Limits...
Fast imaging of live organisms with sculpted light sheets.

Chmielewski AK, Kyrsting A, Mahou P, Wayland MT, Muresan L, Evers JF, Kaminski CF - Sci Rep (2015)

Bottom Line: A telescope composed of two electrically tuneable lenses enables us to define thickness and position of the light-sheet independently but accurately within milliseconds, and therefore optimize image quality of the features of interest interactively.This technique proved compatible with confocal line scanning detection, further improving image contrast and resolution.Finally, we determined the effect of light-sheet optimization in the context of scattering tissue, devising procedures for balancing image quality, field of view and acquisition speed.

View Article: PubMed Central - PubMed

Affiliation: 1] Department of Chemical Engineering and Biotechnology, University of Cambridge, Pembroke Street, Cambridge, CB2 3RA, UK [2] Department of Physiology, Development and Neuroscience, University of Cambridge, Downing Street, Cambridge, CB2 3DY, UK.

ABSTRACT
Light-sheet microscopy is an increasingly popular technique in the life sciences due to its fast 3D imaging capability of fluorescent samples with low photo toxicity compared to confocal methods. In this work we present a new, fast, flexible and simple to implement method to optimize the illumination light-sheet to the requirement at hand. A telescope composed of two electrically tuneable lenses enables us to define thickness and position of the light-sheet independently but accurately within milliseconds, and therefore optimize image quality of the features of interest interactively. We demonstrated the practical benefit of this technique by 1) assembling large field of views from tiled single exposure each with individually optimized illumination settings; 2) sculpting the light-sheet to trace complex sample shapes within single exposures. This technique proved compatible with confocal line scanning detection, further improving image contrast and resolution. Finally, we determined the effect of light-sheet optimization in the context of scattering tissue, devising procedures for balancing image quality, field of view and acquisition speed.

No MeSH data available.


Related in: MedlinePlus

Scattering analysis using color-coded contrast variations through a sample (late stage Drosophila embryo with GFP labelled nuclei).(a): XY cross section through a stack combined from low (0.13) and high (0.3) NA images based on local highest contrast (See main text). Above the excitation beams used to obtain the image color-coded to visualize which raw data stack a given part of the compound image comes from. (b): XZ cross-section through only the data obtained using low NA illumination. (c): XZ cross-section through compound image obtained using high NA illumination data only. (d): resolution measure (HLSFR) for each YZ plane (perpendicular to illumination) along the illumination direction into the sample (-X axis). Rainbow colour ‘High NA’ line corresponds to a compound image created from only high NA beams (c) while ‘Low NA’ line is from a raw stack obtained using low NA illumination (b). All scale bars 40 μm. EX arrows indicate excitation beam direction.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4403519&req=5

f4: Scattering analysis using color-coded contrast variations through a sample (late stage Drosophila embryo with GFP labelled nuclei).(a): XY cross section through a stack combined from low (0.13) and high (0.3) NA images based on local highest contrast (See main text). Above the excitation beams used to obtain the image color-coded to visualize which raw data stack a given part of the compound image comes from. (b): XZ cross-section through only the data obtained using low NA illumination. (c): XZ cross-section through compound image obtained using high NA illumination data only. (d): resolution measure (HLSFR) for each YZ plane (perpendicular to illumination) along the illumination direction into the sample (-X axis). Rainbow colour ‘High NA’ line corresponds to a compound image created from only high NA beams (c) while ‘Low NA’ line is from a raw stack obtained using low NA illumination (b). All scale bars 40 μm. EX arrows indicate excitation beam direction.

