Making agar bridges for electrophysiology


  • agarose
  • potassium chloride (KCl; MW = 74.54 g/mol)
  • glass capillaries
  • ethanol burner
  • syringe


  1. Prepare bent capillary tubes. Alight an ethanol burner by dipping the wick in EtOH, e.g. filling a 1.5 ml tube (eppendorf) with ethanol and dipping the wick in it. Hover the capillary (at 1/3 of its length) over the fire and keep pushing on the short end with a pen until the capillary is bent at a right angle.
  2. Prepare a 20 ml solution at 1% agarose. Weight 0.2 g agarose and dissolve it in 20 ml ddH2O in a 50 ml Falcon tube.
  3. Weight the appropriate amount of KCl in order to achieve a final concentration of 3 M. I need:
    0.020 l x 3 mol/l = 0.06 mol KCl
    0.06 mol x 74.54 g/mol = 4.4724 = 4.47 g KCl

  5. Microwave the agarose solution.
  6. Add the KCl and dissolve it.
  7. Aspire the solution with the syringe.
  8. Fill bent capillary tubes with the solution. Tip: hold the connection between capillary and syringe in order to prevent the mixture from spilling around the capillary. Quickly dry capillaries. Trim the ends of capillaries with a diamond cutter.
  9. Capillaries should not contain any bubbles.


    Store the capillaries, i.e. agar bridges, at room temperature in a 3 M KCl solution.


Day 1: split cells

Matrigel 6 cm diameter Petri dishes:

  • Dilute matrigel 100-fold with DMEM without any additives
  • Pipet 3 ml of diluted matrigel on a 6 cm diameter Petri dish
  • Incubate for 30 min at 37°C

During the incubation, trypsinise a T75 flask of GLUTag cells and resuspend them in 10 ml medium (50/50 fresh/conditioned media).
Remove the matrigel from the dish and add 4 ml full DMEM medium.
Add 2.5 ml of the resuspension in the dish and incubate overnight at 37°C/5% CO2.

Day 2: transfect cells using Lipofectamine-2000

  • 1 hour before transfection, replace the full DMEM medium. Don’t forget to warm the medium in the 37°C water bath.
  • Prepare the plasmid and Lipofectamine-2000 seperately in OptiMEM. Use 3 ug of plasmid and 12 ul of Lipofectamine-2000. Eg. if the plasmid concentration is 1.5 ug/ul:
    1. Pipet 298 ul OptiMEM in a 1.5 ml eppendorf tube labelled “P” (“P” stands for “plasmid”)
    2. Pipet 288 ul OptiMEM in a 1.5 ml eppendorf tube labelled “L” (“L” stands for “Lipofectamine-2000”)
    3. Pipet 2 ul pDNA in the “P” tube.
    4. Pipet 12 ul Lipofectamine-2000 (straight from the fridge) in the “L” tube. Do not mix the lipofectamine-2000 by pipetting up and down. Lipofectamine is sticky and will stick to the pipette if this is done!
  • Let the two tubes sit for 10 min in the hood.
  • Add the tube “P” content to the tube “L” (never do the reverse, the less you pipette Lipofectamine-2000, the better)
  • Incubate for 20 min in the hood
  • Add the 600 ul mixture to the 6 cm diameter dish dropwise.
  • Incubate overnight at 37°C/5% CO2.

Transfer transfected cells into smaller recording dishes

This step is necessary in order to go from a confluent layer of cells to single cells scattered on a dish that are amenable to being patch-clamped.

