Volume imaging of biological specimens by photochemical sectioning

Abstract

The disclosure relates to photocleavable crosslinkers that are suitable for preparing photodegradable hydrogels and to methods for volumetric imaging of biological specimens by photochemical sectioning and imaging of photodegradable hydrogels. A photocleavable crosslinker according to the disclosure can include one or more water-soluble hydrocarbon linkers, two or more reactive functional groups such as ethylenically unsaturated groups, and one or more photocleavable groups. A crosslinked hydrogel can be formed between the photocleavable crosslinker and other hydrogel monomers. The crosslinked hydrogel can be used as a matrix immobilize a biological or other sample material in a hydrogel composite. The hydrogel composite and the sample material therein can be volumetric imaged by sequentially and repeatedly optically imaging an image layer of the hydrogel composite and then photochemically sectioning a depolymerization layer of the hydrogel composite.

Description

VOLUME IMAGING OF BIOLOGICAL SPECIMENS BY PHOTOCHEMICAL SECTIONINGCROSS REFERENCE TO RELATED APPLICATION

[0001] The benefit of priority to U.S. Provisional Application No. 63 / 678,132 filed August 01, 2024, is hereby claimed and the disclosure is incorporated herein by reference in its entirety.STATEMENT OF GOVERNMENT INTEREST

[0002] This invention was made with government support under UG3MH 126864 awarded by the U.S. National Institutes of Health. The government has certain rights in the invention.BACKGROUND OF THE DISCLOSUREField of the Disclosure

[0003] This disclosure relates to 3D, volumetric or tomographic imaging techniques of specimens enclosed in hydrogels susceptible to photochemical degradation, for example by means of using a photocleavable cross-linker and a UV light.Brief Description of Related Technology

[0004] The study of anatomically complex biological systems and how they function relies on volumetric or 3D imaging techniques to understand how different parts are structured and how they interact and influence the overall functioning of organs and organisms.

[0005] Currently, electron and fluorescence microscopy are used for sample-sectioning- based volumetric imaging. Techniques such as serial-section electron microscopy (EM), and focused-ion-beam (FIB)-SEM employ serial sectioning or surface-ablation of the sample followed by on-section or block-face imaging (Fig. 1A). Alternatively, large volumetric imaging can be performed through sequential physical sectioning followed by on-section or on-block fluorescence imaging of fixed or cleared samples. However, applying such physical sectioning processes to, for example, hydrogel-embedded specimens can lead to sample distortion, tearing, and loss, and the repeated exposure to the light beam can cause photobleaching. Additionally, existing vibratomes and (ultra)microtomes struggle to accommodate the increased size of mammalian tissues, especially when they have been expanded. Besides the mechanical sectioning limitations, electron microscopy imaging techniques have long faced limited penetration challenges, which severely limits volumetric imaging. In response, techniques such as serial-section electron microscopy (ssEM) and focused-ion-beam scanning electron microscopy (FIB-SEM) employ mechanical sectioningor surface-ablation followed by on-section or block-face imaging. While analogous serial sectioning has been applied to volumetric fluorescence imaging of fixed tissue, mechanical sectioning of hydrogel-expanded specimens often leads to sample distortion, tearing, and loss. Hydrogel stiffening (e.g., increasing monomer concentration, re-embedding, or modifying monomer chemistry) can reduce these problems, but the resulting changes in polymer density and composition often lead to reduced imaging resolution and compromised compatibility with established expansion protocols that require specific hydrogel densities and compositions.

[0006] Even with a robust physical sectioning and imaging pipeline, the sample loss and distortion from the sectioning process can pose computational challenges when a large sample volume needs to be stitched precisely at high resolution. Combined, these factors drastically restrict the dimensions and shapes of samples that can be practically imaged using conventional nanoscale electron microscopy or fluorescence imaging techniques. Furthermore, in applications such as brain connectomics, sectioning must be near-perfect, as even a single tear or lost section can be catastrophic for tracing and reconstruction of thin processes and structures.SUMMARY

[0007] In an aspect the disclosure relates to a photocleavable crosslinker (PC) comprising: one or more water-soluble hydrocarbon linkers (L) (e.g., heteroatom-containing hydrocarbon group such as ethylene oxide -CH2CH2O-); two or more ethylenically unsaturated groups (U) or other reactive functional groups (F) each bound directly or indirectly to one or more water-soluble hydrocarbon linkers (L) (e.g., vinyl group U such as H2C=CR-, where R is H, methyl, etc.); and one or more photocleavable groups (P) each bound directly or indirectly to at least one of the one or more water-soluble hydrocarbon linkers (L) or other reactive functional groups (F), and to at least one of the two or more ethylenically unsaturated groups (U) (e.g., o-nitrobenzyl group; can be represented by U-P-L sequence in overall PC molecule).

[0008] Various refinements of the PC structures are possible. In a refinement the photocleavable crosslinker (PC) can include two or more photocleavable groups (P) such that the cleaved PC residue after photodegradation is separated from both previously crosslinked hydrogel chains to which it was bonded. Example structures can include U-P-L- P-U, F-P-L-P-F, etc. In another refinement the photocleavable crosslinker (PC) can include a single photocleavable group (P), which can be sufficient to break the crosslink between adjacent hydrogel chains, with the cleaved PC residue remaining as a pendant group on oneof the hydrogel chains. Example structures of this refinement can include U-P-L-U, F-P-L-F, etc. In another refinement, the photocleavable crosslinker (PC) can include more than the minimum number of L, II, F, and / or P groups.

[0009] In a refinement, the photocleavable crosslinker comprises two or more photocleavable groups (P).

[0010] In a refinement, the photocleavable crosslinker (PC) has a structure U-P-L-P-U, e.g., a generic form of the PC compound 3 from the appendices with terminal ethylenically unsaturated groups (II) at opposing ends of the PC molecule, each with an intervening photocleavable group (P) connecting it to the linker (L).

[0011] In a refinement, the photocleavable crosslinker (PC) has a structure C-[(L-P)n-U]2 where n = 1, 2, 3, or 4; or the photocleavable crosslinker (PC) has a structure C-[L-P-U]nwhere n = 2, 3, or 4; and C is a hydrocarbon core (e.g., C1-C10 with or without heteroatoms).

[0012] In another refinement, the one or more water-soluble hydrocarbon linkers (L) in the photocleavable crosslinker comprise a plurality of ethylene oxide repeat units (-CH2CH2O-).

[0013] The number of ethylene oxide repeat units, or the molecular weight of the linker (L) or photocleavable crosslinker (PC) more generally, can be selected based on the solvent and tissue permeability (in the case of PC). The PC preferably is water soluble, and poly(ethylene glycol) (PEG) chains of MW = 1000-4000 g / mol (about 23-91 ethylene oxide repeat units) are suitable from both the water solubility and synthetic standpoints. If the MW is too low or too high, PC becomes less soluble in water. If the MW is too high, the starting materials of PEG polymers become too viscous, making the PC hard to handle and limiting or preventing PC molecule diffusion through biological samples (e.g., as part of the tissue expansion process). The examples below use a PEG linker (L) with an MW of about 2000 for the PC; however, different MWs inside or outside of the general range can be suitable in different embodiments.

[0014] In a refinement, water-soluble hydrocarbon linkers (L) other than those based on or otherwise including ethylene oxide repeat units can be used. In general, any oligomeric or polymeric water-soluble backbone structure can be used. For example, the water-soluble hydrocarbon linkers (L) can include residues from one or more of polypropylene glycol), polyethyleneimine, polyvinyl alcohol, polyacrylic acid, etc.

[0015] In a refinement, the photocleavable crosslinker (PC) comprises the two or more ethylenically unsaturated groups (II); and the two or more ethylenically unsaturated groups(II) each comprise a (meth)acrylate group (H2C=CR-C(=O)O-, where R is H, methyl). More generally, the ethylenically unsaturated groups (II) can be any C=C group, with or without an ester group linking it to the adjacent PC structure (e.g., photocleavable groups (P) or otherwise). Examples can include vinyl groups II such as H2C=CR-, where R is H, methyl, etc.; other unsaturated groups II such as R1R2C=CR3-, where R1, R2, R3can independently be H, methyl, other C1-C10 hydrocarbon groups with or without heteroatoms, etc.