Mentions: The benefits of strong focusing of the illumination sheet into the sample for high NA illumination is ultimately limited by scattering. Complex biological samples like Drosophila or zebrafish embryos, scatter and distort the illumination light, making it impossible to maintain a tight laser focus at increasing penetration depth. We assessed this by comparing the image quality obtained in stage 10 Drosophila embryos with more developed mesodermal tissue for NA = 0.13 and NA = 0.3 beams, the latter translated axially to 5 different positions to cover the same field of view. We analysed local contrast by dividing the six resulting stacks into sub-volumes measuring 8 × 8 × 5 μm3 and determined RMS contrast for each one (for details refer to methods). To visualize the results we selected the boxes (subvolumes) with highest contrast and combined them into a single stack. The result is shown in Figure 4a, which is an XY cross-section from this stack. Near the top of the image, in grey scale, is shown the beam profile of the NA = 0.13 beam. Below the positions of the 5 axially displaced beam foci for the NA = 0.3 beam are shown in different colour. Six stacks were thus taken in total. The boxes (subvolumes) noted above were colour coded depending on which stack they were taken from (e.g. grey indicates the data was taken from the stack with the 0.13 NA beam, red corresponds to the stack with the 0.3 NA beam focus at the right most position, etc.). Figure 4a reveals that at up to 90 μm penetration depth there is good correspondence between the box colour patterns and corresponding beam focus positions, indicating benefits from high NA illumination. Deeper into the tissue however the colour pattern is lost and the box pattern contains many panels from the low NA illumination indicating that no benefit is gained from high NA illumination. Figure 4b–d quantifies this observation. Using the same raw data (but presented as XZ cross-sections) we created two separate stacks: one obtained using a single 0.13 NA illumination (Figure 4b) and one colour-coded using five 0.3 NA beams (Figure 4c). A Fourier analysis (HLSFR – see Methods) was performed for both stacks to quantify image quality in each YZ plane (perpendicular to the displayed XZ cross-sections) and plotted below. Clearly there is considerable improvement (up to 50%) for the first 90 μm of the higher NA (colour line) illumination over the lower NA (grey line) case but the advantages diminish for deeper penetration depth, consistent with the visual observation in Figure 4a. For the depicted example it is thus sufficient to stitch stacks obtained from 2 high NA beams and 1 from a low NA beam to achieve the same image quality as stitching data from 5 high NA beams. This finding confirms the relevance of optimizing light sheet conditions for the required field of view, image quality and imaging speed. The graph in Figure 4d also indicates a very strong coupling between beam thickness and image quality, with red and yellow peaks matching the respective beam thicknesses. Here, further improvements may be possible by positioning the beam waists at positions intermediate to those shown, i.e. displacements over less than 2 Rayleigh ranges. Naturally this comes with increases in acquisition time.


Fast imaging of live organisms with sculpted light sheets.

Chmielewski AK, Kyrsting A, Mahou P, Wayland MT, Muresan L, Evers JF, Kaminski CF - Sci Rep (2015)

Scattering analysis using color-coded contrast variations through a sample (late stage Drosophila embryo with GFP labelled nuclei).(a): XY cross section through a stack combined from low (0.13) and high (0.3) NA images based on local highest contrast (See main text). Above the excitation beams used to obtain the image color-coded to visualize which raw data stack a given part of the compound image comes from. (b): XZ cross-section through only the data obtained using low NA illumination. (c): XZ cross-section through compound image obtained using high NA illumination data only. (d): resolution measure (HLSFR) for each YZ plane (perpendicular to illumination) along the illumination direction into the sample (-X axis). Rainbow colour ‘High NA’ line corresponds to a compound image created from only high NA beams (c) while ‘Low NA’ line is from a raw stack obtained using low NA illumination (b). All scale bars 40 μm. EX arrows indicate excitation beam direction.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4403519&req=5