  1. Warm up medium, PBS and trypsin in the water bath.
  2. Wash cells with 10 ml PBS.
  3. Trypsinise for 3-5 minutes with 2 ml trypsin.
  4. Stop trypsinisation using 6 ml DMEM.
  5. Triturate 30 times.
  6. Centrifuge on a table-top centrifuge at 700 rpm for 5 min at room temperature.
  7. Throw away supernatant (this step gets rid of the trypsin)
  8. Resuspend in 10 ml DMEM and triturate 30 times.
  9. Transfer 5 ml in a new tube and add 45 ml DMEM.
  10. Pipet 2 ml in 3.5 cm diameter dishes for patch-clamp experiments the following day.
  11. Each dish contains 0.5 x 2/50 = 0.5 x 1/25 = 0.5 x 0.04 = 0.02 = 2 % of the initial cells. It is advisable to make another two-fold dilution to get 1% of the initial cells on a few dishes in case the former dilution is not strong enough.

Day 4: patch-clamp

Intracellular (pipette) solution for whole-cell patch-clamp

This solution will be in dialysis with the cytosol and hence is similar to it in composition (high potassium and low sodium).

Chemical mM MW/concentration for 100 ml
KCl 107 74.55 0.7976 g
CaCl2 1 1 M 100 ul
MgCl2 7 1 M 700 ul
EGTA 11 380.35 0.4183
HEPES 10 238.3 0.2383 g
Na2ATP 5 569.16 0.2846

Adjust to pH to 7.2 using potassium hydroxide (KOH).

Always check the molecular weight on the bottles of the chemicals. It may vary, due to different water contents for example.


Chimerel et al., Bacterial metabolite indole modulates incretin secretion from intestinal enteroendocrine L cells, Cell Rep., 2014
Link to the article in PubMed

Extracellular (bath) solution for whole-cell patch-clamp in GLUTag cells

This solution is needed to keep cells in a physiological environment during patch-clamp experiments. It also contains glucose as a nutrient.

Chemical mM MW/concentration for 1 L for 500 ml for 500 ml 10X
NaCl 138 58.44 8.065 g 4.0325 g 40.325 g
KCl 4.5 74.55 0.335 g 0.1675 g 1.675 g
NaHCO3 4.2 84.01 0.3528 g 0.1764 g 1.764 g
NaH2PO4 1.2 120 0.144 g 0.072 g 0.72 g
CaCl2 2.6 1 M 2.6 ml 1.3 ml
MgCl2 2.6 1 M 1.2 ml 0.6 ml 6 ml
HEPES 10 238.3 2.382 g 1.1915 g 11.915 g

Glucose: 1 mM (18 MG/100 ml)
pH 7.4 using NaOH
For a 500 ml 1X solution, I added 2.4 ml of NaOH at 1 M concentration.

Add CaCl2 and adjust the pH when the solution concentration is 1X.
Add glucose to the solution needed on the day, usually 50 ml.

Always check the molecular weight on the bottles of the chemicals. It may vary, due to different water contents for example.


Rogers et al., Electrical activity-triggered glucagon-like peptide-1
secretion from primary murine L-cells, J. Physiol., 2011
Link to the article in PubMed

Culturing GLUTag cells

Starting point: confluent T75 flask containing stuck GLUTag cells and 20 ml DMEM medium

All steps are undertaken in a hood!

Day 1: Splitting a confluent flask

  1. Label one 50 ml falcon tube with “condition” and one with “rubbish”
  2. Transfer the DMEM from the flask into the falcon labelled “condition”
  3. Wash cells with 10 ml PBS and transfer the dirty PBS into the “rubbish” tube
  4. Add 3 ml trypsin to the flask and leave it in the incubator for 3-5 min
  5. Check under the microscope that the cells have detached
  6. Add 5 ml fresh medium to stop the trypsinisation reaction
  7. Spin the cells down at 600 rpm for 5 min
  8. Chuck the supernatant
  9. Resuspend the cells in 5 ml fresh medium and triturate them 20 times with a 5 ml stripette (Note: using a 5 ml stripette rather than a bigger-volume one will help separate the cells due to the narrower aperture of the pipette)
  10. Add 5 ml condition medium from the “condition” tube to the resupsended cells
  11. Add 3 ml of the resuspension to a new T75 flask containing 17 ml fresh DMEM medium
  12. The other 7 ml can be used to seed Petri dishes for use in experiments, e.g.: calcium imaging, electrophysiology, secretion experiments, transfection, etc.