[0016] In a refinement, the photocleavable crosslinker (PC) comprises the reactive functional groups (F) and the reactive functional groups (F) are each independently selected from the group consisting of alkyne groups, hydroxyl groups, amine groups, N- hydroxysuccinimide (NHS)-esters, azide groups, and maleimide groups.

[0017] In a refinement, the photocleavable crosslinker (PC) can include other functional groups (F) capable of crosslinking with hydrogel monomers. The functional groups (F) can be unsaturated groups or other groups. Examples include alkyne groups, hydroxyl groups, amine groups, NHS-ester groups, azide groups, maleimide groups, etc. Unsaturated alkyne groups in the PC can crosslink with saturated groups in the hydrogel monomers. Hydroxyl groups and amine groups in the PC can crosslink with carboxylic groups in the hydrogel monomers, for example forming ester or amide links, respectively to the hydrogel chains. Azide groups in the PC can crosslink with the hydrogel monomers via click-chemistry mechanisms. Maleimide groups in the PC can crosslink with the hydrogel monomers via maleimide-thiol reaction mechanisms (e.g., via a pendant thiol group in the hydrogel monomers).

[0018] In another aspect of the disclosure the one or more photocleavable groups (P) of the photocleavable crosslinker, each comprise an o-nitrobenzyl group (e.g., o-NCh-Ph-, or o-NO2-Ph-(CH*)-(C=O)-NH-, where Ph is a phenyl group, such as a Passerini reaction residue from reaction with o-nitrobenzaldehyde).

[0019] In a refinement, the one or more photocleavable groups (P) each comprise a photolabile group selected from the group consisting of nitrobenzyl groups, phenacyl groups, benzyl groups, and combinations thereof.

[0020] Photolabile groups are chemical functional groups that can be cleaved or otherwise removed via application of light, for example as used in photoresists. The nitrobenzyl group can be an o-nitrobenzyl group as described above. The phenacyl group can be represented by Ph-C(=O)CH2-, where Ph is a phenyl group. The benzyl group can be represented by Ar-CH2-, where Ar can be a phenyl group (Ph) or a polyaromatic residuesuch as from naphthalene, anthracene, phenanthrene, pyrene, or perylene (e.g., having 2, 3, 4, or 5 fused aromatic rings).

[0021] In a refinement, the one or more photocleavable groups (P) is cleavable upon exposure to electromagnetic radiation having a wavelength in a range of 100 nm to 450 nm or 270 nm to 350 nm.

[0022] The electromagnetic radiation is not particularly limited and generally can be in one or more of the ultraviolet (UV) (e.g., UVA, UVB, UVC), visible, or infrared (IR) (e.g., near IR (NIR), deep IR) wavelength ranges. For example, while UV or NIR are typical cleavage wavelengths (e.g., as a characteristic of the group (P) and / or for applied light during photocleavage) for fluorescence imaging applications, cleaving wavelengths more generally in the visible or other IR or UV ranges can be selected.

[0023] In a refinement, photocleavage can be performed under linear single photon absorption, or nonlinear multi-photon absorption (MPA). The latter permits controlled spatial photodegradation within the gel, with less concern regarding the light path. In the MPA case, the same P group can be triggered at different wavelengths. In the examples, MPA is demonstrated with two-photon absorption (2PA) at 740 nm (which is consistent with PC compound 3’s single photon absorption range in the UV range), and the approach can be extended to higher MPA degrees (e.g., three-photon absorption; 3PA) at higher wavelengths. In further refinements, the cleavage wavelength can be at least 10, 100, 270, 280, 315, 380, 400, 410, 420, 450, 485, 500, 565, 590, 625, 750, 800, 1400, or 3000 nm and / or up to 100, 270, 280, 315, 380, 400, 410, 420, 450, 485, 500, 565, 590, 625, 750, 800, 1400, 3000, or 8000 nm. The foregoing wavelength values can represent a peak intensity value for light distributed across a range of wavelengths, such as from a lamp, laser, etc.)

[0024] In a refinement, cleavage can occur between either or both of the groups adjacent to a given photocleavable group (P). For example, cleavage of a U-P-L sequence in an overall PC molecule can result in one or more of (i) U and P-L fragments, (ii) U-P and L fragments, and (iii) U, P, and L fragments. For example, cleavage of a PC compound with a general structure U-P-L-P-U (i.e. , as in example PC compound 3) can result in U and P-L-P fragments (e.g., where the U fragments can still be incorporated in a hydrogel polymer chain).

[0025] In another aspect of the disclosure, the photocleavable crosslinker and photocleaved residues thereof are water soluble. The photocleavable crosslinker and / or the corresponding photocleaved residue suitably have a sufficiently high water-solubility such that the crosslinker is sufficiently (e.g., completely) dissolved during hydrogel formation andcrosslinking, which in turn assists with penetration, immobilization, and eventual expansion of biological tissues by the hydrogel. For example, the photocleavable crosslinker and its corresponding photocleaved residue can have a solubility in water of at least 0.1 , 0.2, 0.5, 1, 2, 5, or 10 mg / ml (e.g., at 25°C in water, such as in deionized water or other purified reference water medium).

[0026] In another aspect of the disclosure, the photocleavable crosslinker has a molecular weight in a range of 200 g / mol to 10000 g / mol (e.g., as number- or weight-average; or a range bracketing a distribution, such as 1 / 99, 5 / 95, or 10 / 90 cuts of a cumulative distribution.

[0027] In a refinement, the photocleavable crosslinker (PC) and / or the corresponding linker (L) can have a molecular weight in a range of 200 g / mol to 10000 g / mol, such as at least 200, 400, 600, 800, 1000, 1200, 1500, 2000, 3000, or 4000 g / mol and / or up to 800, 1200, 1600, 2000, 3000, 4000, 5000, 6000, 8000, or 10000 g / mol. The foregoing values can be number- or weight-average molecular weight; or a molecular weight range bracketing a distribution, such as 1 / 99, 5 / 95, or 10 / 90 cuts of a cumulative distribution.

[0028] In a refinement, the photocleavable crosslinker has a structure according to formula (I):Formula (I) where n is in a range of 10 to 1000 (e.g., as number- or weight-average; or a range bracketing a distribution, such as 1 / 99, 5 / 95, or 10 / 90 cuts of a cumulative distribution).

[0029] In a further refinement, the number of ethylene oxide units n can be in a range of 10 to 1000, whether in the illustrated compound of Formula (I) or in PC compounds with other P and / or II groups. For example, n can be at least 10, 20, 30, 40, 50, 60, 80, 100, 200, or 300 and / or up to 20, 40, 60, 80, 100, 120, 150, 200, 300, 400, 600, 800, or 1000. The foregoing can represent a number- or weight-average; or a range bracketing a distribution, such as 1 / 99, 5 / 95, or 10 / 90 cuts of a cumulative distribution. Larger values of n can reduce relative water solubility, but such PC compounds can be soluble in various organic solvents. For example, the sample structure could first be embedded in an aqueous medium with the PC and hydrogel monomers to crosslink with the biomolecules / labels of interest. Once this is done, the hydrogel need not remain in an aqueous environment (e.g., the PC-gel can beswitched to an organic solvent or other non-aqueous medium with other desired properties, such as higher refractive index to leverage higher numerical aperture objectives, etc.).

[0030] I another aspect, the disclosure relates to a crosslinked hydrogel comprising: a crosslinked (radical) polymerization product between (i) the photocleavable crosslinker of the disclosure (e.g., via the ethylenically unsaturated groups thereof) and (ii) one or more (ethylenically or vinyl functional) hydrogel monomers.

[0031] In a refinement, the photocleavable crosslinker is present in the crosslinked hydrogel in an amount in a range of 5 to 60 wt.%, 0.1 to 5 mol.%, or 0.2 to 10 eq.% relative to the crosslinked polymerization product (e.g., including the combined amount of the photocleavable crosslinker and the one or more ethylenically functional hydrogel monomers as polymerized / incorporated into the crosslinked polymerization product).

