Introduction: Measuring the Invisible
As we were planning our project we considered the yellow fluorescence protein mVenus to be a straight-forward screening tool. With our first successful transformations, we realized how wrong we were. Equipping an organism full of pigments and photoreceptors like C. reinhardtii with a fluorescent protein for concentration measurements is quite brave, but using it for screening seemed to us to be a manageable task. Our first plate reader results taught us otherwise. With every measurement indicating a higher or equal fluorescence for wild type algae than for the transformed, frustration started to rise. Measuring YFP in C. reinhardtii turned out to be like looking for a fluorescent needle in a colorful haystack.
Our first approach was to perform colony PCRs on every colony, that was transformed with a YFP construct - something we were trying to avoid. We had hoped to be able to use YFP as a fast screening tool for our successful transformations. Every clone that showed a positive band in the PCR gel was then observed under a fluorescence microscope. At this point we were able to identify our first YFP clones and that the part worked in principle. Returning to the plate reader our now confirmed YFP clones still showed smaller or equal emission values than our WT strain UVM4 under the plate reader. Only after many measurements and trial and error, we were able to identify the correct parameters to measure our YFP. We are aware that for many experts in the field of fluorescence measurement this kind of measurements can be part of the daily routine. Nevertheless, our goal is to expose our mistakes and troubleshooting when measuring YFP in C. reinhardtii to help other young researchers and iGEM teams that might find themselves in a similar situation. The part we used for these measurements can be found here.
Our Mistakes: What You Can Do Better
There are many factors that play an important role when measuring fluorescence. The first important thing to take into account is the physics of the measuring device. Most fluorescence spectrometers have a 90° angle of excitation and emission length. This greatly improves the quality of the measurement by decreasing drastically the mixing of excitation and emission light. Nevertheless, there are spectrometers and other measuring devices, like plate readers, that measure the emission light in the same angle as the excitation light. This leads to the problem that you measure your own excitation light as part of the emission signal and thereby increase the input into the sensor. This can lead to false positive measurements, over excitation of the sensor and many more problems, especially if the emission and excitation wavelengths are close to one another. This was a problem we had to deal a lot with. Having access to a spectrometer or device with a 90° angle of excitation and emission is indeed of great help.
The emission and excitation peaks of mVenus are very close to one another. The excitation peak is at 515 nm and the emission peak at 528 nm (Kremers et al., 2006). This led to the problem that, due to the same angle of excitation and emission light, we could not excite at 515 nm and measure emission at 528 nm. A very relevant factor is the band width of the excitation and emission. A narrow bandwidth allows measurement were excitation and emission are close to one another. A wide bandwidth demands the use of emission and excitation wavelengths that are wider apart. We had to learn this by adjusting our measurements to the bandwidth of our plate reader.
Another important factor when measuring fluorescence in photosynthetic organisms is to take into account the autofluorescence from the chloroplast. So it is crucial to have comparable cell concentrations in your probes. What we hoped to achieve with our YFP was to screen for positive mutants: if fluorescence was detected, it meant that our part was successfully being expressed. Yet, we encountered problems with the cell concentrations when they varied in our assays. For example, a positive clone with a low cell density in the probe would appear to have an equal or lower fluorescence signal than the autofluorescence of a YFP-negative clone with a high cell density. False positives would be measured and positive clones would go unnoticed. By adjusting and equaling the cell concentration of all your probes you can assure a better chance of success when measuring fluorescence emission of your probes. Unfortunately, this sometimes can require an effort that ultimately defeats the purpose of fast and efficient screening, especially when using 96-well plates.
Additionally, regarding the pigments, photo systems and light antennae in C. reinhardtii, it must be taken into account that there are fluorescence quenching effects. The emission light of YFP is absorbed by the pigments, photo systems and light antennae of the algae in the probe, therefore reducing the measured emission.
Methods: Fluorescence intensity measurement
After many measurements to adjust parameters we found the right protocol to measure the fluorescence intensity of YFP in C. reinhardtii. The measuring device we used was the Tecan Plate Reader 200pro. This device has an emission bandwidth of 20 nm and an excitation bandwidth of 9 nm. Due to the proximity of excitation and emission peaks of YFP, it was not possible to excite at 515 nm and measure emission at 528 nm with our device. To avoid problems when measuring YFP you have to take the bandwidth into account. We found that keeping a distance of 30 to 40 nm between excitation and emission wavelength showed the best results for the measurement of YFP.
To measure the fluorescence intensity we excited at 490 nm and added a lag time to the detection of the emission of 5 µs. The emission wavelength we measured was 528 nm: the emission peak of YFP. Having a distance of 38 nm to the excitation wavelength allowed us to surpass the bandwidth problems. Make sure to always have a wild type (WT) comparison for your measurements! Before measuring the fluorescence intensity of the probes we adjusted the optical density (cell concentration) of the WT and YFP probes to be at the same values. This is imperative to be able to compare the fluorescence of WT and the YFP clone.