Day 3: Changing medium

Exchange 10 ml supernatant for 10 ml fresh DMEM medium

Day 5: Splitting a confluent flask

Refer to Day 1.

Cristian Riccio, Institute of Metabolic Science, University of Cambridge


Thanks to Dr. Edward Emery and Dr. Frank Reimann from the Institute of Metabolic Science for their contribution to the establishment of this protocol.

Production of selenomethionine derivative of a protein

Day 1: Inoculate overnight culture

Autoclave 2 x 200 ml LB medium.
In the evening, inoculate the medium (containing the appropriate antibiotic) with either a bacterial colony or a chunk of glycerol stock. Grow overnight at 37°C with shaking at 220 rpm.

Day 2: Grow SeMet protein

    • Inoculate 10 ml of overnight culture into a fresh medium (200 ml) containing the appropriate antibiotic (e.g. ampicillin at 100 ug/ml final concentration). Grow at 37°C, 220 rpm until the OD600 reaches 0.3.
    • While the bacterial cells are growing, prepare the SelenoMet Medium Base (cf. Fig. 1)
    • Spin down in table-top centrifuge (e.g. Allegra X-15R (Beckman Coulter)) in 4 x 50 ml falcon tubes.
    • Resuspend with 20 ml of ready-to-use SelenoMet Medium Base taken from 1 litre prepared above
    • Grow at 37°C, 220 rpm until OD600 = 0.6.
    • Induce with IPTG (1 mM final concentration) and transfer to 18°C.

Day 3: Purification of the protein

Refer to protocol for purification of native protein.


Secretion experiment with transfected cells

Day 1 : split cells

Starting point: T75 flask with GLUTag cells at a confluence > 80%

End point: 10 cm diameter Petri dish with GLUTag cells

  1. Transfer 5 ml of condition medium into a falcon tube.
  2. Throw away the rest of the condition medium (should be 15 ml)
  3. Wash cells with 10 ml PBS
  4. Add 3 ml trypsin and incubate for 5 min in the 37°C/5% CO2 incubator (check that cells are detached at the end of this short incubation and, if not, tap on the side of the flask to detach cells)
  5. During the trypsinisation, add 2 ml complete DMEM to the 5 ml condition medium
  6. At the end of the trypsinisation, add the 7 ml medium and triturate the cells. The total volume of cells is now 10 ml
  7. Add 6 ml complete DMEM to each of two 10 cm diameter Petri dishes
  8. Add 4 ml of resupspended cells to each of the Petri dishes
  9. Add the remaining 2 ml of resuspended cells to a new T75 flask containing 18 ml complete DMEM
  10. Incubate overnight in the 37°C incubator containing 5% CO2

Day 2 : Transfection day

Starting point:

  • 10 cm diameter Petri dish with attached GLUTag cells x 2

End point:

  • 10 cm diameter Petri dish with transfected GLUTag cells x 1
  • 10 cm diameter Petri dish with non-transfected GLUTag cells x 1


  • Transfection agent: Lipofectamine-2000
  • plasmidic DNA
  • Optimem

Initial and final conditions

  • Initial medium volume: 10 ml complete DMEM
  • Final medium volume: 6 ml complete DMEM + 1.5 ml Optimem
  • Initial pDNA concentration: 0
  • Final pDNA concentration: 0.04 ug/cm^2, i.e. 3 ug/75 cm^2
  • Initial Lipofectamine-2000 concentration: 0
  • Final Lipofectamine-2000 concentration: 12 ul/ug pDNA, i.e. 36 ul/75 cm^2


pDNA stock concentration: 2 ug/ul

Procedure (in hood)