[0032] The foregoing ranges represent alternative weight-, molar-, or ethylenic group equivalent-bases for characterizing the relative amount of the photocleavable crosslinker (PC) in the crosslinked hydrogel. One, two, or all three of the ranges can apply to a given crosslinked hydrogel. For example, the PC can be present in an amount of at least 5, 10, 15, 20, 25, 30, 35, or 40 wt.% and / or up to 20, 30, 35, 40, 45, 50, 55, or 60 wt.%. Alternatively or additionally, the PC can be present in an amount of at least 0.1 , 0.2, 0.4, 0.7, 1 , 1.5, 2, or2.5 mol.% and / or up to 1, 2, 2.5, 3, 4, or 5 mol.%. Alternatively or additionally, the PC can be present in an amount of at least 0.2, 0.4, 0.7, 1 , 1.5, 2, 2.7, or 3.5 eq.% and / or up to 2, 3, 4, 5, 6, 8, or 10 eq.%. Complementary ranges can apply to the combined amount of the ethylenically functional hydrogel monomers (e.g., 40 to 95 wt.%, 95 to 99.9 mol.%, or 90 to 99.8 eq.% relative to the crosslinked polymerization product, including various subranges thereof as defined above).

[0033] In another aspect of the disclosure, the one or more ethylenically functional hydrogel monomers of the crosslinked hydrogel, comprise acrylamide monomer and acrylate monomer (e.g., sodium acrylate).

[0034] In a refinement, the acrylamide monomer is present in the crosslinked hydrogel in an amount in a range of 3 to 30 wt.%, 5 to 50 mol.%, or 5 to 50 eq.% relative to the crosslinked polymerization product; and the acrylate monomer is present in the crosslinked hydrogel in an amount in a range of 10 to 80 wt.%, 30 to 90 mol.%, or 30 to 90 eq.% relative to the crosslinked polymerization product.

[0035] In another aspect of the disclosure, the one or more ethylenically functional hydrogel monomers of the crosslinked hydrogel, comprise one or more of (i) N,N-dimethylacrylamide (DMAA) monomer and acrylate monomer (e.g., sodium acrylate), and (ii) thiolated polyethylene glycol (PEG-SH) monomer and polyethylene glycol-diacrylamide (PEG-diacrylamide) monomer.

[0036] More generally, the hydrogel monomers can include any monomers forming water- soluble, free-radical-chain-growth polymerized polymers that can be crosslinked by the photocleavable crosslinker (PC), whether containing ethylenically unsaturated group or otherwise for polymerization / crosslinking (e.g., in the PC crosslinker or hydrogel monomers). Examples include other hydrogel systems used for expansion microscopy, such as monomer N,N-dimethylacrylamide acid (DMAA) or PEG-SH + PEG-diacrylamide (e.g., thiol-ene reaction between unsaturated PEG-diacrylamide and thiolated PEG-SH having no unsaturated groups). In cases where the photocleavable crosslinker (PC) incorporates other functional groups (F) for crosslinking, other hydrogel monomer(s) / polymer(s) can be used (e.g., those that do not necessarily involve free-radical-chain-growth polymerization).

[0037] In another aspect the disclosure relates to a hydrogel composite comprising: the crosslinked hydrogel of the disclosure as a matrix; and a sample material (e.g., biological tissue) immobilized in the matrix.

[0038] In a refinement, the sample material comprises biological tissue.

[0039] In a refinement, the sample material can include one or more non-biological materials. For example, the sample can include one or more of polymers (e.g., gels and plastics) and crystalline materials (e.g., zeolites, metal organic frameworks). In other cases, the hydrogel composite can be used for non-imaging applications such as light-activated, controlled, and targeted release (e.g., location and dose) of gel-linked drugs, chemicals, small molecules, and biomolecules (e.g., vaccine antibodies / nanobodies / mRNA; hormones, growth factors, etc.), where the sample is the drug, chemical, small molecule, biomolecule, etc. immobilized in the crosslinked hydrogel matrix for subsequent controlled release via photosectioning or other photocleavage mechanism.

[0040] In a further refinement, the sample material comprises mammalian (e.g., human, mouse, or otherwise) tissue (or cells) selected from the group consisting of brain tissue (e.g., olfactory bulb, hippocampus), kidney tissue, and portions thereof (e.g., all or less than all of a particular organ or a component of a particular organ can be sampled). More generally, the sample material can include any type of biological tissue. For example, the sample material can include insect tissue, plant tissue, microbes, yeast colonies, bacteria (biofilms), archaea. The sample also can scale molecules, subcellular organelles, single cells, organoids, explants, tissues, and small organisms.

[0041] In a refinement of the hydrogel composite, the sample material is (linearly) expanded by a factor in a range of 1.5 to 10 (e.g., 2-8, 3-7, or 4-5) relative to an original size (or dimensions) of the sample material before being incorporated into the hydrogel composite. More generally, the sample material can be expanded by a factor of at least 1.5, 1.7, 2, 2.5, 3, 3.5, 4, 4.5, or 5 and / or up to 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 7, 8, 9, 10. The expansion factor can represent a linear expansion factor in one dimension that generally would be essentially isotropic in all three dimensions (up to 5% rms error for expansion isotropy), giving an equivalent volumetric expansion values and ranges equal to the foregoing ranges cubed. For example, a sample isotropically expanded by a linear factor of 5 in three mutually orthogonal directions has an equivalent volumetric expansion factor of 125. The expansion factors in different directions are generally similar, but need not be identical.

[0042] In another refinement of the hydrogel composite, the sample material is fluorescently stained (e.g., labeled or otherwise bound to one or more fluorophores).

[0043] The fluorescent staining or labeling can be performed in a two-step process where the sample tissue is initially bound to a primary antibody, which in turn is bound to a fluorophore-labeled secondary antibody, giving a general structure of tissue-primary antibody-secondary antibody. In a refinement, fluorescent staining or labeling can be achieved using (1) endogenously expressed fluorescent proteins (FPs) and (2) protein tags (HALOTAG, SNAP-tag, etc.). In another refinement, the sample can be stained or labeled with known non-fluorescent materials, but such materials can (1) affect the mechanism of readout and / or (2) retention after hydrogel embedding process.

[0044] In another refinement of the hydrogel composite, the sample material comprises a fluorophore (e.g., as a component of the fluorescent stain / label) having at least one of condition (I) and condition (II): (I) an excitation wavelength at which absorption by the photocleavable crosslinker is not more than 1% (e.g., fraction of incident light energy absorbed by the photocleavable crosslinker at a specific wavelength or range of wavelengths over which photocleavage can appreciably occur); and / or (II) an excitation wavelength at least 50 nm different from (e.g., higher or lower than) a cleavage wavelength of the photocleavable crosslinker (or range of wavelengths over which photocleavage can appreciably occur).

[0045] Suitably, the photocleavage wavelengths of the PC are distinct from those of excitation wavelengths for any fluorophores used as stains or labels for the sample material. For example, this can be expressed as an absorption threshold (particularly suitable when multi-photon absorption is used for PC cleavage), for example where absorption by thephotocleavable crosslinker is not more than 0.0001, 0.001 , 0.01, 0.1 , 1, 2, 5, or 10% at excitation wavelengths of the fluorophores used as a fluorescent stain / label (e.g., also considering the relatively short illumination time for fluorophore excitation). Alternatively or additionally, the excitation wavelengths of the fluorophores can be at least 25, 50, 75, 100, 150, or 200 nm different from (e.g., higher or lower than) a cleavage wavelength of the photocleavable crosslinker. The absorption threshold and / or wavelength difference between the fluorophore excitation wavelength and the PC cleavage wavelength can equivalently apply to the actual EM radiation wavelengths used for excitation and cleavage in an imaging method (i.e., as contrasted with characteristic wavelength properties of the components themselves). The absorption threshold is a particularly suitable characteristic when multiphoton absorption is used for PC cleavage.

[0046] In another aspect the disclosure relates to a method for forming a hydrogel composite, the method comprising: combining (e.g., mixing in an aqueous medium) a sample material with a monomer solution comprising (i) the photocleavable crosslinker of claim 1 and (ii) one or more (ethylenically functional) hydrogel monomers; (radically) polymerizing the photocleavable crosslinker with the one or more ethylenically functional hydrogel monomers to form (i) a crosslinked hydrogel from the photocleavable crosslinker and the one or more (ethylenically functional) hydrogel monomers, and (ii) a hydrogel composite comprising the crosslinked hydrogel as a matrix and the sample material immobilized in the matrix; and expanding the hydrogel composite with water (e.g., contacting with / immersing in water for sufficient time and number of repetitions to swell the hydrogel composite and thereby expand the sample material).

