Oblique plane microscopy system

Abstract

An oblique plane microscopy system, comprising: a first optical system comprising a first objective lens, the first optical system configured to deliver excitation light in the form of an oblique light sheet to an object and gather fluorescence light from the object in response to the excitation light sheet; an excitation light coupling unit configured to couple the excitation light to an optical path of the first optical system; and at least two synchronised scanning mirrors arranged on the optical path, configured to scan the excitation light across the sample and de-scan the fluorescence light

Description

[0001] OBLIQUE PLANE MICROSCOPY SYSTEM

[0002] TECHNICAL FIELD

[0003] The present invention related to an oblique plane microscopy system and methods of using the same.

[0004] BACKGROUND ART

[0005] Light-sheet fluorescence microscopy (LSFM) is a rapidly advancing set of techniques for biological imaging that offer impressive temporal resolution and low phototoxicity. The underlying principle relies on selectively illuminating a thin slice of the sample around the focal plane of the objective that is used to collect the emitted fluorescence. To achieve this, most implementations of LSFM rely on a second, orthogonal, objective lens to focus the excitation light into a thin plane or sheet. However, this orthogonal geometry imposes strict mechanical constrains that limit the selection of available high-numerical aperture (NA) objectives. More importantly, having two orthogonal high NA lenses in close proximity excludes the use of typical sample mounting methods (e.g. slides, dishes, multi-well plates) and limits potential imaging applications.

[0006] The above shortcomings can be avoided by using a variant of LSFM, termed Oblique Plane Microscopy (0PM), which uses a single, high NA, objective to deliver the light-sheet illumination and collect the emitted fluorescence simultaneously. In this approach, a lightsheet is introduced at an oblique angle, with respect to the primary objective, illuminating a thin slice of the sample that extends above and below the primary objective’s focal plane.

[0007] In known implementations, however, the primary objective can only move within a very limited travel range to maintain good conjugation with a scan mirror and a constant lightsheet angle while scanning, which is critical for 0PM operation. This reduced travel range limits the focus capabilities of the microscope base and can compromise the operation of the various other imaging modalities (e.g. Transmitted Light, Epifluorescence Differential Interference Contrast and components (e.g. oculars, stage top heaters, etc) which enable a wide range of imaging experiments in a user-friendly way that is accessible to non-experts. Depending on the sample and / or mounting method, it is likely that the primary objective of the OPM system will move away from its design position as the user selects a focal plane, disturbing the conjugation with the scan mirror, resulting in 3D volumes with non-uniform illumination and poor signal -to-noise ratio.

[0008] It is an object of the present invention to at least partially address one or more of the above problems.

[0009] SUMMARY OF THE INVENTION

[0010] According to an aspect of the disclosure there is provided an oblique plane microscopy system, comprising: a first optical system comprising a first objective lens, the first optical system configured to deliver excitation light in the form of an oblique light sheet to an object and gather fluorescence light from the object in response to the excitation light sheet; an excitation light coupling unit configured to couple the excitation light to an optical path of the first optical system; and at least two synchronised scanning mirrors arranged on the optical path, configured to scan the excitation light across the sample and de-scan the fluorescence light.

[0011] Optionally, the position of an object-side focal plane of the first objective lens is adjustable.

[0012] Optionally, a synchronisation of the at least two synchronised scanning mirrors is configured to be adjusted based on an adjustment of the position of the object-side focal plane of the first objective lens, such that adjustment the position of the object-side focal plane of the first objective lens does not result in a change of an oblique angle of the light sheet.

[0013] Optionally, the scanning of the at least two synchronised scanning mirrors is configured to be adjusted such that a pivot point of the excitation light beam tracks a pupil plane of the first objective lens as the position of an object-side focal plane of the first objective lens is adjusted.

[0014] Optionally, the at least two synchronised scanning mirrors are arranged substantially at a primary image plane of the first optical system. Optionally, the at least two synchronised mirrors are configured such that the excitation light sheet is translated in the focal plane of the first objective lens over a scanning range of the scanning mirrors.

[0015] Optionally, the at least two synchronised mirrors are configured such that an oblique angle of the excitation light-sheet relative to the focal plane of the first objective lens is substantially the same over a scanning range of the scanning mirrors.

[0016] Optionally, the at least two synchronised mirrors are configured to pivot the excitation light path about a pivot point arranged at a pupil plane of the first objective lens.

[0017] Optionally, the at least two synchronised scanning mirrors are galvanometric mirrors.

[0018] Optionally, the first optical system is a telescopic optical system, comprising at least one lens in addition to the first objective lenses.

[0019] Optionally, the system further comprises a second optical system comprising a second objective lens, the second optical system configured to perform remote focussing; wherein the optical path to which the excitation light coupling unit couples the excitation light is common to the first and second optical systems, and the at least two synchronised scanning mirrors arranged on the common optical pat, between the first and second optical systems.

[0020] Optionally, the second optical system is a telescopic optical system, comprising at least one lens in addition to the second objective lenses.

[0021] Optionally, the second optical system is arranged back-to-back with the first optical system.

[0022] Optionally, the first and second optical systems are arranged to have a substantially 4f configuration. Optionally, the system further comprises a third optical system comprising a third objective lens configured to image the object.

[0023] Optionally, the third optical system further comprises an imaging sensor.

[0024] Optionally, the light path of the third optical system is angled obliquely to the light path of the second optical system, the oblique angle corresponding to the oblique angle of the light sheet.

[0025] Optionally, the third objective lens is a glass-tipped objective lens.

[0026] Optionally, the system further comprises an excitation light source system configured to generate the excitation light forming the oblique light sheet.

[0027] Optionally, the excitation light source system comprises a scanning mirror configured to generate an excitation light sheet from an excitation light beam.

[0028] Optionally, an oblique angle of the light sheet with respect to the imaging plane is between 20 and 40 degrees, preferably about 30 degrees.

[0029] Optionally, the first objective lens has a high numerical aperture.

[0030] Optionally, the numerical aperture is between 1 and 1.4.

[0031] According to a second aspect of the disclosure, there is provided a method of three- dimensional fluorescence microscopy imaging, the method comprising imaging an object using the oblique plane microscopy system of the first aspect.

[0032] Optionally, the method further comprises adjusting the position of an object-side focal plane of the first objective lens and adjusting synchronisation of the at least two synchronised scanning mirrors correspondingly.

[0033] BRIEF DESCRIPTION OF THE DRAWINGS Further features of the disclosure will be described below, by way of non-limiting examples and with reference to the accompanying drawings, in which:

[0034] Fig. 1 shows an example OPM system;

[0035] Fig. 2 shows data relating to remote focusing PSF characterization;

[0036] Fig. 3 shows data relating to remote focusing PSF characterization;

[0037] Fig. 4 shows data relating to remote focusing volume distortion characterization;

[0038] Fig. 5 shows data relating to OPM PSF characterization;

[0039] Fig. 6 shows data relating to scanning unit voltage ratio calibration; and

[0040] Fig. 7 shows image data for a volume in a stage 7 Drosophila egg chamber.