  1. Pipet 750 ul of Optimem in each of two eppendorf tubes
  2. Pipet 3 ug/(2 (ug/ul)) = 1.5 ul pDNA into one tube
  3. Pipet 3 x 12 ul/ug = 36 ul Lipofectamine-2000 into the other tube
  4. Wait 5-10 min
  5. Transfer the content of the tube containing the pDNA into the tube containing the Lipofectamine-2000
  6. Incubate in the hood for 20 min
  7. Pipet dropwise evenly on the 10 cm diameter Petri dish containing the cells to be transfected
  8. Put the cells back in the 37°C/5% CO2 incubator

Day 3 : Transfer cells into 24-wells plate

Starting point: Petri dish with transfected cells
End point: 24-wells plate with transfected cells

  1. Dilute matrigel 100 x with cold DMEM (without any additives such as L-glutamine, FBS, pen/strep).
  2. Pipet 250 ul diluted matrigel into each well
  3. Incubate the plate for 30 min in the incubator
  4. In the meantime, wash and trypsinise the cells with 2 ml trypsin (6 cm dish). Centrifuge and resuspend in 13 ml DMEM (6 cm dish).
  5. Throw away the matrigel
  6. Pipet 1 ml cells into each well

Day 4 : Stimulate secretion

Starting point: 24-well plates containing attached transfected and untransfected GLUTag cells
End point: Frozen supernatants from stimulated cells

    1. Prepare washing solution: extracellular solution + 10 mM Glucose + 1 mg/ml BSA (Bovine serum albumin). E.g.: 50 ml extracellular solution + 90 mg Glucose (10 mM final concentration) + 50 mg BSA
    2. Prepare stimulating solutions:
      A: 4 ml washing solution (negative control)
      B: 4 ml washing solution + 4 ul forsklin at 10 mM (1000 X) + 4 ul IBMX at 100 mM (1000 X) (positive control)
      C: 4 ml washing solution + 0.4 ul stimulating agent
      D: 4 ml washing solution + 4 ul stimulating agent
    3. for (j in 1:3) {
      for (i in 1:4) { # 4 rows in a 24-wells plate
      empty row i
      fill row i with 400 ul washing solution
    4. Empty row. Add 250 ul condition 3 wells at a time (the 3 wells containing the same condition)
    5. Incubate 2h at 37°C
    6. 30 min before the end of the incubation, prepare tubes. The total number of samples being N, we’ll need 2 x N tubes. The first set of tubes will be stored on wet ice (water ice) and the second set of tubes will be stored on dry ice (carbon dioxide ice). Both sets of tubes are labelled from 1 to N and the tubes in the second set are labelled with a barcode going from X to X + N – 1 (X being any integer).
    7. At the end of the incubation, transfer the plates from the incubator to wet ice and transfer the supernatants in the first set of tubes
    8. Spin the tubes for 5 min at 2000 rpm at 4°C
    9. Transfer the supernatant (be careful not to transfer any pellet) from the first set of tubes to the second set of tubes
    10. Store at -80°C until ready for analysis

Cristian Riccio, Laboratory of Prof. Gribble and Dr. Reimann, Institute of Metabolic Science

Tirage de pipettes pour patch clamping

Aujourd’hui, nous allons découvrir comment tirer des pipettes utilisées en patch clamping, une technique du domaine de l’éléctrophysiologie. Pour cela, une machine qui tire les pipettes (les rend très pointues pour que la pointe puisse s’attacher à une cellule) est nécessaire. La machine en question est montrée en Fig. 1.