[0047] In a refinement of the method for forming a hydrogel composite, the initial monomer solution can include one or more polymerization reaction components, such as a radical initiator, a promoter (to stabilize radicals), and an inhibitor (to delay polymerization until all reagents diffuse through the sample), for example including APS, TEMED, and 4HT as in the examples.

[0048] In another refinement of the method for forming a hydrogel composite, the sample material is fluorescently stained (e.g., in a step prior to being combined with the monomer solution).

[0049] In another aspect the disclosure relates to a method for volumetric optical (e.g., fluorescent) imaging of a sample material, the method comprising:(a) providing a hydrogel composite according to the disclosure; (b) optically imaging a (top) image layer of the hydrogel composite (e.g., acquiring and storing in a computer medium a 3D / volumetricimage of the sample material in the image layer); (c) photochemically sectioning a (top) depolymerization layer of the hydrogel composite (e.g., forming free water-soluble cleaved PC residue, free non-crosslinked hydrogel polymer chains, and free sample material / biological tissue from the depolymerization layer, which then dissipate / diffuse / dissolve into a surrounding aqueous medium); and (d) repeating steps (b) and (c) a plurality of times for sequential image layers and depolymerization layers. In a refinement of the method for volumetric optical imaging, optically imaging the top image layer of the hydrogel composite is performed to obtain a volumetric section image of any sample material in the top image layer; and the method further comprises (e) stacking the sequential image layers to form a composite volumetric (or 3D) image (e.g., to perform tomographic reconstruction).

[0050] In a refinement of the method for volumetric optical imaging, the image layer has a thickness greater than a thickness of the depolymerization layer (e.g., such that sequential image layers have overlapping volumes to facilitate computational assembly of a composite 3D / volumetric image of the sample material as a whole).

[0051] Various refinements of the disclosed thicknesses of the polymerization layers and the image layers are possible. The image layer can have a thickness (or depth) of about 0.1- 2.6 mm or about 1.2-2 mm, for example at least 0.1, 0.2, 0.4, 0.6, 0.8, 1, 1.2, 1.4, 1.7, or 2 mm and / or up to 0.5, 0.7, 0.9, 1.3, 1.6, 2, 2.3, or 2.6 mm. The depolymerization layer can have a thickness (or depth) of about 0.05-2.2 mm or about 0.7-1.4 mm, for example at least 0.05, 0.1 , 0.2, 0.4, 0.5, 0.7, 1, 1.2, or 1.5 mm and / or up to 0.5, 0.7, 0.9, 1.2, 1.4, 1.7, 2, or 2.2 mm. The image layer can be thicker than the depolymerization layer by a factor of 1.1 to 2 (i.e., 10%-100% thicker), such as at least 1.1, 1.2, 1.3, 1.4, 1.5, or 1.7 and / or up to 1.2, 1.3, 1.4, 1.6, 1.8, or 2. Alternatively or additionally, the image layer can be thicker than the depolymerization layer by about 0.2-1.4 mm or 0.4-1 mm, such as at least 0.2, 0.4, 0.6, 0.8, or 1 mm and / or up to 0.5, 0.6, 0.7, 0.8, 0.9, 1 , 1.2, or 1.4 mm.

[0052] In another refinement of the method for volumetric optical imaging, a 3D / volumetric image of the sample material in the image layer comprises voxels having a length dimension in a range of 5 nm to 100 nm or 5 nm to 300 nm (e.g., with one, two, or three (mutually orthogonal) voxel dimensions being generally in this range, although they can have different individual values).

[0053] Various refinements of the disclosed voxel dimensions are possible. Generally, the voxels can have up to three independent (e.g., same or different) voxel length dimensions (e.g., edge lengths) that are at least 5, 10, 20, 30, 40, 50, 60, or 70 nm and / or up to 30, 40,50, 60, 70, 80, 90, 100, 150, 200, 250, or 300 nm. The voxel dimensions / imaging resolution values can represent effective voxel size accounting for the expansion ratio (e.g., physical imaging voxel size divided by expansion factor). For example, given a (50 nm)3voxel resolution with a hydrogel isotropic linear expansion factor of 5, the physically imaged voxel size is a (250 nm)3volume in the hydrogel composite, which is then size-adjusted to (50 nm)3.

[0054] In another refinement of the method for volumetric optical imaging, optically imaging the image layer comprises performing on-block fluorescence imaging (e.g., imaging a pre-determined volumetric depth from an external top (or block-face) surface).

[0055] In another refinement of the method for volumetric optical imaging, optically imaging the image layer comprises exposing the image layer to electromagnetic radiation at one or more excitation wavelengths in a range of 400 nm to 800 nm (e.g., one excitation wavelength corresponding to each different fluorophore in the fluorescently stained sample; for example as applied by a single-wavelength (e.g., laser), or broadband (e.g., LEDs, halogens, arc) light source); photochemically sectioning the depolymerization layer comprises exposing the image layer to electromagnetic radiation at a photocleavage wavelength in a range of 100 nm to 450 nm; and at least one of condition (I) and condition (II) is satisfied: (I) absorption by the photocleavable crosslinker is not more than 1% at the one or more excitation wavelengths; and / or (II) the one or more excitation wavelengths are each at least 50 nm different from (e.g., higher or lower than) the photocleavage wavelength. Photochemically sectioning the depolymerization layer can include exposing a portion or sub-volume of the image layer corresponding to the eventual depolymerization layer.

[0056] More generally, the excitation wavelengths and the photocleavage wavelengths can be independently selected from wide ranges, such as at least 10, 100, 270, 280, 315, 380, 400, 410, 420, 450, 485, 500, 565, 590, 625, 750, 800, 1400, or 3000 nm and / or up to 100, 270, 280, 315, 380, 400, 410, 420, 450, 485, 500, 565, 590, 625, 750, 800, 1400, 3000, or 8000 nm. The foregoing wavelength values can represent a peak intensity value for light at a single wavelength or light distributed across a range of wavelengths, such as from a laser, lamp, etc.

[0057] Suitably, the photocleavage wavelengths of the PC are distinct from those of excitation wavelengths for any fluorophores used as stains or labels for the sample material. For example, this can be expressed as an absorption threshold (particularly suitable when multi-photon absorption is used for PC cleavage), for example where absorption by the photocleavable crosslinker is not more than 0.0001 , 0.001 , 0.01 , 0.1 , 1 , 2, 5, or 10%excitation wavelengths of the fluorophores used as a fluorescent stain / label (e.g., also considering the relatively short illumination time for fluorophore excitation). Alternatively or additionally, the excitation wavelengths of the fluorophores can be at least 25, 50, 75, 100, 150, or 200 nm different from (e.g., higher or lower than) a cleavage wavelength of the photocleavable crosslinker.

[0058] In another refinement of the method for volumetric optical imaging, photochemically sectioning the depolymerization layer comprises exposing the image layer to a multi-photon light source (e.g., two-photon, three-photon, etc. illumination for photocleavage via multi-photon absorption (MPA)).

[0059] In another refinement of the method for volumetric optical imaging, the hydrogel composite is in an aqueous environment. The aqueous environment can be bath or reservoir, which is typically at least 10* volume relative to initial hydrogel composite size. The hydrogel composite can be positioned on (or mounted to) a (bottom) support substrate (e.g., glass or other optically transparent material). The aqueous environment suitably can include water, buffers (e.g., 1x PBS, 10x PBS, 1 mM Tris), and weak base (e.g., 1 mM NaOH) as an imaging medium. The sample expansion factor can change (e.g., shrink in higher ionic strength) according to the imaging medium.

[0060] In a refinement of the method for volumetric optical imaging, photochemically sectioning the depolymerization layer comprises (completely) photodegrading the depolymerization layer to form a remaining bulk portion of the hydrogel composite. In another refinement of the method for volumetric optical imaging, photochemically sectioning the depolymerization layer comprises photodegrading a portion (only) of the depolymerization layer at an interface between the depolymerization layer and the image layer, thereby releasing a cleaved (or photosliced) hydrogel composite portion and forming a remaining bulk portion of the hydrogel composite. In either case, the remaining bulk portion of the hydrogel composite with its newly formed exposed or outermost surface then forms the “starting” hydrogel composite for the next volumetric imaging cycle.