f4: Scattering analysis using color-coded contrast variations through a sample (late stage Drosophila embryo with GFP labelled nuclei).(a): XY cross section through a stack combined from low (0.13) and high (0.3) NA images based on local highest contrast (See main text). Above the excitation beams used to obtain the image color-coded to visualize which raw data stack a given part of the compound image comes from. (b): XZ cross-section through only the data obtained using low NA illumination. (c): XZ cross-section through compound image obtained using high NA illumination data only. (d): resolution measure (HLSFR) for each YZ plane (perpendicular to illumination) along the illumination direction into the sample (-X axis). Rainbow colour ‘High NA’ line corresponds to a compound image created from only high NA beams (c) while ‘Low NA’ line is from a raw stack obtained using low NA illumination (b). All scale bars 40 μm. EX arrows indicate excitation beam direction.
Mentions: The benefits of strong focusing of the illumination sheet into the sample for high NA illumination is ultimately limited by scattering. Complex biological samples like Drosophila or zebrafish embryos, scatter and distort the illumination light, making it impossible to maintain a tight laser focus at increasing penetration depth. We assessed this by comparing the image quality obtained in stage 10 Drosophila embryos with more developed mesodermal tissue for NA = 0.13 and NA = 0.3 beams, the latter translated axially to 5 different positions to cover the same field of view. We analysed local contrast by dividing the six resulting stacks into sub-volumes measuring 8 × 8 × 5 μm3 and determined RMS contrast for each one (for details refer to methods). To visualize the results we selected the boxes (subvolumes) with highest contrast and combined them into a single stack. The result is shown in Figure 4a, which is an XY cross-section from this stack. Near the top of the image, in grey scale, is shown the beam profile of the NA = 0.13 beam. Below the positions of the 5 axially displaced beam foci for the NA = 0.3 beam are shown in different colour. Six stacks were thus taken in total. The boxes (subvolumes) noted above were colour coded depending on which stack they were taken from (e.g. grey indicates the data was taken from the stack with the 0.13 NA beam, red corresponds to the stack with the 0.3 NA beam focus at the right most position, etc.). Figure 4a reveals that at up to 90 μm penetration depth there is good correspondence between the box colour patterns and corresponding beam focus positions, indicating benefits from high NA illumination. Deeper into the tissue however the colour pattern is lost and the box pattern contains many panels from the low NA illumination indicating that no benefit is gained from high NA illumination. Figure 4b–d quantifies this observation. Using the same raw data (but presented as XZ cross-sections) we created two separate stacks: one obtained using a single 0.13 NA illumination (Figure 4b) and one colour-coded using five 0.3 NA beams (Figure 4c). A Fourier analysis (HLSFR – see Methods) was performed for both stacks to quantify image quality in each YZ plane (perpendicular to the displayed XZ cross-sections) and plotted below. Clearly there is considerable improvement (up to 50%) for the first 90 μm of the higher NA (colour line) illumination over the lower NA (grey line) case but the advantages diminish for deeper penetration depth, consistent with the visual observation in Figure 4a. For the depicted example it is thus sufficient to stitch stacks obtained from 2 high NA beams and 1 from a low NA beam to achieve the same image quality as stitching data from 5 high NA beams. This finding confirms the relevance of optimizing light sheet conditions for the required field of view, image quality and imaging speed. The graph in Figure 4d also indicates a very strong coupling between beam thickness and image quality, with red and yellow peaks matching the respective beam thicknesses. Here, further improvements may be possible by positioning the beam waists at positions intermediate to those shown, i.e. displacements over less than 2 Rayleigh ranges. Naturally this comes with increases in acquisition time.

Bottom Line: A telescope composed of two electrically tuneable lenses enables us to define thickness and position of the light-sheet independently but accurately within milliseconds, and therefore optimize image quality of the features of interest interactively.This technique proved compatible with confocal line scanning detection, further improving image contrast and resolution.Finally, we determined the effect of light-sheet optimization in the context of scattering tissue, devising procedures for balancing image quality, field of view and acquisition speed.

View Article: PubMed Central - PubMed

Affiliation: 1] Department of Chemical Engineering and Biotechnology, University of Cambridge, Pembroke Street, Cambridge, CB2 3RA, UK [2] Department of Physiology, Development and Neuroscience, University of Cambridge, Downing Street, Cambridge, CB2 3DY, UK.

ABSTRACT
Light-sheet microscopy is an increasingly popular technique in the life sciences due to its fast 3D imaging capability of fluorescent samples with low photo toxicity compared to confocal methods. In this work we present a new, fast, flexible and simple to implement method to optimize the illumination light-sheet to the requirement at hand. A telescope composed of two electrically tuneable lenses enables us to define thickness and position of the light-sheet independently but accurately within milliseconds, and therefore optimize image quality of the features of interest interactively. We demonstrated the practical benefit of this technique by 1) assembling large field of views from tiled single exposure each with individually optimized illumination settings; 2) sculpting the light-sheet to trace complex sample shapes within single exposures. This technique proved compatible with confocal line scanning detection, further improving image contrast and resolution. Finally, we determined the effect of light-sheet optimization in the context of scattering tissue, devising procedures for balancing image quality, field of view and acquisition speed.

No MeSH data available.


Related in: MedlinePlus