Fig.1 Machine à tirer les pipettes. La manette sert à régler la hauteur de la boucle. La boucle chauffe pour modeler le verre. Les vis servent à fixer la pipette.
Fig.1 Machine à tirer les pipettes. La manette sert à régler la hauteur de la boucle. La boucle chauffe pour modeler le verre. Les vis servent à fixer la pipette. Le stoppeur arrête la première descente de la boucle chauffante.
  1. Ouvrir la protection en verre.
  2. S’assurer que la manette est complètement tournée dans le sens des aiguilles d’une montre.
  3. Insérer la pipette pour ne laisser qu’un centimètre sortir au dessus de la vis supérieure.
  4. Insérer la pipette derrière la vis inférieure
  5. Serrer les vis supérieure et inférieure.
  6. Fermer la cage en verre.
  7. Positionner le stoppeur à gauche.
  8. Appuyer sur start.
  9. La boucle va chauffer et affiner le cylindre en son milieu.
  10. Attendre quelques secondes pour que le verre se resolidifie.
  11. Utiliser la manette pour positionner la boucle au milieu de la partie affinée.
  12. Déplacer le stoppeur à droite.
  13. Appuyer sur start.
  14. La boucle va chauffer et séparer le cylindre en deux pipettes pointues.
  15. Ouvrir la cage en verre et sortir les pipettes inférieure puis supérieure. Les sortir sur le côté et attention à ne pas endommager la pointe.
  16. Conserver les pipettes sur de la gomme d’architecte dans une boîte de Pétri (Fig. 2) ou passer au cirage (Fig. 3).
Fig. 2 Stockage des pipettes à patch clamp sur de la gomme d'architecte dans des boîtes de Pétri.
Fig. 2 Stockage des pipettes à patch clamp sur de la gomme d’architecte dans des boîtes de Pétri.
Fig. 3 Cirage des pipettes.
Fig. 3 Cirage des pipettes.


Merci au Dr. Ed Emery pour ses explications détaillées de la procédure.

Théorème de Bayes et diagnostics

Un article très intéressant est paru sur la BBC traitant des problèmes liés au diagnostic de maladies comme le cancer. Les problèmes naissent du fait que les les personnes saines sont plus nombreuses que les personnes malades (prévalence de la maladie basse) et que le test diagnostic peut être positif (i.e. la maladie est détectée) bien que le patient soit sain (taux de faux positifs différent de 0). Cela veut dire que même si le test est positif, les chances sont toujours hautes que la personne soit saine.

Exercice tiré de l’article

Quelle est la probabilité qu’une femme testée positive (i.e. malade) à une mammographie soit malade étant données les données suivantes ?

1. La probabilité qu’une femme ait le cancer du sein est de 1% (prévalence).

2. Si une femme a le cancer du sein, la probabilité que le test soit positif est de 90% (sensibilité du test).

3. Si une femme n’a pas le cancer du sein, la probabilité que le test soit quand même positif est de 9% (taux de fausse alerte).

Réponses possibles

a. 90%

b. 80%

c. 10%

d. 1%

Danse l’étude présentée par l’article, plus de trois quarts des médecins présentés avec cette question ont répondu faux ! Beaucoup (presque la moitié) ont répondu 90%, peut-être en raison du point 2 ci-dessus (probabilité que le test soit positif si la femme a le cancer). Cependant, la question est la probabilité que la femme ait le cancer si le test est positif ! Ce n’est pas la même chose !

Pour répondre à la question, il faudra rouvrir les cours de probabilité, en particulier il faut utiliser le théorème de Bayes.

P(A|B) = \frac{P(B | A) P(A)}{P(B|A)P(A) + P(B|\bar A)P(\bar A)}

Adaptons le théorème à notre cas particulier.

A : la femme a (vraiment) le cancer

B : le test est positif, le diagnostic suggère que la femme a le cancer

P(A | B) : étant donné que le test est positif (condition), la probabilité que la femme ait vraiment le cancer du sein

Les barres au-dessus des lettres indique la négation.

En remplaçant les probabilités données dans l’exercice dans la formule, nous obtenons ceci:



ce qui est environ égal à 10% ! bien loin des 90 % suggérés par les médecins.

Notez que le 99% (probabilité de ne pas avoir le cancer du sein) est le complémentaire du 1% (probabilité d’avoir le cancer du sein, sans a priori).

Cristian Riccio, le 7 juillet 2014


article de la BBC
Théorème de Bayes sur Wikipédia