[0061] In another aspect the disclosure relates to a kit comprising: the photocleavable crosslinker of the disclosure; and one or more of ethylenically (or vinyl) functional hydrogel monomers, polymerization reagents (e.g., initiators, promotors, and / or inhibitors), fluorescent stain reagents (e.g., primary and secondary antibodies), light sources (e.g., for photosectioning or fluorescence excitation), gelation chambers (e.g., mold or other container to accommodate different sizes of samples), and brushes (e.g., paint brushes to handle gels / gelled samples.

[0062] In another aspect, the disclosure relates to a photocleavable crosslinker (PC) comprising: a water-soluble hydrocarbon linker (L) comprising a plurality of ethylene oxide repeat units; two ethylenically unsaturated groups (II) each comprising an acrylate group; and two photocleavable groups (P) each comprising an o-nitrobenzyl group; wherein the photocleavable crosslinker (PC) has a structure U-P-L-P-U. In a refinement, the photocleavable crosslinker can have a structure according to formula (I) as illustrated above, for example where n is in a range of 10 to 100.

[0063] In another aspect, the disclosure relates to a crosslinked hydrogel comprising: a crosslinked polymerization product between (i) a photocleavable crosslinker as disclosed herein, (ii) an acrylamide monomer, and (iii) an acrylate monomer; wherein: the photocleavable crosslinker is present in the crosslinked hydrogel in an amount in a range of 5 to 60 wt.%; the acrylamide monomer is present in the crosslinked hydrogel in an amount in a range of 3 to 30 wt.%; and the acrylate monomer is present in the crosslinked hydrogel in an amount in a range of 10 to 80 wt.%.

[0064] In another aspect, the disclosure relates to a hydrogel composite comprising: a crosslinked hydrogel as disclosed herein as a matrix; and a sample material immobilized in the matrix, the sample material comprising mammalian tissue and a fluorophore having at least one of condition (I) and condition (II): (I) an excitation wavelength at which absorption by the photocleavable crosslinker is not more than 1%, and (II) an excitation wavelength at least 50 nm different from a cleavage wavelength of the photocleavable crosslinker.

[0065] In another aspect, the disclosure relates to a method for volumetric optical imaging of a sample material, the method comprising: (a) providing a hydrogel composite as disclosed herein, wherein the hydrogel composite is in an aqueous environment; (b) optically imaging an image layer of the hydrogel composite to obtain a volumetric section image of any sample material in the image layer, wherein optically imaging the image layer comprises exposing the image layer to electromagnetic radiation at one or more excitation wavelengths in a range of 400 nm to 800 nm; (c) photochemically sectioning a depolymerization layer of the hydrogel composite, wherein photochemically sectioning the depolymerization layer comprises exposing the image layer to electromagnetic radiation at a photocleavage wavelength in a range of 100 nm to 450 nm; (d) repeating steps (b) and (c) a plurality of times for sequential image layers and depolymerization layers; and (e) stacking the sequential image layers to form a composite volumetric image comprising voxels having a length dimension in a range of 5 nm to 100 nm.BRIEF DESCRIPTION OF THE DRAWINGS

[0066] The figures described below depict various aspects of the compositions and methods disclosed herein. It should be understood that each figure depicts an embodiment of a particular aspect of the disclosure, and that each of the figures is intended to accord with a possible embodiment thereof.

[0067] Fig. 1A are schematics of volumetric imaging techniques used in the study of biological specimens based on serial sectioning (microtoming, left), and focused-ion-beam (FIB) serial block-face electron microscopy (EM, right).

[0068] Fig. 1 B is a schematic showing volumetric imaging of a specimen using photochemical sectioning (PS) serial on-block volumetric optical imaging (VIPS).

[0069] Fig. 2A is a schematic of the photochemical degradation (photodegradation) of acrylamide (AA) / sodium acrylate (SA) polymer gel via photocleavage of the crosslinker (photocleavable crosslinker, PC) (top) and a schematic of the photocleavage of the PC under UV illumination [hv (UV)] (bottom).

[0070] Fig. 2B is a schematic illustrating a method for forming a crosslinked hydrogel and sample-containing hydrogel composite according to the disclosure.

[0071] Fig. 3A is a plot of the (PC)-crosslinked polyacrylamide / sodium polyacrylate hydrogel (PC-gel) expansion factor vs. PC concentration ( w). The PC concentration ( w) can be expressed as the molar ratio of PC relative to bis-acrylamide at a standard reference concentration of 1.5 g / L (or about 0.00972 M based on a molecular weight of 154.2 g / mol for bis-acrylamide), a crosslinking monomer in a standard non-photodegradable, bis- acrylamide-crosslinked polyacrylamide / sodium polyacrylate gel (Bis-gel). The amount of PC in a PC-gel can be expressed as a ratio relative to bis-acrylamide, because corresponding Bis-gels are standard gel recipes commonly used for tissue expansion.

[0072] Fig. 3B is a graph of the photodegradation time of expanded PC-gel-embedded HEK cells under UV (405 nm) laser illumination at different PC concentrations (error bars = standard error of the mean (SEM); n = 3 gels).

[0073] Fig. 4 are photographs of a piece of fluorescently labeled PC gel in water before (left), during (center), and after (right) exposure to light-sheet UV illumination (A = 405 nm).

[0074] Fig. 5 is a schematic of an optical setup for two photon volume imaging by photochemical sectioning method (VIPS).

[0075] Fig. 6A are confocal microscope images of HEK cells labeled with far-red fluorescent dyes for p-tubulin, and embedded and expanded in PC-gel. The images aretaken before (left) and after (right) exposure to a 405 nm UV laser, the dotted line indicates the limit of the region exposed to the laser. Scale bar: 10 pm (48 pm post-expansion).

[0076] Fig. 6B are confocal microscope images of HEK cells labeled with far-red fluorescent dyes for p-tubulin, and embedded and expanded in Bis-gel (control). The images are taken before (left) and after (right) exposure to a 405 nm UV laser, the dotted line indicates the limit of the region exposed to the laser. Scale bar: 10 pm (45 pm postexpansion).

[0077] Fig. 7 is a system schematic for performing lattice light sheet excitation, two-color detection, and light-sheet photochemical sectioning for light-sheet VIPS.

[0078] Fig. 8A is a schematic showing a tissue embedded and expanded in PC-gel, the depth of the imaging slices, and overlap between imaging slices used in confocal VIPS imaging (top panel); and schematics illustrating photochemical sectioning via photodegradation and photoslicing (bottom panel).

[0079] Fig. 8B is a schematic showing imaging sub-volumes and the depth of imaging slices within sub-volumes as implemented in 2P and light sheet VIPS.

[0080] Fig. 9 is a schematic of a kit including a photocleavable cross-linker and other components and tools used in preparing photodegradable hydrogels and hydrogel composites.

[0081] Fig. 10 is a schematic of steps of a method for volumetric imaging by photochemical sectioning.DETAILED DESCRIPTION

[0082] Hydrogel-based tissue clearing and expansion techniques have improved the ability to study cellular structures and molecular details within intact tissues at resolutions beyond the diffraction limit of light. However, current fluorescence imaging methods face a fundamental trade-off between imaging depth and resolution when applied to large, expanded tissue samples. The deeper the imaging depth, the lower the achievable resolution, thus limiting either the size of the sample that can be imaged at high resolution or the level of detail that can be captured in three dimensions (3D). The disclosed compositions and methods address this limitation, providing for volume imaging of biological specimens by photochemical sectioning (VIPS), a nanoscale volume fluorescence imaging method. VIPS employs a non-physical cutting process called (referenced herein as "photochemical sectioning") to enable high-resolution imaging of large tissue volumes.

[0083] Several features of the VIPS process and its corresponding compositions and articles are summarized below.

[0084] Photodegradable Hydrogel: The disclosed hydrogel used for sample immobilization and expansion incorporates a photocleavable crosslinker, such that the hydrogel rapidly and completely degrades upon exposure to UV light under physiological conditions.