[0041] DETAILED DESCRIPTION

[0042] Fig. la shows an example OPM system according to the disclosure. As shown in Fig. 1, the OPM system comprises a first optical system OBJ 1, TL 1 comprising a first objective lens OBJ 1. The first optical system is configured to deliver excitation light in the form of an oblique light sheet to an object (e.g. a sample) and gather fluorescence light from the object in response to the excitation light sheet. Preferably, the first objective lens has a high numerical aperture, e.g. a numerical aperture between 1 and 1.4. As shown, the first optical system may additionally comprise a tube lens TL 1. As shown, the first optical system may be a telescopic optical system. Further, the first optical system may be provided within a microscope base 1X81, e.g. a commercially available microscope base such as an Olympus 1X81 microscope base. The position of an object-side focal plane of the first objective lens OBJ 1 (e.g. the position of the primary objective focal plane within the sample) may be adjustable, e.g. by a mechanism forming part of the microscope base.

[0043] As shown in Fig. la, the OPM system comprises a second optical system OBJ 2, TL2, comprising a second objective lens OBJ 2. The second optical system is configured to perform remote focussing. As shown, the second optical system may additionally comprise a tube lens TL 2. As shown, the second optical system may be a telescopic optical system. As shown, the second optical system may be arranged back-to-back with the first optical system. The first and second optical systems may be arranged to have a substantially 4f configuration, for example. Preferably, the overall magnification of the first telescopic system comprised of OBJ 1 and TL 1 and the second telescopic system comprised of OBJ 2 and TL 2 should be about equal to the ratio of the refractive indices of the immersion media in the first and second objective lenses, respectively.

[0044] As shown in Fig. la, the OPM system comprises an excitation light coupling unit DM configured to couple the excitation light to a common optical path of the first and second optical systems. The excitation light coupling unit DM may be a dichroic mirror arranged on the common light path, configured to reflect excitation light along the common light path from an excitation light source system configured to generate the excitation light forming the oblique light sheet. As shown, the excitation light coupling unit DM may be arranged on a light path within the second optical system OBJ 2, TL 2, e.g. between the second objective lens OBJ 2 and the second tube lens TL2.

[0045] As shown in Fig. la, the OPM system comprises a third optical system OBJ 3, QBF, TL 3. The third optical system is configured to image the object at an imaging sensor CAMERA. As shown, the third optical system comprises a third objective lens OBJ 3 and a third tube lens TL 3. As shown, between the third objective lens OBJ 3 and the third tube lens TL 3, an optical filter QBF may be provided, such as a quadband emission filter or other single or multiband filter. As shown, the light path of the third optical system is angled obliquely to the light path of the second optical system. The oblique angle corresponds to the oblique angle of the light sheet relative to the imaging axis. As shown, the third objective lens may be a glass-tipped objective lens.

[0046] Preferably, the angle of the light sheet with respect to the imaging plane is between 20 and 40 degrees, more preferably about 30 degrees, e.g. 28 degrees.

[0047] As shown in Fig. la, the OPM system comprises two synchronised scanning mirrors Gl, G2 arranged on the common optical path, between the first and second optical systems. The two synchronised scanning mirrors Gl, G2 are configured to scan the excitation light across the sample and de-scan the fluorescence light. In other examples, more than two synchronised scanning mirrors may be provided. As shown, the synchronised scanning mirrors Gl, G2 are arranged substantially at a primary image plane of the first optical system OBJ 1, TL 1. Preferably, the synchronised scanning mirrors Gl, G2 are galvanometric mirrors. Fig. lb schematically shows a scanning of an image across the intermediate image plane of the microscope base by the synchronised scanning mirrors Gl, G2. As shown, the synchronised mirrors Gl, G2 are configured such that the excitation light sheet is translated in the focal plane of the first objective lens over a scanning range of the scanning mirrors Gl, G2. Preferably, an oblique angle of the excitation light-sheet relative to the focal plane of the first objective lens is substantially the same over the scanning range of the scanning mirrors Gl, G2.

[0048] Fig. 6 shows the synchronisation of the scanning mirrors Gl, G2 in further detail. As shown in Fig. 6a, the synchronised mirrors may pivot the excitation light path about a pivot point during scanning. As shown in the central image, preferably, the pivot point is arranged at a pupil plane of the first objective lens OBJ 1.

[0049] As shown in Fig. 6, the synchronisation of the synchronised scanning mirrors Gl, G2 is configured to be adjusted. Preferably, the adjustment is based on an adjustment of the position of the object-side focal plane of the first objective lens OBJ 1, such that adjustment the position of the object-side focal plane of the first objective lens does not result in a change of an oblique angle of the light sheet. Preferably, the scanning of the synchronised scanning mirrors Gl, G2 is configured to be adjusted such that a pivot point of the excitation light beam tracks a pupil plane of the first objective lens OBJ 1 as the position of an object-side focal plane of the first objective lens OBJ 1 is adjusted.

[0050] As shown in Fig. la, the OPM system comprises an excitation light source generating system. The excitation light source generating system is configured to generate the excitation light forming the oblique light sheet. As shown, the excitation light source generating system may comprise a plurality of laser light sources. As shown, the excitation light from the laser light sources may be combined via one or more dichroic mirrors. The excitation light is collimated and formed into a light sheet by an optical system LI, L2, CY 1, CY 2, CY 3. The excitation light may be directed to the excitation light coupling unit DM by a mirror Ml, as shown.

[0051] Further details of a specific example OPM system, namely an OPM microscope, according to the disclosure are provided below, described with reference to work carried out by the inventors. The terms “we”, “our”, or similar refer to the inventors. Method steps described below refer to the work carried out by the inventors to put an OPM system according the disclosure into practice and test its performance.

[0052] 2. Experimental methods

[0053] 2.1 Optical setup

[0054] The OPM microscope comprises three telescopes as shown in Fig. 1. The first telescope resides in the Olympus 1X81 microscope base and is comprised of an Olympus 100X, 1.35NA silicone oil objective (UPLSAPOIOOXS) (OBJ 1, Fig. 1) and a 180mm tube lens (TL1).

[0055] The second telescope is arranged back-to-back with the first telescope and forms an aberration free 3D image of the sample space in the remote volume superintended by the objective lens of this second telescope (OBJ 2, Fig. 1), an Olympus 40X, 0.95NAair objective (UPLXAPO40X).

