William and Mary Team
As part of their project, the William and Mary iGEM Team were imaging cells expressing a variety of fluorescent protein including RFP, CFP, GFP, and YFP. The constructs are being used for the Interlab Measurement study. As part of the measurement track the W&M team assembled fluorescent constructs, and measured and reported their fluorescence. The samples were sent to us to confirm the correct assembly of the parts by validating their fluorescence. Their interest in seeing how their constructs performed (in terms of florescent output) perfectly matched our own need to test OpenScope on cell samples rather than fluorescent beads. In addition, optical components were obtained to extend our imaging capabilities to both GFP and RFP. In particular, RFP imaging has not been tested previously.
The W&M team were kind enough to send us dried DNA samples containing constructs for GFP and RFP expression in E. coli. Details on the protocols used to resuspend the DNA and to transform the cells can be found below.
Transformation Protocol for William and Mary iGEM Team 2015
Context
Using E. Coli K12 Dh5 alpha cells
Dry plasmid for two different GFP constructs in duplicate (so 4 samples total) sent on filter paper
All constructs on Amp resistance backbones pSB1C3
Constructs are J23106 + I13504 on pSB1C3, and J23117 + I13504 on pSB1C3
Constructs generated through DNA synthesizing the insert (J231XX + I13504) and using Gibson Assembly to insert it into the Amp backbone
Objective:
Transform cells with the constructs provided
Image the cells for RFP and GFP fluorescence using OpenScope to confirm functionality of the inserts, and the microscope itself
DNA Extraction:
Adapted from: Extraction protocol
Cut center of circles out of the filter paper
Added 50 µl of EB buffer to resuspend DNA
Pipetted up and down to mix
Transformation:
Adapted from: Transformation protocol
Add 1 µl of DNA extract to thawed out competent Top10 E. coli cells (except control plate)
Mixture allowed to stand on ice for 15 minutes
Cells heat shocked for 30 seconds at 42˚
Mixture allowed to stand on ice for 5 minutes
100 µl of LB added at room temperature
Cells incubated for 1hr at 37˚ with shaking at 120 rpm
Cells plated out on Amp plates containing rich medium
Results:
After preliminary testing using the RFP epi-cube, it was decided that imaging of RFP at this stage would not be possible. Hence only bacterial strains expressing GFP (as confirmed earlier) were tested against the control.
Earlier testing using the standard set-up (single LED illumination, GFP epi-cube) indicated that visualising fluorescent beads was possible with OpenScope (Fig. 1).
i) William and Mary
Testing using the commercial fluorescence microscope confirmed that the samples had phenotypes as reported by William and Mary. Both constructs resulted in GFP expression in E. coli. J23106+I13504 demonstrated increased fluorescence intensity relative to J23117+I13504.
A control slide (sample taken from the agar of the control plate, untransformed cells plated on Amp) was first tested to establish a fluorescence-free baseline (Fig. 2a). The duplicate J23106+I13504 samples were then tested using the standard set-up (100mW LED) and non-standard set-up (3W LED). Fluorescence was detected (Fig. 2b and 2c).
Duplicate J23117+I13504 samples were then tested using standard set-up and the non-standard set-up, but the brightness was insufficient to image the fluorescence. This confirms the reduced fluorescence intensity of the J23117+I13504 samples compared to the J23106+I13504 samples.
ii) Glasgow
Testing using the commercial fluorescence microscope confirmed that the samples had phenotypes as reported by Glasgow. From tube 5, a small proportion of the cells were expressing RFP and the majority expressed GFP as predicted.
Results from preliminary testing of DH5α cells with p126.1 and p56.1 (confirmed: GFP expression only) using the standard set-up indicated illumination brightness was insufficient to detect GFP. The non-standard set-up was used, and GFP expression was confirmed in p126.1, p126.+p56.1 and p126.1+p80.1 cells as expected (Fig. 3a-d). The images display an artefact of the square shape of the LED used, as there is an area of increased fluorescence in the outline of a square at the centre of the images (Fig 3c and d).
Conclusions:
The first objective of the collaboration was to confirm the phenotypes of the bacterial strains and the expression of fluorescent proteins. Due to the technical limitations of OpenScope (see below), RFP could not be visualised. Hence a commercially available microscope was used to confirm RFP expression. In addition, GFP was reliably confirmed using the same microscope. Across the board, expression of the fluorescent proteins was as reported by Glasgow and W&M iGEM teams. This confirms the functionality of the plasmids.
After successful visualisation of fluorescent beads labeled with GFP (Fig. 1), it was expected that OpenScope would enable visualisation of E. coli expressing GFP. However, results indicate that reliable detection was not possible using the standard set-up (single LED illumination, GFP epi-cube). Possible explanations for this are as follows:
The fluorescent beads used are far brighter than biological samples, and therefore can be visualised with low-brightness LEDs
The bacteria, when prepared on slides, form a relatively uniform thin surface. Flourescence from this is difficulty to detect, as OpenScope cannot resolve individual bacterial cells
Replacement of the 100mW LED with the 3W LED allowed visualisation of samples with reduced fluorescence intensity in the case of the p126.1, p126.+p56.1 and p126.1+p80.1 cells. The 100mW LED was sufficient only to image the J23106+I13504 samples. Overall, the results suggest that in order to reliably detect fluorescence the 3W LED is more appropriate. However, this is still not sufficiently reliable to make OpenScope useful for fluorescence screening at this stage. In addition, the artifact (image of the LED itself) seen when using the 3W LED means that the uniformity of illumination must be improved.
RFP is more challenging to image, as it has a narrow gap between the excitation (584nm) and emission (607nm). Hence it is difficult to find low-cost dichroic mirrors that are transparent to wavelengths around 607 nm while being reflective to wavelengths around 584 nm. In addition, sourcing LEDs with an emission peak in the region of 584 nm was not possible. As such, the LEDs used had an emission peak at 591nm, which is closer to the transparency region for the dichroic mirrors. Overall, RFP imaging has not yet been demonstrated as a proof of concept. Perhaps this could be starting point for future iGEM teams looking to build on the OpenScope project.
