Macroscale fluorescence imaging against autofluorescence under ambient light

Macroscale fluorescence imaging is increasingly used to observe biological samples. However, it may suffer from spectral interferences that originate from ambient light or autofluorescence of the sample or its support. In this manuscript, we built a simple and inexpensive fluorescence macroscope, which has been used to evaluate the performance of Speed OPIOM (Out of Phase Imaging after Optical Modulation), which is a reference-free dynamic contrast protocol, to selectively image reversibly photoswitchable fluorophores as labels against detrimental autofluorescence and ambient light. By tuning the intensity and radial frequency of the modulated illumination to the Speed OPIOM resonance and adopting a phase-sensitive detection scheme that ensures noise rejection, we enhanced the sensitivity and the signal-to-noise ratio for fluorescence detection in blot assays by factors of 50 and 10, respectively, over direct fluorescence observation under constant illumination. Then, we overcame the strong autofluorescence of growth media that are currently used in microbiology and realized multiplexed fluorescence observation of colonies of spectrally similar fluorescent bacteria with a unique configuration of excitation and emission wavelengths. Finally, we easily discriminated fluorescent labels from the autofluorescent and reflective background in labeled leaves, even under the interference of incident light at intensities that are comparable to sunlight. The proposed approach is expected to find multiple applications, from biological assays to outdoor observations, in fluorescence macroimaging.

µL quartz cuvette with 1.5 mm light path (Hellma Optics, Jena, Germany). The absorbance (0.16) at 480 nm yielded 29 µM concentration using ε(480)=37000 M −1 · cm −1 (evaluated after recording the absorbance at 447 nm of a denaturated Dronpa-2 solution in 1 M NaOH by using 44000 M −1 · cm −1 for the molar absorption coefficient of the deprotonated chromophore 6  Mixed bacterial culture Dronpa-2 and Padron were expressed in E. coli DH10B strain. Cells were grown into 2 mL of lysogeny broth (LB) at 37 • C, 220 RPM for 1 h. Cells were platted at low density on LB agar plates, and plates were incubated overnight at 37 • C. Dronpa-2 and Padron single colonies were then separately transfered to LB-ampicillin media and incubated at 37 • C, 220 RPM for 4 h. The pre-cultures were diluted to reach OD(600 nm)=0.6, then mixed at the same concentrations, and were finally plated onto LB agar and incubated at 37 • C.

Plant transformation and growth
Arabidopsis thaliana and Camelina sattiva(cv Celine) were transformed with Arabidopsis floral-dip method and transgenic were selected as described previously. 7 Arabidopsis and Camelina seeds were sown respectively on sucrose-supplemented medium 8 or water-soaked paper and grown for 7 days in a growth chamber under cycles of 16 h light / 8 h dark at 22 • C.

Illumination system
The illumination system aims at increasing the distance between the last optical surface of our device and the sample without significant loss of light intensities delivered to the sample. The optical layout of the illumination system is shown in Fig.S1. Our design integrates one divergent doublet (ACN254-040-A, f = -40 mm, Thorlabs, NJ, US) and two convergent doublets (AC508-  Figure S1: Optical layout of the illumination system. Optical elements denoted from 1 to 4 are given in Supplementary Table S2.

Imaging system
The macroscope aims at observing fluorescence emission from a small area within 4 × 4 mm 2 with both green (525 nm) and red (585 nm) channels. Before being imaged by the objective, the fluorescence emission is first collected by the beam-expanding system used for excitation, through which optical aberrations (especially the chromatic one) are introduced. Our objective consists of three lenses with different glass materials (BK7, SF11) to correct for chromatic aberration. The shape of each lens minimizing the spherical aberration at large NA and off-axis aberrations up to 2 4 mm from the optical axis was optimized using the OSLO software. Commercial lenses shapes and focal lengths at the closest to the optimized elements were chosen and the air space between each element was optimized to yield the final design (Fig. S2), which surpasses the imaging performances of singlet or achromatic doublet lenses ( Fig. S3 and Fig. S4) for generating high quality images (Fig. S5).  ; c,f: Designed lens system. Images taken with green filter ET525/36m and red filter ET585/20m, aperture F/4.0, focusing in the red channel. The system succeeds to improve the sharpness for both channels and corrected the defocus in the green channel. Scale bars: 1 mm. 6

Estimated cost of the Speed OPIOM macroscope
We currently estimate the cost of our imaging macroscope in the lower 10 ke range as follows: camera 0.8 ke , optomechanical parts /filters 7 ke, electronics (Teensy 3.5, LEDs and printed circuits boards) 0.2 ke, computer 1 ke. Figure S6: Dependence of the Pre-OPIOM and OPIOM images, of the signal profiles, and the associated signal-to-noise ratios of the nitrocellulose membrane on the Dronpa-2 amount (in pg) deposited on a 400 µm-diameter blot. In the signal profiles, the error bar measures the standard deviation over three independent measurements. The analyses have been performed along three lines: one crosses the middle of the blot (displayed in red) whereas the two others (shown in green and blue) are located out of it. The images have been recorded at resonance for Dronpa-2 (see also Table S1). One should notice that at low concentrations, the signal-to-noise analysis of the Pre-OPIOM images is no longer reliable since the signal observed at the blotted area is lower than the observed standard deviation of the membrane signal and much similar to the Pre-OPIOM signal observed for the control.     1The radius were set at 6.973 and 6.787 mm for LED emitting light at 405 and 480 nm respectively.