[0056] To optimise the performance of the remote focusing system, the overall magnification between the primary objective (OBJ 1) space and the remote focusing objective (OBJ 2) space should equal the ratio of the refractive indices of the immersion media in the primary and remote focusing objective spaces, respectively. Given that our choice for the primary and remote focusing objectives is an Olympus 100X Silicone oil (n=1.406), and an Olympus 40X Air objective, respectively, the ideal focal length for the tube lens of the remote focusing objective (TL2, Fig. 1) should be 320mm. As no 320mm focal length lenses are commercially available, we adopted a tube lens assembly that consists of two commercially available achromatic lenses (Thorlabs, ACT508-500-A-ML, and ACT508-750-A-ML) with an effective focal length of 321mm. Finally, the third telescope is tilted at 28°with respect to the optical axis of the remote focusing objective (OBJ 2) and employs a glass tipped objective lens (ASI, AMS-AGY, 54-10-5, vl. l) (OBJ 3, Fig. 1). This tertiary objective lens is mounted on a linear translation stage (Newport, 9066-X-M) driven by a motorized adjuster (Thorlabs, PIA13) to relay fluorescence signal from individual planes within the remote focusing volume through a quad-band emission filter (Chroma, ZET405 / 488 / 561 / 647m) to an sCMOS camera (Hamamatsu, Orca-Flash4.0v3) with the help of a 200mm tube lens (Thorlabs, TTL200-A). Light sheet excitation is provided by four excitation lasers at 405 (Coherent, Obis 405nmLX), 488 (Coherent, Obis 488nmLS), 560 (MPB Communications, 2RU-VFL-P-2000-560-B1R) and 642nm (MPB Communications, 2RU-VFL-P-2000-642-B1R), all coupled into a single mode optical fibre (Thorlabs, PM-S405-XPCustom). Laser light emitted from the fibre is collimated with an f = 30mm achromat (Thorlabs, AC254-030-A) and passes through an adjustable slit (Thorlabs, VA100CP) to control the beam size in one dimension and, subsequently, the light sheet NA. Next, collimated light reflects from the surface of a 16 kHz resonance mirror (EOPC, SC-30). The mirror surface is imaged onto the sample plane by downstream optics for the purpose of pivoting the light sheet, in plane, to de-shadow during imaging. After reflecting from the 16kHz resonance mirror, the excitation beam passes through three cylindrical lenses. The first (f=25mm, Edmund Optics, 69-717) and last (f=100mm, Thorlabs, LJ1567RM-A) cylindrical lenses act on one dimension of the excitation beam as a telescope and expand the beam diameter, only in that dimension, by a factor of 4 and set the width of the light sheet upon exiting the primary objective lens at the sample. The middle cylindrical lens (f=150mm, Thorlabs, LJ1629RM-A) is oriented orthogonally to the first and third and is responsible for creating the light sheet. After the set of cylindrical lenses, an adjustable iris further controls the extent of the light sheet (width) at the sample plane to match the desired field of view (FOV). These elements are all mounted on a translating platform whose position is adjusted to displace the excitation light in the primary objective back aperture and set the final light sheet angle at the sample.

[0057] The excitation beam reflects from a mirror that is conjugated to the primary objective back focal plane. This mirror allows the position of the light sheet to be adjusted, at the sample plane, without changing the angle. Next, the excitation beam reflects from a quad-band dichroic mirror (Chroma, ZT405 / 488 / 561 / 640rpcv2), passes through the 321mm tube lens assembly (Fig. 1, TL2), and reaches two galvo mirrors (Thorlabs, QS30X-AG). Both galvo mirrors are connected to separate analogue outputs from an FPGA based data acquisition (DAQ) board (National Instruments, PCIe-7852R) and are actuated synchronously to scan the excitation light sheet across the sample and de-scan the corresponding fluorescence signal. Microscope control and data collection are carried out within the Lab View environment using custom control software.

[0058] In this configuration the pair of galvo mirrors can support an image size of 20mm in diameter and a scan range of ±10mm at the primary image plane of the microscope base (Fig. 1(c)). For 100X magnification this translates to an area in the sample space of 200 pm in diameter over a scan range of±100 pm which extends beyond the supported FOV of the 100X silicone oil objective used in this work. The field curvature introduced by the scanning motion of the mirror results in an axial focal plane shift up to 0.92mm and, because this scales with the square of the magnification in sample space that amounts to 92nm of focal shift across the 200 pm FOV, which is below the depth of field of the primary objective.

[0059] The parameters of the light-sheet illumination were estimated from its NA using the software provided by Remacha et al.

[0021] , To measure the NA of the light-sheet illumination, we mounted a board level camera (Edmund Optics 84-933) on the objective turret with the camera sensor at the expected position of the primary objective back focal plane and measured the size of the excitation profile for different openings of the adjustable slit (API, Fig. 1).

[0060] 2.2. Remote focusing volume characterization

[0061] To evaluate the remote focusing performance, we used lOOnm green / yellow beads (Thermo Fisher, F8803) immobilized with Poly-L-lysine (Sigma, P4707-50mL) on a No 1.5 cover glass bottomed 8 well dish (ibidi, 80807) and immersed in silicone oil. To achieve uniform illumination of the beads across the FOV we used the epifluorescence module of the Olympus microscope base which was equipped with an Olympus filter cube (Olympus, U-MWIBA3) which selected the spectrum for green excitation of a mercury arc lamp (X-Cite, EXFO120).

[0062] We mounted the remote focusing objective (OBJ 2) on a linear stage (Thorlabs, XR25P / M) that allowed displacement along its optical axis ±5mm away from its design position. This enabled us to systematically characterize how the performance of the remote focusing system is affected by axial misalignments between the primary and the remote focusing objective and experimentally confirm the optimal axial position of the latter.

[0063] We also replaced the tilted third telescope downstream of the remote focusing objective with one comprised of an Olympus 60X, 0.9NA air objective (UPLFLN) and a 150mm tube lens (Thorlabs, AC254-150-A-ML) set on a straight optical path (Fig. 1(b)). This configuration simplifies data analysis as the mapping between the primary and remote focusing volumes can be readily described in cartesian coordinates without any transformations.

[0064] Finally, we mounted the Olympus 60X air objective lens on a long travel range piezo walk stage capable of 14mm travel with nanometer step sizes (Physik Instrumente, N-664.3A). In this way, we could probe different parts of the remote volume with high resolution regardless of the axial position of the remote focusing objective and collect bead Z-stacks for PSF measurements.

[0065] 2.2.1. PSF measurements

[0066] To measure the PSF we mounted the bead sample on the microscope’s XYZ stage (Applied Scientific Instrumentation MS-2000XYZ) with the Z piezo set at the centre of its range of motion. Then, the beads were epi-illuminated and brought into focus using the eye pieces and adjusting the focus knob (objective Z position) of the microscope base. We consider this the design or reference position of the primary objective and we kept the primary objective at this position for all PSF measurements. Subsequently, we used the Z piezo of the microscope’s stage to position the beads at 8 different depths away from the focal plane of the primary objective at Z = -40, -30, -20, -10, 10, 20, 30, 40pm. For each Z position of the bead sample in the primary objective space, a bead image was formed at a corresponding depth in the remote volume. We could then probe these parts of the remote volume with the help of the third telescope and collect a bead image stack by scanning the tertiary objective with 54 steps and a 150 nm step size. To measure the axial and lateral FWHMs of the PSF we used custom Matlab scripts. First, the frame with the best overall focus was determined based on the second derivate method presented by Pech-Pachecoet al., candidate bead locations were identified using a wavelet transform algorithm as described in Y. Zhang et al. and fit with a 2D Gaussian to determine the XY position for each bead across the FOV [22,23],

[0067] Subsequently, the Z profile for each bead at its fitted XY position was extracted from all the images in the stack and fit with a IDGaussian to determine the axial (Z) FWHM of the PSF and to identify the best focal plane for each bead. Finally, a second 2D Gaussian fit was performed at the best focal plane of each bead to determine the lateral (X, Y) PSF FWHM. 2.2.2. Remote volume distortion due to axial misalignment

[0068] To investigate how the axial misalignment between the primary and remote focusing objective affects the PSF and the remote focusing volume we kept the primary objective lens fixed at its design position and translated the remote focusing objective lens in 1mm steps to cover a total range of ± 5mm around its design position, which was determined from a Zemax model of the microscope. At each new axial position of the remote focusing objective lens, we performed the PSF analysis routine described in Section 2.2.1 to determine the PSF lateral and axial FWHM and the position of each bead, at different depths of the remote focusing volume.

