Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 7 (8), e43942

Programmable Illumination and High-Speed, Multi-Wavelength, Confocal Microscopy Using a Digital Micromirror

Affiliations

Programmable Illumination and High-Speed, Multi-Wavelength, Confocal Microscopy Using a Digital Micromirror

Franck P Martial et al. PLoS One.

Abstract

Confocal microscopy is routinely used for high-resolution fluorescence imaging of biological specimens. Most standard confocal systems scan a laser across a specimen and collect emitted light passing through a single pinhole to produce an optical section of the sample. Sequential scanning on a point-by-point basis limits the speed of image acquisition and even the fastest commercial instruments struggle to resolve the temporal dynamics of rapid cellular events such as calcium signals. Various approaches have been introduced that increase the speed of confocal imaging. Nipkov disk microscopes, for example, use arrays of pinholes or slits on a spinning disk to achieve parallel scanning which significantly increases the speed of acquisition. Here we report the development of a microscope module that utilises a digital micromirror device as a spatial light modulator to provide programmable confocal optical sectioning with a single camera, at high spatial and axial resolution at speeds limited by the frame rate of the camera. The digital micromirror acts as a solid state Nipkov disk but with the added ability to change the pinholes size and separation and to control the light intensity on a mirror-by-mirror basis. The use of an arrangement of concave and convex mirrors in the emission pathway instead of lenses overcomes the astigmatism inherent with DMD devices, increases light collection efficiency and ensures image collection is achromatic so that images are perfectly aligned at different wavelengths. Combined with non-laser light sources, this allows low cost, high-speed, multi-wavelength image acquisition without the need for complex wavelength-dependent image alignment. The micromirror can also be used for programmable illumination allowing spatially defined photoactivation of fluorescent proteins. We demonstrate the use of this system for high-speed calcium imaging using both a single wavelength calcium indicator and a genetically encoded, ratiometric, calcium sensor.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1
Figure 1. Schematic diagrams of the DMD confocal optical pathway and the mirror arrangement for pinhole formation and scanning.
A. Light from the excitation source was expanded and collimated and sent to the DMD via a dichroic mirror. Mirrors in the “on” position on the DMD send the excitation light through a tube lens to the microscope objective and the specimen. Light emitted from the specimen travels back along the same path but it then passes through the dichroic mirror (Dchr). An emission filter (Em) ensures that only the desired excitation wavelength can reach the camera. A modified Offner Triplet arrangement consisting of three separate curved mirrors (CCv1 and CCv2 concave mirrors, r = 207; Cvx convex mirror, r = 103.5) and a single plain mirror (Pm) were then used to form an image on the CCD of an EM camera positioned at a second conjugate image plane. The double-headed arrow next to CCv2 illustrates the approximate plane of movement that allows a change in the magnification of this optical relay. B. A scanning unit consists of n×n micro mirrors in the “on” position, acting as a pinhole, in a square of p×p pinholes. Full scan of the unit is obtained by presenting a series of p2 scanning units with the pinhole in different positions. Different values of n and p define different configurations, referred to as “n×p”, and give different levels of confocality. The figure illustrates a 2×4 confocal configuration.
Figure 2
Figure 2. The effect of pinhole size on image contrast and resolution.
Images were taken through the axial centre of 6 um beads using pinhole sizes ranging from 22 mirrors to 102 mirrors. A constant proportional separation was used such that each pinhole was separated in x and y dimensions by a distance of 6 pinholes. Line profile measurements through the centre of the beads for each configuration are plotted below. The dashed horizontal lines illustrate the position from where the line profile measurements were taken.