[0085] Photochemical Sectioning: By embedding and expanding tissue specimens in this hydrogel, volume fluorescence imaging of the specimens can be extended beyond the objective's working distance. This is achieved through sequential on-block (or block-face) imaging and controlled photodegradation of the hydrogel ("photochemical sectioning"). This photochemical sectioning process selectively removes a thin layer of the tissue sample using photochemical reactions.

[0086] On-Block Fluorescence Microscopy: After each sectioning step, an upper or outer volumetric layer defined by the newly exposed tissue surface is imaged using high-resolution fluorescence microscopy. This generates a series of 3-dimensional images that represent different depths within the tissue.

[0087] Petabyte-Scale Imaging: The photochemical sectioning process can be combined with lattice light-sheet microscopy to image the entire tissue or organ samples at sub-100 nm resolution levels, for example a wild-type mouse olfactory bulb at a voxel size of 66 nm x 66 nm x 87 nm as illustrated in the examples, which generates ~0.6 petabytes of raw data, which is substantially larger than datasets acquired by other known monolithic superresolution imaging techniques.

[0088] Whole-Organ Imaging at Sub-100nm Resolution: By combining VIPS with high numerical aperture objectives, isotropic 3D imaging is possible for whole brains, organs, or even entire organisms at resolutions below 100 nm or even 50 nm. This allows the mapping complex biological structures and processes at the nanoscale.

[0089] Protein-Specific Connectomics at High Speed: VIPS, coupled with protein-specific labeling techniques, permits mapping of neuronal connections with molecular specificity at a speed that is an order of magnitude faster than current volumetric electron microscopy (EM) methods. This significantly accelerates connectomics research, allowing imaging, analysis, characterization, etc. of the intricate wiring of the brain and other complex neural systems more efficiently.

[0090] High-Throughput Assessment of Neurodegeneration: The examples below demonstrate the ability of VIPS to assess axon and myelin degeneration in a mouse modelof Niemann-Pick Disease Type C1 (NPC1) at single-axon resolution. This opens new avenues for studying neurodegenerative diseases (or a variety of pathophysiology).

[0091] The examples described in the examples below demonstrate the capabilities of VIPS by reconstructing the axonal projection and myelination patterns within the entire murine olfactory bulb at ~100nm sub-diffraction-limit resolution. This illustrates the potential of VIPS to map dense, complex neuronal networks and to study cellular interactions in large tissue volumes.

[0092] Several potential applications of the VIPS process and its corresponding compositions and articles are summarized below.

[0093] Alternative to Electron Microscopy: VIPS can provide a more accessible and scalable alternative to electron microscopy techniques (e.g., ssTEM, FIB-SEM) for subcellular imaging and reconstruction.

[0094] Whole-Brain Studies: VIPS can be used to perform comparative studies of neuronal organization and disease-related changes at the whole mammalian brain level, opening new avenues for neuroscience research.

[0095] Neurobiology: VIPS can be used to study the detailed structure of neuronal networks and brain mapping, aiding in understanding neurological diseases such as Alzheimer's, Parkinson's, and autism. The technology can be used for connectomics, and to study changes in synaptic connectivity in normal and neurodegenerative diseases, providing insights into the underlying mechanisms of synaptic loss and dysfunction.

[0096] Oncology: The method can help in the detailed imaging of cancerous tissues, enabling researchers to better understand tumor microenvironments and metastatic processes, potentially leading to the discovery of new therapeutic targets.

[0097] Similarly, the visualizing across scales from molecules to organs has an impact across multiple subdisciplines of biological research including: Developmental Biology, Immunology, Regenerative Medicine and Stem Cell Research, Cardiovascular Research, Metabolic Disorders, Infectious Disease Research, Genomic Visualization, Wound Healing and Tissue Repair. In addition, there are applications in the pharma sector for drug testing, the pharmacokinetics to map the therapeutic index on tissues with subcellular resolution.

[0098] Plant Biology: Beyond biomedical applications, VIPS can be used to study plant tissues at high resolution, aiding in the understanding of plant physiology and pathology, which can lead to the development of more resilient crop varieties.

[0099] Environmental Monitoring: The technology can be applied to study the effects of environmental pollutants on various organisms at the cellular level, contributing to environmental protection and remediation efforts.

[0100] Petabyte-to- exabyte scale datasets for developing foundation models: The vast amount of imaging data generated by VIPS can be used to train large Al models capable of automated tissue analysis, diagnostics. VIPS imaging data can provide the highest- resolution foundation model in life sciences. The high-resolution imaging data generated by VIPS can be used to create virtual lab experiences, providing students and researchers with remote access to detailed biological specimens.

[0101] The disclosure relates to methods of embedding, sectioning and imaging samples, e.g., biological samples like tissues or cells, in hydrogels that allow expansion of the sample and are susceptible to photodegradation by the use of a photocleavable crosslinker. Furthermore, the disclosure relates to photocleavable crosslinkers suitable for crosslinking hydrogels and prone to cleaving or decomposing when exposed to light having wavelengths relevant to imaging and staining of biological samples, e.g., UV light, near infrared light.

[0102] The disclosure relates to methods and techniques to overcome the challenges of traditional imaging and physical sectioning techniques (Fig. 1A). The disclosed methods particularly relate to volumetric imaging via photochemcial sectioning (VIPS), which is generally advantageous in the imaging of biological specimens. VIPS is a light-based sample sectioning and imaging process where a labelled or stained sample is embedded in a photodegradable gel or hydrogel (Fig. 1 B). The photodegradable hydrogel is crosslinked with a photocleavable crosslinker. The photocleavable crosslinker allows for the spatially confined degradation of the hydrogel under single and / or multi-photon illumination. The method combines sequential on-block volumetric illumination, imaging and photo-sectioning with, for example, UV lattice light-sheet illumination. These techniques can be combined with large scale computation (e.g., petabyte-scale) to image and reconstruct entire structures, for example neural structures like the olfactory bulb.Definitions

[0103] As used herein, the terms embedding, encapsulating, and gelating are used interchangeably and refer to a process of firmly placing a sample within a medium (i.e., a gel or hydrogel) to support it during sectioning and imaging processes.

[0104] Photochemical sectioning refers to a light-based process in which a photodegradable gel / hydrogel and consequently a sample embedded in such hydrogel can be selectively and rapidly degraded when exposed to light, particularly UV light.

[0105] As used herein, photocleavable refers to a chemical bond, a section of a molecule, or an entire molecule that can be degraded or broken with light, particularly with light of a specific wavelength or range of wavelengths, e.g., UV light, visible light, infrared light.

[0106] Photolabile groups are chemical functional groups that can be cleaved or otherwise removed via the application of light, for example as used in photoresists.

[0107] A Passerini reaction is a three-component reaction between a carboxylic acid, a carbonyl compound such as a ketone or aldehyde, and an isocyanide, for example as illustrated below in Scheme 1 for suitable selections of Ri, R2, R3, and R4 corresponding to an eventual photocleavable crosslinker product. For example, the carboxylic acid can be (meth)acrylic acid or other carboxylic compound with an ethylenically unsaturated group or a reactive functional group, the carbonyl compound can be o-nitrobenzaldehyde or other ketone / aldehyde with a photocleavable group, and the isocyanide can be a diisocyanide- terminated polyethylene oxide or other water-soluble hydrocarbon linking group.„ Scheme 1

[0108] As used herein, the term voxel is considered an equivalent to volume pixel and refers to the smallest distinguishable, box-shaped part of a three-dimensional image (3D image) of a sample or a sub-volume or section of a sample.

[0109] As used herein the terms 3D-rendered dataset and composite 3D / volumetric images are used interchangeably and refer to a 3D image reconstructed or formed from performing image processing and stacking techniques on a sequence of images captured by methods of the disclosure.Photocleavable Crosslinkers

[0110] A photocleavable crosslinker according to the disclosure can have one or more water-soluble hydrocarbon linkers (L), two or more ethylenically unsaturated groups (U) or other reactive functional groups (F), and one or more photocleavable groups (P). The two or more ethylenically unsaturated groups (U) can be bound directly or indirectly to one or more water-soluble hydrocarbon linkers (L). The one or more photocleavable groups (P) can eachbe bound directly or indirectly to at least one of the one or more water-soluble hydrocarbon linkers (L) or other reactive functional groups (F) and to at least one of the two or more ethylenically unsaturated groups (II) or other reactive functional groups (F).