[0069] We observed that as the axial misalignment between the primary and remote focusing objective lenses was increased, bead positions appeared to radially shift further from the image centre with increasing remote focusing depth. Furthermore, the amount of radial shift was observed to be proportional to the distance from the image centre.

[0070] To quantify this radial shift at each remote focusing depth, we converted the position of each bead into polar coordinates (r, 9) and found the corresponding bead with the same 9 value and smallest r separation in the centre of the remote focusing volume, where the bead sample is positioned at the focal plane of the primary objective (Z=9). Subsequently, the radial bead displacement of all beads in the FOV was plotted as a function of the distance from the image centre and fit with a linear function to determine the rate of radial displacement at that depth of the remote focusing volume.

[0071] 2.3. Scan mirror voltage calibration

[0072] To determine the optimal voltage ratio between the scanning mirrors as a function of the primary objective Z position, we used 1 9nm diameter beads (Thermo Fisher, F8893) embedded in a 2% agarose gel. The primary objective was initially set to the lowest z position (9mm) allowed by the microscope base and the 3Dbead sample was brought into focus using a custom Z stage insert that allowed the sample to be translated over a 25mm Z range using a micrometre driven linear stage (Thorlabs, XR25C / M). We used a light-sheet with NA=O.18 to illuminate an oblique FOV with dimensions of 98 pm x 16pm (Y axis, oblique axis), set the voltage ratio between the scanning mirrors to a starting estimate and performed 100 steps to scan a total length of 140 pm along the X axis. For each step of the scanning mirrors, we collected an image and summed the intensity of all the pixels to obtain the summed intensity value. Next, we normalized this intensity value to the average intensity of all the 100 images collected at each mirror step, plotted the normalized intensity values as a function of step number and fit with a linear function to obtain the slope for the new voltage ratio.

[0073] Subsequently, we changed the voltage ratio between the scanning mirrors and imaged the same volume again to obtain the slope for the new voltage ratio. In total we investigated 15 different voltage ratios from 1.013 to 1.041 and, by plotting the corresponding slope values and performing a linear fit we predicted the voltage ratio that resulted in a slope of zero for the lowest primary objective position.

[0074] The primary objective lens was then moved up by +1 mm and the above procedure was repeated to obtain the voltage ratio that produces a zero slope for the new objective position. We investigated a total of 10 primary objective lens Z positions, with a step size of 1mm, spanning the entire range of motion offered by the microscope base.

[0075] 2.4. Microscope characterization oblique path

[0076] To measure the PSF in the oblique path we mounted the bead sample on the microscope’s XYZ stage with the Z piezo set at the centre of its range of motion. The beads were epiilluminated and brought into focus using the eye pieces and adjusting the focus knob of the microscope base to bring the primary objective to its design position. We then switched to the 0PM modality, used a 488nmexcitation and a light-sheet with an NA of ~0.1, and collected a volume of 77 pm x 90pm x 26pm (X, Y, Z) by scanning the mirrors with a 117nm step size along the X direction.

[0077] Subsequently, we used the microscope stage’s Z piezo to position the beads at 4 different depths away from the focal plane of the primary objective at Z= -10, -5, 5, 10pm. For each Z position of the bead sample, we imaged the same volume under identical imaging conditions. To analyse the beads and obtain the lateral and axial FWHMs we first deskewed the raw volumes to convert to cartesian coordinates and then followed the same analysis steps described in section 2.2.1. To obtain a summary image, we took an XZ section of each volume and max projected them into a single image.

[0078] 2.5. Drosophila sample preparation

[0079] Drosophila egg chambers were prepared as previously described with minor modifications

[0024] ,

[0080] Briefly, flies expressing the microtubule-associated protein, Jupiter, endogenously tagged with eGFP were fattened on dry yeast at 25°C for one day before dissection

[0025] , Their ovaries were dissected at room temperature in a drop of Voltalef oil IOS (VWR, 24627.188) on a cover glass (22x50, thickness No. 1.5, VWR, 631-0138) and individual egg chambers were isolated and positioned at the centre of the coverslip.

[0081] 3. Results and discussion

[0082] 3.1. Remote focusing volume PSF characterization

[0083] 3.1.1. Axially aligned primary and remote focusing objectives

[0084] To characterize the performance of the remote focusing system we worked on the straight optical path (Fig. 1(b)). In this configuration both the primary and remote focusing objective volumes can be intuitively mapped in cartesian coordinates without any postprocessing. We began by positioning the fluorescent beads at the focal plane of the primary objective and measuring their corresponding lateral and axial FWHM across the available FOV. As expected, beads closer to the centre of the FOV exhibited slightly improved FWHMs (255 ± 10 nm laterally and 688 ± 38nm axially) compared to the edges (272 ± 14nm laterally and 764 ± 49nm axially) (Fig. 2(a)). But, on average, throughout a circular area of 200 pm in diameter we obtained a relatively uniform performance with FWHMs of 264 ± 15 and 730 ± 49 nm (mean ± standard deviation, n=974) for the lateral and axial dimensions respectively. To provide some context for the above measurements, we imaged the same beads with the Epifluorescence modality of the Olympus microscope base via the camera port on the eyepiece adaptor, and obtained 238 ± 3 nm and 593 ± 30 nm (mean ± standard deviation, n=795) for the lateral and axial FWHMs respectively. Given that in the epifluorescence configuration only the primary objective and its tube lens inside the microscope base affect the size and shape of the PSF, we reasoned that the epifluorescence values for the axial and lateral FWHMs would provide a good experimental reference for the best performance that can be expected from the remote focusing system. Therefore, the good agreement between the axial and lateral FWHMs measured with the remote focusing system and the epifluorescence module indicate that minimal aberrations were introduced by the additional optical components of the remote focusing system, which led to an increase of 11% in the lateral and 23%in the axial size of the PSF.

[0085] To study how the PSF changes throughout the remote focusing volume, we kept the primary objective fixed at its designed position while the bead sample Z position was discretely varied from -40 to +40 pm in 10 pm steps. Bead stacks were collected by scanning the tertiary objective as described in section 2.2.1. Representative PSF cross sections are shown in Fig. 2(b). The PSF of the system retained its overall shape over the investigated depth range indicating that no significant aberrations were introduced by the remote focusing system. However, we noticed an increase in the lateral size and an elongation along the axial direction of the PSF, which became more pronounced with distance from the image centre and remote focusing depth. These changes on the size of the PSF indicate the presence of residual spherical aberrations that are not eliminated by the remote focusing system, possibly due to deviations of the magnification between the object space and the remote focusing volume from its optimal value

[0026] , Nonetheless, the size of the PSF does not change more than ~20% within a 200 pm diameter FOV and delivers better performance at depths closer to the primary objective focal plane (Fig. 2(c)).