Fig. 1 shows one of our successful measurements of the fluorescence intensity of a YFP clone in comparison to the autofluorescence of the wild type. We did a sequential dilution of the probes and the fluorescence intensity of the YFP clone steadily showed a higher intensity than the autofluorescence of the WT, proving that our protocol works for the measurement of the fluorescence intensity of YFP in C. reinhardtii.
Methods: Confocal microscopy
We used confocal microscopy for two reasons: At first we searched for a second affirmation that our observed flourescence coressponded to the mVenus expression, secondly we wanted to get a deeper understanding where mVenus is located in the cell. A confocal microscope uses punctual focused laser light of distinct wavelength. This laser scans the entire probe and excitating flurophores within this probe when the right wavelength is matched. This punctal fluorescence emission is then digital rendered into a pciture of the entire probe. The scaning method can be used in different levels of focus creating pictures of different heights, which also can be rendered into a 3D model of the flourescence emission. The process of recording several focus layers can be observed in figure 3 and the resulting 3D model is shown in figure 4.Since the microscope is working with lasers, the excitation bandwith is much more narrow and allows a better differentiation, between auto- and mVenus fluorescence. We compared YFP expression without (Fig. 2, YFP BBa_K2984019) and with the ARS secretion signal combined with the SP20 enhancer (Fig. 2, ARS-YFP-SP20 BBa_K2984030) to the fluorescence properties of the wild type strain UVM 4. To evaluate wether the seen images represented auto or mVenus fluorescence, the probes where excited with light of the wavelength of 655 nm (autofluorescence of chlorophyll) and 508 nm (mVenus fluorescence). Pictures for each wavelength were taken separately and then compared in an overlay (figure 2). The first coloumn in figure 2 display the pictures taken under excitation of 655 nm. These measurements serve as base for a comparison. It can bee seen that the algae structure in the first picture of the first row, does not vary much between the pictures of the first row. This structure is the result of the chlorophyll fluorescence. A different behavior can be observed in the second row of figure 2. The first picture is quite comparable to the first one of row 1, but a stark contrast is present between the second picture of row one and two.
In the second picture of the second row, a bright dot is visible in the center of the depicted algae structure emphasized by the white arrow. This area is made up of the cytosol where protein expression takes place, verifying the expression of our mVenus construct. The second row does show a different situation. In this case, a bright dot is visible in the second and third picture of the third row. This implies a translocation of mVenus protein inside, which could verify the functionality of the arylsulfatase secretion signal.
Methods: Fluorescence Emission Scan
Measuring the emission spectrum of YFP in our algae turned out to be one of our main challenges. The presence of chlorophyll was a hindering factor due to its own fluorescence and strong absorption. And the main problem while measuring the emission spectrum of our YFP-expressing clones was the bandwidth of our plate reader and the fact that we kept measuring our own excitation pulse in the spectrum. Ideally, we would have liked to measure the emission spectrum for the optimal excitation wavelength of mVenus: 515 nm. But the measurements repeatedly showed us that the excitation was too close to the emission and interfered with our measurements. Through trial and error we discovered that the optimal excitation wavelength for this measurement was 485 nm. This allowed us to scan the emission between 490 nm and 590 nm and record the spectrum of our probes. The excitation pulse is still measured and leads to an over stimulation of the detector up to a wavelength of around 515 nm, as can be seen on Fig. 4 and 5. Nevertheless, by recording the spectra of WT and a YFP-expressing clone we were able to make a difference spectrum, where the over-stimulated peak disappears and the fluorescence spectrum of YFP is revealed (Fig. 3).
Methods: Time Resolved Measurements
Time resolved measurements are of great interest to examine dynamic processes in cells. Once we had achieved measurements of the fluorescence intensity and fluorescence spectrum of our YFP clones, we wanted to do time resolved measurements of the fluorescence intensity in our algae. Now that we knew how to measure the fluorescence intensity of our YFP-expressing clones (Fig. 1), we could use the same protocol to measure time resolved fluorescence. The main objective of these measurements was the characterization of the promoter PsaD, which we used in most of our genetic constructs during the iGEM competition. How is the expression controlled over a light-dark cycle? One of the advantages when working with C. reinhardtii is the easy synchronization of the cell cycle of the cell culture by a light-dark regime. We did a fluorescence intensity measurement every 20 minutes for approximately 40 hours. It is important to shake the plate with algae before each measurement to avoid the aggregation of algae on the bottom of the wells, which ultimately affects the measurement of the YFP. During our time resolved measurement we observed a light-activated expression of YFP controlled under the PsaD promoter.
We wanted to quantify the light induction of PsaD by alternating between dark phases and light phases, which we successfully achieved, as can be seen in Fig. 7.
Sources
- Kremers, G. J., Goedhart, J., van Munster, E. B., & Gadella, T. W. (2006). Cyan and yellow super fluorescent proteins with improved brightness, protein folding, and FRET Förster radius. Biochemistry, 45(21), 6570-6580.