[0111] Water-soluble linkers (L) of the disclosure can include heteroatom-containing hydrocarbons such as one or more ethylene oxide units (-CH2CH2O-), e.g., polyethylene oxide (PEG). The number of ethylene oxide repeat units, or the molecular weight of the linker (L) or photocleavable crosslinker (PC) more generally, can be selected based on the solvent and tissue permeability (in the case of PC). The PC preferably is water soluble, and poly(ethylene glycol) (PEG) chains of MW of about 1000-4000 g / mol (about 23-91 ethylene oxide repeat units) are suitable from both the water solubility and synthetic standpoints. In various embodiments, the MW of the PEG chains can be about 1000, 1200, 1300, 1500, 1700, 1800, 2000, 2200, 2500, 2700, 3000, 3200, 3500, or about 4000 g / mol and / or have about 23, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, or about 90 ethylene oxide repeat units. If the MW is too low or too high, PC becomes less soluble in water. If the MW is too high, the gel precursor including the PC becomes too viscous, making the PC and consequently the hydrogel, hard to handle, and limiting or preventing diffusion of the hydrogel precursors through biological samples (e.g., as part of the tissue embedding process). The examples use a PEG linker L with a MW of about 2000 g / mol for the PC; however, different MWs inside or outside of the general range can be suitable in different embodiments.

[0112] In embodiments, water-soluble hydrocarbon linkers (L) other than those based on or otherwise including ethylene oxide repeat units can be used. In general, any oligomeric or polymeric water-soluble backbone structure can be used. For example, the water-soluble hydrocarbon linkers (L) can include residues from one or more of polypropylene glycol), polyethyleneimine, polyvinyl alcohol, polyacrylic acid, etc.

[0113] Ethylenically unsaturated groups (II) can include any C=C group, with or without an ester group linking it to the adjacent PC structure (e.g., photocleavable groups (P) or otherwise). Examples can include vinyl groups such as H2C=CR-, where R is H, methyl, etc.; other unsaturated groups II such as R1R2C=CR3-, where R1, R2, R3can independently be H, methyl, or other C1-C10 hydrocarbon groups with or without heteroatoms, etc. For example, C1 , C2, C3, C4, C5, C6, C7, C8, C9, or C10 hydrocarbon groups with or without heteroatoms can be selected. Ethylenically unsaturated groups (II) can include, for exampleunsaturated carboxylic acids, e.g., acrylic acid0or methacrylic acid.

[0114] Photocleavable groups can include an o-nitrobenzyl group, for example, where Ri can be an amide groupnitrophenyl)propanamidewhere Ph is a phenyl group, such as a Passerini reaction residue from reaction with o- nitrobenzaldehyde such as.

[0115] Photocleavable groups can include one or more photolabile groups selected from the group consisting of nitrobenzyl groups, phenacyl groups, benzyl groups and combinations thereof. The nitrobenzyl group can be an o-nitrobenzyl group as described herein. The phenacyl group can be represented by Ph-C(=O)CH2-, where Ph is a phenyl group. The benzyl group can be represented by Ar-CH2-, where Ar can be a phenyl group (Ph) or a polyaromatic residue such as from naphthalene, anthracene, phenanthrene, pyrene, or perylene (e.g., having 2, 3, 4, or 5 fused aromatic rings).

[0116] The photocleavable crosslinker can include reactive functional groups (F). The reactive functional groups (F) can each be independently selected from the group consisting of alkyne groups (c=c), hydroxyl groups (-OH), amine groups (-NH3), N-hydroxysuccinimide H(NHS)-ester groups, azide groups (-N=N+=N-), or maleimide groups

[0117] The photocleavable crosslinker (PC) can crosslink hydrogel monomers through bonding via the ethylenically unsaturated groups (II) or the reactive functional groups (F) described herein. Ethylenically unsaturated groups in the PC can react and crosslink with ethylenically unsaturated groups in the hydrogel monomers. Unsaturated alkyne groups in the PC can crosslink with saturated groups in the hydrogel monomers. Hydroxyl groups and amine groups in the PC can crosslink with carboxylic groups in the hydrogel monomers, for example forming ester or amide links, respectively to the hydrogel chains. Azide groups in the PC can crosslink with the hydrogel monomers via click-chemistry mechanisms.Maleimide groups in the PC can crosslink with the hydrogel monomers via maleimide-thiol reaction mechanisms (e.g., via a pendant thiol group in the hydrogel monomers).

[0118] In some embodiments, the photocleavable crosslinker (PC) can include two or more photocleavable groups (P) such that the cleaved PC residue after photodegradation is separated from both previously crosslinked hydrogel chains to which it was bonded. Example structures of this embodiment can include U-P-L-P-U, F-P-L-P-F. In other embodiments, the photocleavable crosslinker (PC) can include a single photocleavable group (P), which can be sufficient to break the crosslink between adjacent hydrogel chains, with the cleaved PC residue remaining as a pendant group on one of the hydrogel chains. Example structures of this embodiment can include U-P-L-U, F-P-L-F, etc. In other embodiments, the photocleavable crosslinker (PC) can include more than the minimum number of L, II, F, and / or P groups.

[0119] The photocleavable crosslinker (PC) can have a hydrocarbon core (C) including a carbon chain with about 1 to about 10 carbon atoms (i.e., C1-C10) with or without heteroatoms. For example, C1 , C2, C3, C4, C5, C6, C7, C8, C9, or C10 hydrocarbon groups with or without heteroatoms can be selected as the hydrocarbon core (C). The photocleavable crosslinker (PC) can have a structure C-[(L-P)n-U]2 where n = 1 , 2, 3, or 4. Alternatively, the PC can have a structure C-[L-P-U]nwhere n = 2, 3, or 4.

[0120] A photocleavable crosslinker according to the disclosure can include one or more photocleavable groups (P) cleavable upon exposure to electromagnetic radiation having a wavelength in a range of about 100 nm to 450 nm or 270 nm to 350 nm. For example, electromagnetic radiation with a wavelength of about 100, 120, 150, 200, 220, 250, 270, 300, 350, 370, 400, 400, 420, 450 nm or any numbers and ranges therebetween.

[0121] The electromagnetic radiation is not particularly limited and generally can be in one or more of the ultraviolet (UV) (e.g., UVA, UVB, UVC), visible, or infrared (IR) (e.g., near IR (NIR), deep IR) wavelength ranges. For example, while UV or NIR are typical cleavage wavelengths (e.g., as a characteristic of the group (P) and / or for applied light during photocleavage) for fluorescence imaging applications, cleaving wavelengths more generally in the visible or other IR or UV ranges can be selected. In embodiments, photocleavage can be performed under linear single photon absorption, or nonlinear multi-photon absorption (MPA). The latter permits controlled spatial photodegradation within the gel, with less concern regarding the light path. In the MPA case, the same P group can be triggered at different wavelengths. In the examples, MPA is demonstrated with two-photon absorption (2PA) at 740 nm (which is consistent with PC compound 3’s single photon absorption range in the UV range), and the approach can be extended to higher MPA degrees (e.g., three- photon absorption; 3PA) at higher or lower wavelengths. In various embodiments, thecleavage wavelength can be at least 10, 100, 270, 280, 315, 380, 400, 410, 420, 450, 485, 500, 565, 590, 625, 750, 800, 1400, or 3000 nm and / or up to 100, 270, 280, 315, 380, 400, 410, 420, 450, 485, 500, 565, 590, 625, 750, 800, 1400, 3000, or 8000 nm. The foregoing wavelength values can represent a peak intensity value for light distributed across a range of wavelengths or a monochromatic wavelength, such as from a lamp, gas laser, solid-state laser, etc.

[0122] According to the disclosure, cleavage can occur between either or both of the groups adjacent to a given photocleavable group (P). For example, the cleavage of a ll-P-L sequence in the overall PC molecule can result in one or more of (i) II and P-L fragments, (ii) U-P and L fragments, and (iii) II, P, and L fragments. For example, cleavage of a PC compound with a general structure U-P-L-P-U can result in II and P-L-P fragments, U-P and L-P-U fragments, U fragments, P fragments, L fragments, and combinations or subcombinations of the foregoing. In general, after cleavage, the U fragments (or U-containing fragments) can still be attached to a hydrogel polymer chain. For example, a U fragment can remain as a monomer unit in a cleaved hydrogel polymer chain, a U-P fragment can remain as a monomer unit with a pendant P group in a cleaved hydrogel polymer chain, etc.