[0086] 3.1.2. Axially misaligned primary and remote focusing objectives

[0087] A key incentive for using a microscope base is the ability to navigate the sample and focus on the plane of interest. This is typically achieved by moving the primary objective axially. Although this type of motion does not affect standard microscope operation it may disturb the axial alignment between the primary and remote focusing objectives. To investigate how axial misalignment between the primary and remote focusing objectives would impact the PSF, we fixed the primary objective at its nominal position and mounted the remote focusing objective on a translation stage that supported a large axial displacement range of ± 5mm as described in section 2.2.2. Given that the primary objective has a working distance of 200 pm and this system is built around an 1X81 base with a microscope stage equipped with inserts for slides and dishes, the deviations from the nominal primary objective position during routine operation are not expected to exceed ± 1mm. Therefore, displacing the remote focusing lens by ± 5mm already covers axial misalignments between the primary and remote focusing objectives introduced by the focusing motion of the primary objective. Average X, Y and ZFWHM values for each sample Z position and remote focusing objective position are presented in Fig. 3(a). Despite axially misaligning the remote focusing objective by ± 5mm, the PSF maintained its overall shape, and only for axial misalignments larger than ± 2.5mm and depths larger than 30pm away from the focal plane, we observed a more pronounced increase in the lateral size of the PSF and an elongation along its axial dimension, Fig. 3(b). Although, these changes in the PSF size indicate the presence of more pronounced spherical aberrations and consequently a reduction in the volume over which the remote focusing system exhibits good performance, the PSF does not seem to be very sensitive to axial misalignments between the primary and remote focusing objectives.

[0088] For axial misalignments up to 2.5mm and volumes extending ± 10pm away from the focal plane of the primary objective an increase of <3% in the lateral and <11% in the axial size of the PSF was observed. Additionally, the lateral and axial FWHMs of the PSF exhibited a <20% increase in size with distance from the image centre across a 200 pm diameter FOV (Fig- 3(c)).

[0089] 3.2. Remote focusing volume distortion characterization

[0090] Another effect of the axial misalignment between the primary and remote focusing objectives is a distortion of the remote focusing volume, which is perceived as a change in magnification with sample Z-depth [27,28], To investigate the amount of this distortion in our system, we used the same sample and experimental procedure, as above for the PSF measurements, but in this case for each axial position of the remote focusing objective, we first identified all the beads in an image recorded with the sample positioned at the Z=0 focal plane of the primary objective, and subsequently we tracked the beads through the rest of the images that were recorded above and below the focal plane of the primary objective as described in section 2.2.2. We then measured the bead’s lateral displacements relative to their position at the focal plane and calculated the rate of radial separation, which provides a good description of the changes in magnification introduced by the remote focusing system at different depths of the remote focusing volume and for different degrees of axial misalignment between the primary and remote focusing objective. The results are summarized in Fig. 4(a). The magnification remains relatively constant throughout the remote focusing volume for axial positions of the remote focusing objective near its design position. As the remote focusing objective moves to axial positions further above its design position we observe increasing levels of distortion in the remote focusing volume with the apparent magnification increasing for planes above and decreasing for planes below the focal plane and vice versa for axial positions below the design position. The design position of the remote focusing objective was estimated from a Zemax model of the microscope, but without access to an optical model for the Olympus objective and tube lens and a CAD model for the 1X81 base. Given these limitations, we considered the design position of the remote focusing objective as a good initial guess and proceeded to determine its optimal axial position experimentally. By plotting the rate of the magnification change versus axial position and performing a linear fit we obtained the optimal axial position of the remote focusing objective lens at ~0.4mm away from its initial design position (Fig. 4(b)).

[0091] For our applications, where we will operate the OPM microscope at a tilt angle of ~30° and employ light sheets with lengths up to ~40 pm, the working depth of the primary objective space is not expected to exceed ±10 pm [7,21], Furthermore, given that the primary and remote focusing objectives are conjugated via their corresponding tube lenses with focal lengths of 180 and 321mm, respectively, the amount of misalignment introduced during routine microscope operation, where the primary objective could be displaced by up to ±lmm from its designed position while the remote focusing objective remains fixed is equivalent to the scenario where the primary objective is fixed and the remote focusing objective is displaced by 3 mm. Consequently, under routine microscope operation, the PSF of the system will not be significantly affected and distortions of the remote focusing volume are expected to remain at levels below 0.5% (Fig. 4(a)).

[0092] 3.3. Microscope characterization, oblique path

[0093] Next, we moved to the oblique path by inserting, downstream of the remote focusing objective, a third telescope comprised of the glass-tipped objective and a 200mm tube lens (Fig- 1(a)).

[0094] The optimum choice of illumination angle and, consequently, the tilt of this third telescope with respect to the focal plane of the remote focusing objective has been discussed in detail by Millet-Sikking et al., where it was shown that for an 1.35 NA silicone oil objective a ~30° tilt offers a good balance between optical sectioning and size of illuminated volume [6,7], In our system the tilt angle was 28 degrees.

[0095] To characterize the performance of the overall system we first imaged lOOnm yellow beads as described in section 2.4 (Fig. 5(a)). Near the focal plane of the primary objective, we measured average PSFFWHMs of ~290nm, ~274nm and ~712nm in the X, Y and Z dimensions respectively (Fig. 5(b), c). The anisotropy in the FWHM values between the X and Y dimension in our measurements is attributed to light clipping by the tertiary objective along the X dimension and is typical for 0PM systems [7,10,29], Nonetheless, the overall size of the PSF is in good agreement with our measurements on the straight path in section 3.1 (Fig. 5(b)). But, as the distance from the focal plane increased these values also increased to ~315nm, ~288nm and ~903nm for the X, Y and Z FWHM respectively at depths of ±10 pm. Depths further away from the focal plane of the primary objective will be imaged further away from the centre of the tertiary objective and are expected to experience more aberrations. This is consistent with our observations and could explain the increase in the axial and lateral FWHMs. Despite these changes, the PSF maintains its overall shape and size and exhibits an increase of ~9% in the lateral and 27% in the axial size within the investigated range.

[0096] 3.4. Scanning unit characterization We adopted a laser scanning approach for volumetric imaging as this method allows for high frame rates and does not mechanically agitate the sample. Most OPM implementations to date, rely on a scanning unit comprised of two scan lenses to maintain optimal remote focusing performance and a galvo mirror for beam steering [6,11], In this scanning geometry, it is important that the galvo mirror is conjugated to the back focal plane of the primary objective. This restricts the motion of the objective close to an optimal position determined by the relay optics between the galvo mirror and the primary objective. Deviations from this optimal position result in a field depended tilt of the excitation beam. This problem does not significantly affect other microscope modalities such as confocal microscopy, that work with a diffraction limited excitation spot but for OPM systems, where the light sheet can extend tens of microns above and below the focal plane, even small amounts of tilt can produce volumes with non-uniform illumination and poor signal to noise ratio.