Figure 3
Figure 3. Characterisation of axial resolution.
A. Maximum projection of wide-field images of a 4 µm bead taken over an axial distance of 120 µm at 0.5 µm intervals. Below is an axial projection from a frontal plane. B. Images taken of the same bead using a 4×6 pinhole configuration and processed in the same way. Axial projections were made for a range of different pinhole configurations and line profiles drawn through the centre of the bead. C. Axial line profiles comparing 4 different pinhole configurations with images taken in wide-field are shown. Data were fitted with a Gaussian curve to estimate the axial height of the bead. Figures D–F illustrate the relationship between axial resolution, pinhole size and pinhole separation.
Figure 4
Figure 4. The relationship between pinhole size and separation on output power measured at the specimen.
A. The average power at the specimen plane was measured for each pinhole configuration. Separations of 4, 6 and 8 pinhole widths are shown. B. The average power measured at the specimen plane for a 32 pinhole at different separations is shown. Superimposed is an exponential fit of the data.
Figure 5
Figure 5. Examples of wide-field and confocal images of biological specimens.
A. Wide-field (left) and confocal images (right) of cucurbita pollen grains. Excitation wavelengths were 470±nm and 560±20 nm and emission wavelengths of 500–540 nm and 600–670 nm were collected. Images represent maximum projections of stacks of images taken in the axial plane. B. Wide-field and confocal images of a snail neuron filled with Alexa Fluor 568. A cyan look up table was used for clarity. C. A comparison of cucurbita maximum projections captured with the DMD confocal (top) and Leica SP2 CLSM (bottom) with 60 and 63× objectives with N.A of 1 and 0.9 respectively D. Axial profiles of 4 µm beads measured with the DMD confocal with a 4×8 pinhole configuration and Leica SP2 CLSM with an optimal pinhole setting of 1 airy unit.
Figure 6
Figure 6. High-speed calcium imaging.
A Purkinje cell was filled with the high affinity calcium indicator Oregon green BAPTA-1 and voltage clamped at −70 mV. Regions of interest were position over somatic and dendritic regions as shown in panel A. The cell was then pre-hyperpolarised to −110 mV, stepped to −40 mV and then to 0 mV to activate low- and high-voltage activated calcium channels respectively. Images were collected at 30 frames per second in wide-field (upper traces) and confocal (lower traces) modes. Note the differences in peak amplitude and time course for wide-field and confocal modes for each cellular compartment. The holding potential of the cell is shown under the calcium trace. C. Fluorescence measurements were taken from a hippocampal neurone transfected with a ratiometric, genetically encoded calcium indicator named GCaMP2-mCherry. Graphs C–E represent measurements of the GCaMP2, mCherry and ratio signals respectively. 180 pairs of images were collected at 40 ms intervals. The arrows indicate the onset of stimulation with 50 pulses at a rate of 20 Hz.
Figure 7
Figure 7. Photoactivation of mEOS2.
A layer of bacteria expressing mEO2 was photo-activated with 405 nm light. The image on the left shows the resulting loss of green fluorescence where the photoactivation took place. The image on the right shows the appearance of red fluorescence.

Similar articles

See all similar articles

Cited by 7 PubMed Central articles

See all "Cited by" articles

References

    1. Conchello JA, Lichtman JW (2005) Optical sectioning microscopy. NatMethods 2: 920–931. - PubMed
    1. Callamaras N, Parker I (1999) Construction of line-scan confocal microscope for physiological recording. Methods Enzymol 307: 152–169. - PubMed
    1. Nakano A (2002) Spinning-disk confocal microscopy – a cutting-edge tool for imaging of membrane traffic. Cell StructFunct 27: 349–355. - PubMed
    1. Liang M, Stehr R, Krause A (1997) Confocal pattern period in multiple-aperture confocal imaging systems with coherent illumination. OptLett 22: 751–753. - PubMed
    1. Hanley QS, Verveer PJ, Gemkow MJ, rndt-Jovin D, Jovin TM (1999) An optical sectioning programmable array microscope implemented with a digital micromirror device. J Microsc 196 (Pt 3): 317–331. - PubMed

Publication types

MeSH terms

LinkOut - more resources

Feedback