energy of our system is 1.5kJ/mol
3.Optimum DNA Concentration : The optimum DNA concentration for our system is 2-4ug DNA

Aims

Having determined that our system worked at a basic level in our implementation phase we set out to ascertain a clearer picture of the performance of our system to ultimately reveal whether our design had realised our specifications.

Results

Investigating properties of system under isothermal scenarios

To estimate the properties of our system we looked at synthesis of GFPm with time for experiments where temperature is carefully kept constant. (Figure 1).
General Behaviour:

  • In all experiment we can observe a linear synthesis of GFP corresponding to a steady rate of GFP synthesis.
  • This steady rate is thought to be because of the degradation of GFP is negligible in our in vitro chassis, see the Degradation of GFP testing


Figr 1:Isothermal Data - Averages GFPmut3b molecules synthesised for pTet-GFPmut3b construct at different temperatures

Operating Range:

  • Our system seems to 'turn off' at around 4 degrees C, having negligible rates of GFP synthesis.
  • At 8 degrees minimal expression occurs, when the temperature is increased to 37oC the GFP expression increases substantial expression.

Synthesis rate vs Temperature

  • Over the range of temperatures considered here the GFP synthesis and hence the rate was shown to increases with temperature.
  • To investigate the relation between synthesis rate and temperature we extracted the rate of synthesis of GFP for each experiment and plotted it against 1/T (Figure 2).
  • A strong linear correlation in the log plot supports an arrhenius type dependence on temperature with an activation energy of 1.5kJ/mol.
Fig 2: Arrhenius Plot

For more detail on this analysis click here


Other practical considerations:

  • Unfortunately we have not been able to use a visible reporter and so have no idea whether the construct can in fact produce a visible signal. However we are using a stronger visual reporter such as DsRed express the system could potentially produce a strong visual signal
  • In terms of lifespan we have had major problems with evaporation meaning that the lifespan of our system is limited to a few days as discussed in our packaging experiment below. In addition work we worked on vesicles to potentially increase the stability of our system, click here for some more information on [http://partsregistry.org/Chassis/Cell-Free_Systems/Vesicle Vesicles]

Investigating properties of system under dynamic temperature scenarios

After carrying out the experiments above we saw great potential to expand on our work on rate of synthesis and temperature. As per our specifications show we are also wanted to expand on the idea of looking at dynamic temperature scenarios which have potential use in monitoring the maintenance of a cold chain. The cold chain is key for transport of many key temperature sensitive products, of particular concern are the transport of vaccines in third world countries.

Below in figure 3 are the results from one of our experiments in which we subjected our system to a number of steps.

Fig 3:Molecules of GFPmut3b produced on average over time for two samples at constant temperatures of 4&degC and 20°C and two with a temperature step (4-20°C and 20-4°C)

General Behaviour :

  • Step Down from 20 to 4 Degrees : Rate of fluoresence decreases in comparison to the 20 degree isothermal curve. This is a positive result as it correlates with Koutsoumanis' meat model in our modelling section. If temperature increases the rate of synthesis of GFP increases.
  • Step Up from 4 to 20 Degrees : Rate of fluoresence increases rapidly in comparison to the 20 degree isothermal curve. This is a positive result as it correlates with Koutsoumanis' meat model in our modelling section in which the response time of spoilage bacteria in beef is very small.

Other Experiments

1. DNA Concentration : We conducted this experiment to tune our system, by varying the concentration of DNA in our cell extract,to realise the optimum performance of our pTet-GFPmut3b construct. We found that the optimum concentration of DNA is between 2-4µg on the basis of material cost and peak expression rate. For more information please click here

2. Packaging : We conducted this experiment to determine the effect of different packaging methods on the evaporation rate of our system. We found that out none of the packaging methods investigated decreased the rate of evaopration such that our system could survive for a week, as requested by our specifications. For more information please click here



<< Implementation | Testing/Validation | Conclusions >>