[0097] To circumvent the above limitations and simplify microscope alignment, we opted for a two-mirror scanning geometry, where two galvo mirrors are positioned close to the primary image plane of the microscope base and are synchronously actuated to pivot the excitation beam around a point along the optical path after the tube lens [18,30], The axial position of the pivot point can be controlled by appropriately adjusting the amount of angular displacement of each mirror as shown in Fig. 6(a). Hence, by adjusting the ratio of the voltage signals that control the mirrors it is possible to compensate for the focusing motion of the primary objective during routine microscope operation and position the pivot point of the excitation beam at its back focal plane. This achieves tilt-invariant scans of the light sheet along the FOV regardless of the final focus position of the primary objective. Furthermore, this two-mirror scanning geometry produces a simpler setup with smaller footprint and avoids additional losses to the fluorescence signal compared to the traditional scan-lens / mirror configuration

[0012] , To determine the optimal voltage ratio between the two mirrors for a given position of the primary objective, we looked at scan-dependent variations in the fluorescence intensity collected from lOOnm beads uniformly distributed in an agarose gel for different mirror voltage-ratios as described in Section 2.3 (Fig. 6(a)). We reasoned that when the mirrors are operated under the optimal mirror-voltage-ratio the light sheet is expected to maintain a constant angle and provide uniform illumination of the oblique plane imaged by the camera at each step of the scan mirrors. In this scenario a linear fit of the summed intensity values as a function of mirror step would produce a slope that equals zero (Fig. 6(a)). Slope values above or below zero suggest non- uniform illumination of the imaged volume and consequently suboptimal mirror-voltage- ratios. In our setup this slope, which we use as a metric to evaluate the illumination uniformity of the scanned volume, appears to follow a linear relationship with the mirrorvoltage-ratio. By performing a linear fit we can therefore obtain the optimal mirror-voltage ratio for a given objective position as shown in Fig. 6(b). By repeating this process for other objective positions, we can produce a calibration curve to map the entire travel range of the primary objective (Fig. 6(c)). It should be noted that adjusting the mirror voltage ratio to its optimal value as the primary objective is changing positions is straightforward in our setup and allows for seamless operation of the microscope. The objective Z position is read from the microscope base by the acquisition computer in real-time and the mirror-voltage-ratio is adjusted on the fly while the user adjusts the focal plane with the focus knob

[0098] 3.5. Microtubule dynamics

[0099] To demonstrate the capabilities of the microscope for high resolution fast volumetric imaging, we imaged microtubule dynamics in the follicular epithelium of Stage 7 Drosophila egg chambers. Microtubules are highly dynamic tubular structures with a characteristic diameter of 25nm

[0031] ,

[0100] In Drosophilla follicle cells, microtubules are organized by apical microtubule organizing centres (MTOCS) and exhibit polarized growth extending from the apical to the basal side of the cell

[0032] ,

[0101] Previous work, either with whole microtubule labelling, or EB1 comets that track the growing end of microtubules, employed imaging rates in the range of ~0.5-lHz [33,34], Imaging at this speed is possible with other microscope modalities e.g., spinning disk, but is usually limited to a single plane. Therefore, given the 3Dorganizationof the microtubule network, collecting data from a whole cell, instead of a single plane, would provide more useful datasets and a better understanding of the microtubule network dynamics

[0035] , Here, we imaged an oblique FOV of 100 pm x 20 pm which in cartesian coordinates corresponds to 100 pm x 10 pm (Y, Z) and is large enough to accommodate the entire dorsal- ventral axis of a Stage 7 Drosophila egg chamber up to a depth of ~10 pm. This depth is sufficient to capture the entire volume of several epithelial cells, which, at this stage of development, have a height of ~5 pm (Fig. 7(a)).

[0102] We were able to observe changes in the microtubule network with time and detect filamentous structures in both the apical and basal side of the follicle cells. The volumetric imaging speed mainly depends on the scan range, the step size and the sample brightness. Here, we used a camera exposure time of 10ms and scanned the mirrors along the X dimension with a step size of 117nmto achieve square pixels in the final deskewed volume and ensure Nyquist sampling.

[0103] Under these conditions we imaged a total cartesian volume of 50 pm xlOO pm xlO pm (X, Y, Z) at a rate of 0.5Hz (Fig. 7(b)). However, by tuning the above experimental parameters, faster imaging rates can be achieved to match the speed requirements of different biological questions.

[0104] For example, previous work tracking tips of microtubules suggested that an imaging rate of ~lHz is required to observe rapid shrinking and growing events

[0031] , By halving the scan range from 50 to 25 pm, the imaging rate will double and volumes of 25 pm x 100 pm x 10 pm can be imaged at rates of 1Hz.

[0105] 4. Conclusion

[0106] In this work we built and characterized an 0PM microscope designed to work seamlessly with a commercially available microscope base. To support all the functionality offered by the microscope base, where the position of the objective lens is not fixed we adopted a 2-mirror scanning geometry that offers a straight forward way to compensate for changes in the position of the objective lens during routine microscope operation. We showed that by adjusting the angular displacement of the mirrors in a calibrated way it is possible to capture 3D volumes with uniform illumination and high signal-to-noise ratio for every position of the objective lens within its travel range. This adjustability is not supported by the more traditional scan-lens / mirror scanning configurations Furthermore, compared to the scan- lens / mirror, the proposed two-mirror scanning geometry provides improved light efficiency and a more compact footprint, offering potential advantages for all OPM designs, whether they employ a commercial base or not. We also showed that within the expected displacement range of the 100X, 1.35NA objective lens from its design position, and for most practical applications, there is no significant effect on the resolving power, or the fidelity of the 3D data produced by the microscope to warrant further improvements in its optical path. To address applications beyond the scope of this work that require larger objective lens displacements or light sheets that illuminate volumes at greater depths, we proposed and discussed a modified optical path that incorporates a pair of glass wedges and allows for uncompromised performance of the microscope throughout the entire application space.

[0107] Given that microscope bases support standard sample mounting methods and various imaging modalities that enable non-expert users to perform a wide range of imaging experiments, we believe that this work describing the incorporation of an OPM modality in this familiar and user-friendly environment will further facilitate the adoption of this technology and make it available to a wider group of researchers.

