Exeter/7 June 2013

Dr. George Littlejohn and Professor Rob Beardmore spoke to us about genetically encoded biosensors and modelling respectively. George's talk gave us some excellent ideas about working with biosensors and downstream responses, whilst Rob gave us a basic explanation of how modelling can be utilised in biology and chemistry, which gave our physics students some good points to think about for when we chose a project.

Genetically encoded biosensors

Dr. George Littlejohn

Stimulus produce downstream responses

• Stimulus could be light/chemical/hormone/change in pH/etc

• Receptor affects gene expressions, response is generally production of a new protein/enzyme/metabolites/more signalling molecules to start a chain reaction in the cell

Biosensors

• Been done: metal ions (Zn, Ca), cAMP, reactive oxygen species (ROS), IP3

Genetically encoded reporters of gene expression

• Expression of GUS generates a blue colour in tissues where a gene is expressed, but damages the tissues. Gives localised and scaled information (more blue = more gene expression).

• Luciferase is non-invasive and give real-time data (dynamic).

Aequorin

• Uses a jellyfish protein which binds Ca and coelenterazine and produces light

• Easily visualised

Green fluorescent protein (GFP)

• From Aqueoria victoria jellyfish

• VERY widely used, can have colours other than green

Biosensors

• Usually use luminescence or fluorescence

• pH sensitive probes will only work in certain cellular compartments (eg. areas of mitochondria have a very low pH due to H+ pumping in ATP synthesis)

FRET – Fluorescence resonance energy transfer

BRET – Bioluminescence resonance energy transfer

Alkene BioSensor

• AlkS bind to its own promoter and a promoter for AlkB, it also binds octane

• Exposed cells to different chain length alkanes; optimum response for octane, and octane alone

• Useful for a screening experiment; trying to get colonies to automatically report of they’ve made octane, used when working with very large samples and systems

Yellow CaMeleons

• No Ca present, only get blue light

• Ca present, bring 2 fluorescent proteins together and exciting the blue protein transfers energy to the yellow protein, so yellow and blue light emitted

What concentrations do we want our biosensor to work under? Will they be affected by pH?

OxyR

• E. coli binds H2O2 using a sensor molecules

• Associated with a YFP, so presence of hydrogen peroxide will generate fluorescence

Amy idea

• Two red fluorescent proteins, if they associate into a dimer, fluorescence is increased 10 fold (MUCH larger signal than FRET/BRET, which has smaller and smaller signals the futher down a “cascade” you go)

• No papers published so far on this

Sensors that have not been covered yet

• Jasmonic acid (JA)

• Abscisic acid (ABA)

• Ascorbate

• What do they mean in a plant? What do we find out by sensing their levels?

• Biosensors for build-up of materials; makes it easy to see problem areas in, for example, car engines, pipelines, storage tanks, fermenters.

You need SPECIFICITY – if you have 2 fluorescent dimers which have a natural affinity for one another, you get false positives and false negatives. But you can exploit dimer’s natural affinity for one another (the Amy idea where fluorescence is massively increased)

Maths and Biology - The Art of Modelling

Professor Rob Beardmore

First slide – two videos

• Left hand video is using GFP in Salmonella. Some cells have much higher expressions (more green colour/less gene colour) so the cells are not homogenous. Looking at gene that encode “rotor” for FliC, used to move gradients of resources in the cells.

• Current ideas are that treating a patient with a massive dose of antibiotic is the best plan, but can we use antibiotics to turn genes off? (eg. switch of the gene for FliC, bacteria can’t move and won’t be able to have as much control over its environment and food supply)

Simplify the process of making a protein

Have genes which regulate each other’s expression to give a reliable output which can adapt to a changing environment (lots of genes are repressed by the protein they code for, so when enough protein has been produced by transcription/translation, the system switches itself off to prevent wasting amino acids and energy)

• Can model these processes using simple equations, however this is going to be incredibly inaccurate! It’s like a pendulum equation but including the decay of mRNA and final protein. You can show if the presence of a protein activates or represses gene expression

• The processes in a cell are very noisy! Loads of genes interact, as do the proteins they code for.

Rob says: try and work with small numbers of genes! And break down questions to their “lowest levels”

Take me back to the notebook.