Chi.Bio not detecting fluorescence from GFP,RFP cultures

[jmarken]

Hello, we are running into some issues with reproducing the GFP/RFP fluorescence expression readouts from Figure 5 in the PLoS Biology Chi.Bio paper. In summary, our Chi.Bio cannot distinguish between fluorescent culture and blank media.

We have cloned constructs expressing sfGFP and mRFP1 with fairly high expression (J23100 promoter, B0034 RBS, pSC101 low-copy origin) and transformed them into E. coli BW25113. We have confirmed that these cells are appropriately fluorescent using non-Chi.Bio methods.

We then grew monocultures of the GFP,RFP strains to saturation in M9 media and diluted them down to an OD of ~0.5 (as measured on the Chi.Bio). We then calibrated our fluorescence measurement settings to match those used in the paper (briefly, laser power 0.1, gain 512x, 457/550 excitation/emission for GFP and 595/670 for RFP, ‘Clear’ as baseband). We measured fluorescence from the two monocultures and a blank M9 control with the following procedure:
1. Toggle on the ‘Active’ button for FP1 and FP3 (which are calibrated for the GFP,RFP settings, respectively).
2. Manually switch on the relevant excitation LED from the ‘Light Outputs’ box.
3. Click ‘Measure Fluorescent Proteins (FPs)’.
4. After a few seconds, the LED automatically toggles off. Then, record the values reported for Baseband and the relevant Emit.

We found that across all three conditions (blank M9, GFP cells, and RFP cells), the Emit readings are essentially identical, with FP1 (GFP) reporting a value of around 0.017 and FP3 (RFP) reporting a value of around 0.016. The Baseband value itself does change more dynamically: when excited by 457nm, it returns a value of around 9000 for blank M9 and 20,000 for the two cell cultures; when excited by 595nm, it returns a value of around 12,000 for blank M9 and 27,000 for the two cell cultures.

We have so far performed this measurement on two different reactors and obtained the same general result– although the values of the reported numbers change, there is still no way to use the reported FP values to distinguish between empty M9 culture, GFP+ cells, and RFP+ cells.

We would appreciate any thoughts on this matter.

Thank you,
John

[harrison]

Hi John,

A few points/questions
– How “fluorescent” are the “appropriately fluorescent” cells? Do you mean they are at some level that (for example) is visible to the eye? How much expression did you measure from them as a fraction of e.g. “the brightest GFP E.Coli you have ever seen”
– You shouldn’t need to manually switching on the excitation LEDs first. When you click “Measure FPs” it will turn them on itself (and then off again ater the measurement).
– For the GFP have you tried exciting with the lower (395nm) LED?

Generally the system works best when you are not taking tubes in and out of the reactors. There is some wiggle room around the tube and this can impact absolute values of readings. Furthermore, when you start/stop stirring and heat/cool the reactor these will all make some (small) impact on the measurement. Instead, what the system is really designed to do is to track a single test tube (and its contained culture) over time. This is what is done in Figure 5 of the PLoS Biology Paper! So, to best assess your setup I would either a) Test with an inducible construct, which you can maintain at a set OD and then induce and observe fluorescence production or b) Grow your GFP cells to a fixed OD and maintain them there for some time, then pipette into the reactor concentrated RFP cells (say, with the goal of the resultant culture being 50% GFP and 50% RFP) and see how this changes the GFP/RFP balance measured over time.

[jmarken]

Hi Harrison,

Thanks so much for your prompt and thorough answer. We’ll try your suggestions and get back to you with the results.

John

[fbertaux]

Hi John and Harrison,

I can comment on our experience with yeast using light-induced expression mNeonGreen (single integrated copy of the promoter, full induction around pTDH3 strong promoter). We can clearly see the fluorescence reading change when switching ON or OFF the induction for a single culture in turbidostat mode. But the signal is low, so raw numbers cannot be compared between reactors. Also the OD matters a lot, even a small ‘dither’ will be visible on the reading. To be able to compare data between different runs and different reactors, we are now testing if a single calibration (specific to a given reactor, OD and media obviously) provides week-to-week reproducibility.

François

[harrison]

Thanks for the update François, that sounds about what I expect – given it is a single copy. One challenge of week-to-week reproducibility for small signals might be movement of the test-tube when you take it in or out. Your best bet might be to have it “auto calibrate” at the start of experiments (or whenever you expect there is ~zero fluorescence) or something like that.

Do your reactors have (white plastic) diffusers in their spectrometers? There could be a small white sheet just behind the apperture of the spectrometer (i.e. the circular part with hole in the middle). If not, I can mail you some – they can significantly improve signal-to-noise for small measurements (mostly it improves OD measurement, but also fluorescence to some extent).

Harrison

[Author]

Posts