[0108] Further detailed descriptions of the Figures are provided below:

[0109] Fig. 1. Experimental setup, (a) Excitation laser light is combined with a series of dichroic mirrors, coupled into a single mode fibre collimated with L2, passes through adjustable slits (API) for light sheet NA control and reflects off a 16k Hz resonance mirror (RM) for deshadowing at the sample. The beam passes a set of 3 cylindrical lenses. CY 1 and CY 3 act as a telescope and adjust the beam size in one dimension while CY2, rotated to be orthogonal with CY1 / 3, creates a focused (light sheet) at AP2. AP2 controls the light sheet (FOV) extent at the sample. Next, light is focused (collimated) in one dimension with LI before reflecting off a mirror (Ml) conjugated to the primary objective (OBJ 1) back focal plane for controlling the light sheet position at the sample. Components L2 to LI are all mounted on a translation stage (TS) that allows the excitation beam to be shifted, causing the position to move in OBJ 1’s back aperture and the light sheet to tilt. Excitation laser light is coupled into the common path with a quadband dichroic mirror (DM) and then focused by tube lensTL2before reflecting from two galvo mirrors (G1 and G2) and entering the side port of the microscope base (1X81). Finally, excitation light is focused in one dimension and collimated in the other by the microscope base’s tube lens TL1 before entering the primary objective lens. Fluorescence is collected by the same primary objective (OBJ 1) and follows the same path back to dichroic mirror DM while being descanned by G1 and G2. After separating from excitation laser light, fluorescence enters the remote focusing objective lens (OBJ 2) and forms a focal volume at the front focal plane of OBJ 2. Finally, the tertiary objective (OBJ 3) is positioned with its front glass surface at the remote focusing plane matched to the light sheet at the primary objective tilted imaging plane. OBJ 3 is tiled at 28° and the subsequent quadband emission filter (OBF), tube lens (TL3), and camera follow this tilted path, (b) The tertiary objective, tube lens, and camera are placed on a straight path for initial remote focusing characterization, setup, and diagnostic purposes, (c) Top row, Zemax schematic illustrating that galvos G1 and G2 can support a ± 10mm scan of a 20mmdiameter image across the intermediate image plane of the microscope base without clipping. Bottom row, Zemax footprint diagram showing where rays land on the mirror surface. For a 100X magnification this translates to an area in the sample space of 200 pm in diameter over a scan range of ±100 pm which extends beyond the supported FOV of the primary objective (OBJ 1).

[0110] Fig. 2. Remote focusing PSF characterization, (a) Lateral (XY) and axial (Z) PSFFWHMs as a function of radial distance from the image centre with a bead sample at the primary objective focal plane. Error bars represent one standard deviation (b) Representative PSF cross sections and mean fitted XY, Z FWHM and number of beads with the bead sample at nine Z positions. Each example PSF’s grey scale is set to its min / max. Scale bar 1pm. (c) Lateral and axial FWHMs as a function of radial distance from the image centre for different depths relative to the focal plane.

[0111] Fig. 3. Remote focusing PSF characterization for different levels of axial misalignment between the primary and remote focusing objectives, (a) Lateral and axial FWHM of the PSF as a function of imaging depth for different axial positions of the remote focusing objective relative to its design position (remote focusing objective position indicated in bold text on each plot). Error bars represent one standard deviation. Each data point represents more than 25 bead measurements, (b) The corresponding example PSFXZ cross sections from beads within 50 pm of the image centre. Scale bar is 1 m. Cross sections have been scaled to their min / max values. (c)Lateral and axial FWHMs as a function of radial distance from the FOV centre for different depths relative to the primary objective focal plane.

[0112] Fig. 4. Remote focusing volume distortion characterization, (a) The percent error in magnification is plotted as a function of imaging depth relative to focal plane for different axial positions of the remote focusing objective, (b) The magnification error rate is plotted as a function of the axial objective displacement to obtain the optimal position of the remote focusing objective where the magnification remains constant throughout the remote volume.

[0113] Fig. 5. OPMPSF characterization. Volumes of 100 nm beads immobilized on cover glass were collected at depths between-10 to +10pm with a 5 pm step, separately, and combined into a composite image with an XZ cross section shown in (a). XZ and YZ cross-sections of individual beads from the boxed regions in (a) are shown below at larger scale, (b) The mean X, Y and ZFWHMs as a function of imaging depth relative to the focal plane. Error bars represent ± 1 standard deviation, (c) The X, Y and ZFWHMs of various beads as a function of radial distance from the centre of the FOV for different imaging depths relative to the primary objective focal plane.

[0114] Fig. 6. Scanning unit voltage ratio calibration, (a) The axial position of the pivot point of the excitation beam can be adjusted by changing the voltage ratio of the two scan mirrors to compensate for the motion (focusing) of the primary objective during routine microscope operation. The left panel (voltage ratio A) shows closely matched wave forms (top plot) for each galvo mirror and, below, a schematic showing the excitation scanning pivot point is displaced AX above the primary objective pupil plane. As the light sheet is scanned, its angle changes and progressively misaligns from the camera imaging plane, leading to varying image intensity while scanning. This is observed by imaginglOOnmbeads embedded in a 2%agarose gel, as shown in the oblique image inserts. A sum of all pixels for each oblique image at each scan-step is shown at the bottom of the panel. The summed image intensity varies across the lOOpmscan and is fit with a linear function to determine the slope. The centre panel (voltage ratio B) shows an increased voltage ratio matching the scanning pivot point to the primary objective pupil plane. Here, the light sheet angle remains fixed while scanning, as shown in the oblique image inserts and summed intensity vs. scan-step plot. The right panel (voltage ratio C) shows a further increase in the galvo mirror waveform voltage ratio, resulting in the light sheet angle again changing while scanning. For each objective position, the galvo mirror wave form voltage ratio was scanned and plotted as in (b) and the ratio yielding the minimum slope (dashed vertical blue line) was determined for that objective position, (c) The best ratio for each objective position was determined and fit with a linear function to generate a calibration for the mirror voltage ratio as a function of objective position.

[0115] Fig. 7. A volume in stage 7 Drosophila egg chamber with several epithelial cells was imaged at a rate of 0.5Hz. (a) Overview images of the microtubule network, labelled with Jupiter- GFP, at the apical and basal planes of several epithelial cells, (b) Time series of a zoomed in region spanning a single epithelial cell showing the evolution of the microtubule network with the characteristic filamentous structures changing over time.

[0116] References:

[0117] 6. A. Millett-Sikking and A. York, “High NA single-objective light-sheet,” Zenodo, 2019, 10.5281 / zenodo.3244420.

[0118] 7. E. Sapoznik, B.-J. Chang, J. Huh, et al., “A versatile oblique plane microscope for large- scale and high-resolution imaging of subcellular dynamics,” Elife 9, 1 (2020).

[0119] 10. B. Yang, X. Chen, Y. Wang, et al., “Epi-illumination SPIM for volumetric imaging with high spatial -temporal resolution,” Nat. Methods 16(6), 501-504 (2019).

[0120] 11. J. Kim, “Recent advances ino blique plane microscopy,” Nanophotonics 12(13), 2317- 2334 (2023).

[0121] 12. S.Daetwyler, B.-J. Chang, B.Chen, et al., “Mesoscopic oblique plane microscopy via lightsheet mirroring,” bioRxiv, bioRxiv:2023.08.10.552834 (2023).

[0122] 18. P. Bingen, M. Reuss, J. Engelhardt, et al., “Parallelized STED fluorescence nanoscopy,” Opt. Express 19(24), 23716 (2011).

