The aim of the Fifth International InterLab measurement study is to determine if we can
reduce lab-to-lab variability related to fluorescence measurements, by normalizing to absolute cell count or colony forming units rather than OD measurements which are highly variable between different labs.
To answer this question two approaches were used. First, the absorbance of dilution of monodisperse silica microspheres were measured. These microbeads have the same size as Esherichia coli cells, which allows the conversion of the absorbance measurements into equivalent concentrations of beads. The second approach consisted in counting the colony forming units
(CFU) in positive and negative control samples allowing the acquirement of a conversion
factor from absorbance to CFU.
Calibration with Ludox, Microspheres and Fluorescein (day 1)
Three sets of measurements calibrations were performed:
First, Absorbance at 600 nm of LUDOX CL-X (45% colloidal silica suspension) was measured to use it as reference in order to acquire a conversion factor which is the result of the ratio between the corrected Absorbance at 600 nm (Arith. Mean absorbance of LUDOX CL-X - Arith. Mean absorbance of water) and a reference A600. When multiplying this factor by the measured A600, comparable A600 values are obtained as they take into account the variations of volumes in the different wells
The second calibration was performed through a serial dilution of “monodisperse silica microspheres”. These microspheres were used since they have similar size and optical properties as cells, and the number of particles per volume is precisely set. We obtained a standard curve for particles (Figure 2) allowing the conversion of the A600 into an approximate number of cells.
Figure 2. Particle standard curve. The A600 of “monodisperse silica microspheres” is plotted against particle count per 100 µL. This plot allowed the conversion of measured absorbance at 600 nm of the cells in suspension to the number of particles (or cells) in 100 µL.
The last calibration was performed through series dilution of fluorescein. The fluorescence standard curve (
Figure 3) obtained, allowed the normalization of the quantity of GFP fluorescence in our transformed cells due to the analogous excitation and emission properties of the two molecules.
Figure 3. Particle standard curve. The particle count per 100 µL is plotted against the A600 of the “monodisperse silica microspheres”. This plot allowed the conversion of measured absorbance at 600 nm of the cells in suspension to the number of particles (or cells) in 100 µL.
Then, E. coli K-12 DH5-α cells were transformed with the different test devices and spread on plates.
Colony-Picking and Inoculation (Day 2)
Figure 4 : Two colonies were picked from the transformation plates and grown overnight at 37°C in LB
Figure 3. Fluorescein standard curve. The fluorescence of fluorescein is plotted against fluorescein concentration (µM). This plot allowed the conversion of the fluorescence measurements of the cell suspension to a GFP concentration.
Cell measurements were done to convert the absorbance of the cell suspensions into the absorbance of a known concentration of beads. Then the Colony-Forming Units (CFU) were determined and allowed to obtain a conversion factor from absorbance to CFU.
Figure 4 : The cultures were diluted to an A600 of 0.02 then their absorbance at 600 nm and fluorescence (excitation: 410 nm, emission: 520 nm, gain= 75) measured at 0 and 6 hours.
The results showed an increase in the fluorescence and the absorbance after 6 hours compared to the values at 0 hours, indicating an increase in the number of cells (Figure 5). Devices 1 and 4 showed the highest fluorescence while the device 3 was within the same range than the negative control (Figure 5A). Devices 2, 5 and 6 displayed fluorescence levels similar to the positive control. However, we saw no significant difference in absorbance amongst the cells expressing the different devices and the controls at t = 0 h and t = 6 h (Figure 5B). This observation confirms that the fluorescein level observed were not due to a difference in cell number but rather due to different expression levels of the GFP reporter of the devices.
Figure 5. Net cell measurements data. A. Net fluorescence measurements (Excitation= 410 nm, Emission= 520 nm) average of the 2 colonies cultures expressing the different devices, sampled at 0 and 6 hours of growth.
B. Net absorbance at 600 nm, average of the 2 colonies cultures expressing the different devices, sampled at 0 and 6 hours of growth.
Cell Measurements and CFU protocols (Day 3)
Figure 6: Scheme of Colony Forming Units (CFU) protocole
In order to know the number of cells in solution with an A600 = 0.1, the overnight cultures of the negative and positive controls were diluted to this A600, and the absorbance was measured in triplicates of each diluted culture. Serial dilutions of this cell suspension were then performed and 3 dilutions were plated on LB medium (+ Chloramphenicol) and incubated overnight at 37°C. The next day, the number of colonies on each plate were counted, and assuming that one colony comes from one original cell, we calculated the Colony Forming Units (CFU) in 1 mL for the original culture (A600 = 0.1). As expected, the number of colonies formed in inversely proportional to the dilution excepted in the second and third replicates of the positive control from colony 1. This is probably due to a mistake during the manipulations.
Table 2. Colony counting data. Ta. Number of colonies after overnight culture of samples with dilution factors of 8x104, 8x105 and 8x106 respectively.
b. Colony Forming Units (CFU) per mL of cell culture, calculated by multiplication of the number of colonies by the respective dilution factor.
Conclusion
Since the purpose of InterLab study relies on the data coming from the different teams, only then would the results be mostly interesting to analyze and discuss. This 2018 InterLab study was a great opportunity for us to learn many skills through some of the challenges we faced while following the protocol. And we hope to have contributed with our results to decrease the lab-to-lab variability