[0123] 21. E. Remacha, L. Friedrich, J. Vermot, et al., “How to define and optimize axial resolution in light sheet microscopy: a simulation-based approach,” Biomed. Opt. Express 11(1), 8 (2020). 22. Y.-D. Zhang, T. Chen, and L.-Y. Xu, “Application of improved wavelet transform in biological particle detection,” Prog. Biochem. Biophys. 37(10), 1144-1150 (2010).

[0124] 23. J. L. Pech-Pacheco, G. Cristobal, J. Chamorro-Martinez, et al., “Diatom autofocusing in brightfield microscopy a comparative study,” in Proceedings 15th International Conference on Pattern Recognition. ICPR-2000 (IEEE Comput. Soc, n.d.), pp. 314-317.

[0125] 24. D. Nashchekin, L. Busby, M. Jakobs, et al., “Symmetry breaking in the female germline cyst,” Science 374(6569), 874-879 (2021).

[0126] 25. X.Morin, R. Daneman, M. Zavortink, et al., “A protein trap strategy to detect GFP -tagged proteins expressed from their endogenous loci in Drosophila,” Proc. Natl. Acad. Sci. 98(26), 15050-15055 (2001).

[0127] 26. S.Mohanan and A. D. Corbett, “Sensitivity of remote focusing microscopes to magnification mismatch, , JMicrosc 288(2), 95-105 (2022).

[0128] 27. A. D. Corbett, R. A. B. Burton, G. Bub, et al., “Quantifying distortions in two-photon remote focussing microscope images using a volumetric calibration specimen,” Front. Physiol. 5, 1 (2014).

[0129] 28. W. Hong, H. Sparks, and C. Dunsby, “Alignment and characterization of remoterefocusing systems,”Appl. Opt. 62(28), 7431 (2023).

[0130] 29. J.Kim, T. Li, Y.Wang, et al., “Vectorial point spread function and optical transfer function in oblique plane imaging,” Opt. Express 22(9), 11140 (2014).

[0131] 30. F. Gorlitz, P. Hoyer, H. Falk, et al., “A STED microscope designed for routine biomedical applications (Invited Paper),” Prog. Electromagn. Res. 147, 57-68 (2014).

[0132] 31. A. J. Zwetsloot, G. Tut, and A. Straube, “Measuring microtubule dynamics,” Essays Biochem62(6), 725-735 (2018).

[0133] 32. D. Nashchekin, A. R. Fernandes, and D. St Johnston, “Patronin / Shot Cortical Foci Assemble the Noncentrosomal Microtubule Array that Specifies the Drosophila Anterior Posterior Axis,” DevCell 38(1), 61-72 (2016).

[0134] 33. R. Nieuwburg, D. Nashchekin, M. Jakobs, et al., “Localised dynactin protects growing microtubules to deliver oskar mRNA to the posterior cortex of the Drosophila oocyte,” Elife 6, 1 (2017).

[0135] 34. 1. Viktorinova and C. Dahmann, “Microtubule polarity predicts direction of egg chamber rotation in drosophila,” Curr. Biol. 23(15), 1472-1477 (2013). 35. E. Betzig, N. Yamashita, M.Morita, et al., “Three-dimensional tracking of plus-tips by lattice light-sheet microscopy permits the quantification of microtubule growth trajectories within the mitotic apparatus, , J. Biomed. Opt. 20(10), 1 (2015).

Claims

CLAIMS1. An oblique plane microscopy system, comprising: a first optical system comprising a first objective lens, the first optical system configured to deliver excitation light in the form of an oblique light sheet to an object and gather fluorescence light from the object in response to the excitation light sheet; an excitation light coupling unit configured to couple the excitation light to an optical path of the first optical system; and at least two synchronised scanning mirrors arranged on the optical path, configured to scan the excitation light across the sample and de-scan the fluorescence light.

2. The system of claim 1, wherein the position of an object-side focal plane of the first objective lens is adjustable.

3. The system of claim 2, wherein a synchronisation of the at least two synchronised scanning mirrors is configured to be adjusted based on an adjustment of the position of the object-side focal plane of the first objective lens, such that adjustment the position of the object-side focal plane of the first objective lens does not result in a change of an oblique angle of the light sheet.

4. The system of claim 3, wherein the scanning of the at least two synchronised scanning mirrors is configured to be adjusted such that a pivot point of the excitation light beam tracks a pupil plane of the first objective lens as the position of an object-side focal plane of the first objective lens is adjusted.

5. The system of any preceding claim, wherein the at least two synchronised scanning mirrors are arranged substantially at a primary image plane of the first optical system.

6. The system of any preceding claim, wherein the at least two synchronised mirrors are configured such that the excitation light sheet is translated in the focal plane of the first objective lens over a scanning range of the scanning mirrors.

287. The system of any preceding claim, wherein the at least two synchronised mirrors are configured such that an oblique angle of the excitation light-sheet relative to the focal plane of the first objective lens is substantially the same over a scanning range of the scanning mirrors.

8. The system of any preceding claim, wherein the at least two synchronised mirrors are configured to pivot the excitation light path about a pivot point arranged at a pupil plane of the first objective lens.

9. The system of any preceding claim, wherein the at least two synchronised scanning mirrors are galvanometric mirrors.

10. The system of any preceding claim wherein the first optical system is a telescopic optical system, comprising at least one lens in addition to the first objective lenses.

11. The system of claim 1, further comprising a second optical system comprising a second objective lens, the second optical system configured to perform remote focussing; wherein the optical path to which the excitation light coupling unit couples the excitation light is common to the first and second optical systems, and the at least two synchronised scanning mirrors arranged on the common optical pat, between the first and second optical systems.

12. The system of claim 11, wherein the second optical system is a telescopic optical system, comprising at least one lens in addition to the second objective lenses.

13. The system of any one of claims 11 and 12, wherein the second optical system is arranged back-to-back with the first optical system.

14. The system of claim 13, wherein the first and second optical systems are arranged to have a substantially 4f configuration.

15. The system of any preceding claim, further comprising a third optical system comprising a third objective lens configured to image the object.

16. The system of claim 15, wherein the third optical system further comprises an imaging sensor.

17. The system of claim 15 or 16, wherein the light path of the third optical system is angled obliquely to the light path of the second optical system, the oblique angle corresponding to the oblique angle of the light sheet.

18. The system of any one of claims 15 to 17, wherein the third objective lens is a glasstipped objective lens.

19. The system of any preceding claim, further comprising an excitation light source system configured to generate the excitation light forming the oblique light sheet.

20. The system of claim 19, wherein the excitation light source system comprises a scanning mirror configured to generate an excitation light sheet from an excitation light beam.

21. The system of any preceding claim wherein an oblique angle of the light sheet with respect to the imaging plane is between 20 and 40 degrees, preferably about 30 degrees.

22. The system of any preceding claim, wherein the first objective lens has a high numerical aperture.

23. The system of claim 22, wherein the numerical aperture is between 1 and 1.4.

24. A method of three-dimensional fluorescence microscopy imaging, the method comprising imaging an object using the oblique plane microscopy system of any preceding claim.

25. The method of claim 24, wherein the system is the system of any one of claims 2 to 4 and the method further comprises adjusting the position of an object-side focal plane of the first objective lens.

Images