[0123] The photocleavable crosslinker and / or the corresponding photocleaved residue suitably have a sufficiently high water-solubility such that the crosslinker is sufficiently (e.g., completely) dissolved during hydrogel formation, which in turn assists with penetration, immobilization, and eventual expansion of biological tissues by the hydrogel. For example, the photocleavable crosslinker, its corresponding photocleaved residue, including the photocleaved hydrogel, can have a solubility in water of at least 0.1, 0.2, 0.5, 1, 2, 5, or 10 mg / mL (e.g., at 25°C in water, such as in deionized water or other purified reference water medium).

[0124] The photocleavable crosslinker according to the disclosure can have a molecular weight in a range of about 200 g / mol to 10000 g / mol (e.g., measured as number- or weightaverage; or a range bracketing a distribution, such as 1 / 99, 5 / 95, or 10 / 90 cuts of a cumulative distribution). In embodiments, the photocleavable crosslinker (PC) and / or the corresponding linker (L) can have a molecular weight in a range of about 200 g / mol to about 10000 g / mol, such as at least about 200, 400, 600, 800, 1000, 1200, 1500, 2000, 3000, or 4000 g / mol and / or up to about 800, 1200, 1600, 2000, 3000, 4000, 5000, 6000, 8000, or 10000 g / mol. The foregoing values can be number- or weight-average molecular weight; or a molecular weight range bracketing a distribution, such as 1 / 99, 5 / 95, or 10 / 90 cuts of a cumulative distribution.

[0125] According to the disclosure, the crosslinker can have a structure according to Formula (I):Formula (I)

[0126] In embodiments, the number of ethylene oxide units n can be in a range of about 10 to about 1000, whether in the illustrated Formula (I) or in PC compounds with other P and / or II groups. For example, n can be at least about 10, 20, 30, 40, 50, 60, 80, 100, 200, or about 300 and / or up to about 20, 40, 60, 80, 100, 120, 150, 200, 300, 400, 600, 800, or about 1000. The foregoing can represent a number- or weight-average; or a range bracketing a distribution, such as 1 / 99, 5 / 95, or 10 / 90 cuts of a cumulative distribution. Larger values of n can reduce relative water solubility, but such PC compounds can be soluble in various organic solvents. For example, the sample structure could first be embedded in an aqueous medium with the PC and hydrogel monomers to crosslink in the presence of the biomolecules / labels of interest. Once cross-linking has occurred, the hydrogel need not remain in an aqueous environment (e.g., the PC-gel can be switched to an organic solvent or other non-aqueous medium with other desired properties, such as higher refractive index to leverage higher numerical aperture objectives, etc.).Photodeqradable Hydrogels

[0127] Photocleavable crosslinkers of the disclosure can be used to prepare photodegradable crosslinked hydrogels as shown schematically in Fig. 2A and Fig. 2B. A crosslinked hydrogel 101 according to the disclosure can be the product of radical polymerization between (i) the photocleavable crosslinker 104 (e.g., via the ethylenically unsaturated groups thereof) and (ii) one or more hydrogel monomers 103. In Fig. 2A, the reacted photocleavable crosslinker 104 is illustrated as photocleavable crosslinker segments 104A and 104B, and the reacted hydrogel monomers 103 are illustrated as polymer chains 103A and 103B, for example where each chain incorporates hydrogel monomer units and II and / or F monomer units from the photocleavable crosslinker.

[0128] Hydrogel monomers can be ethylenically or vinyl functional, for example, hydrogel monomers can include one or more of (i) N,N-dimethylacrylamide (DMAA) monomerO 0 ®^laI and acrylate monomer, e.g., sodium acrylate O . acrylamide0; and (ii) a PEG-SH monomer and / or a PEG-diacrylamide monomer.

[0129] More generally, the hydrogel monomers can include any monomers forming water-soluble, free-radical-chain-growth polymerized polymers 103 that can be crosslinked by the photocleavable crosslinker (PC), whether containing ethylenically unsaturated group or otherwise for polymerization / crosslinking (e.g., in the PC crosslinker or hydrogel monomers). Examples include other hydrogel systems used for expansion microscopy, such as monomer N,N-dimethylacrylamide acid (DMAA) or PEG-SH + PEG-diacrylamide (e.g., thiol-ene reaction between unsaturated PEG-diacrylamide and thiolated PEG-SH having no unsaturated groups). In cases where the photocleavable crosslinker (PC) 104 incorporates other functional groups (F) for crosslinking, other hydrogel monomers / polymers 103 can be used (e.g., those that do not necessarily involve free-radical-chain-growth polymerization).

[0130] The hydrogel monomers can be selected such that the crosslinked hydrogel 101 is highly hydrophilic, such that it absorbs water and swells up to many times its original size while maintaining mechanical integrity, shape, and optical transparency.

[0131] The crosslinked hydrogel 101 of the disclosure can contain the photocleavable crosslinker 104 in an amount in a range of about 5 to about 60 wt.%, or about 0.1 to about 5 mol.%, or about 0.2 to about 10 eq.% relative to the crosslinked polymerization product (e.g., including the combined amount of the photocleavable crosslinker and the one or more ethylenically functional hydrogel monomers as polymerized / incorporated 103 into the crosslinked polymerization product, without accounting for any residual solvent, i.e. , relative to the non-solvent components of the hydrogel). For example, a monomer solution can contain: 8.625% (w / v) sodium acrylate, 2.5% (w / v) acrylamide, and 4.980% (w / v) PC, 2M NaCI, dissolved in 1x PBS buffer.

[0132] The foregoing ranges represent alternative weight-, molar-, or ethylenic group equivalent-bases for characterizing the relative amount of the photocleavable crosslinker (PC) 104 in the crosslinked hydrogel 101. One, two, or all three of the ranges can apply to a given crosslinked hydrogel 101. For example, the PC can be present in an amount of at least about 5, 10, 15, 20, 25, 30, 35, or 40 wt.% and / or up to about 20, 30, 35, 40, 45, 50, 55, or 60 wt.% relative to the crosslinked polymerization product (e.g., the non-solvent components thereof). Alternatively or additionally, the PC can be present in an amount of at least about0.1 , 0.2, 0.4, 0.7, 1, 1.5, 2, or 2.5 mol.% and / or up to about 1 , 2, 2.5, 3, 4, or 5 mol.% the crosslinked polymerization product (e.g., the non-solvent components thereof). Alternatively or additionally, the PC can be present in an amount of at least about 0.2, 0.4, 0.7, 1, 1.5, 2, 2.7, or 3.5 eq.% and / or up to about 2, 3, 4, 5, 6, 8, or 10 eq.% the crosslinked polymerization product (e.g., the non-solvent components thereof). Complementary ranges can apply to the combined amount of the ethylenically functional hydrogel monomers (e.g., 40 to 95 wt.%, 95 to 99.9 mol.%, or 90 to 99.8 eq.% relative to the crosslinked polymerization product, including various subranges thereof as defined above). Illustrative values for the Basic Protocol in the examples are shown in Table 1 below.Table 1. Basic Protocol Crosslinked Hydrogel Components

[0133] If the concentration of PC is too low (e.g., lower than 5 wt.% relative to the nonsolvent components), the hydrogel may not hold a desired shape, it may be easily deformed and the embedded tissue may not be adequately protected during subsequent expansion, handling, sectioning, and imaging.

[0134] If the concentration of PC is too high (e.g., higher than 60 wt.% relative to the non-solvent components), the hydrogel may be too rigid and consequently it may not expand sufficiently to allow high resolution imaging of a tissue sample embedded within (Fig. 3A and Fig. 3B). Additionally, the degradation time of the hydrogel may be too long, which can result in long exposure times potentially leading to excessive photobleaching and impractical imaging times.

[0135] Hydrogels of the disclosure can contain an acrylamide monomer present in the crosslinked hydrogel in an amount in a range of about 3 to 30 wt.%, about 5 to 50 mol.%, or about 5 to 50 eq.% relative to the crosslinked polymerization product (e.g., relative to thenon-solvent components of the hydrogel). For example, the acrylamide monomer can be present in an amount of at least about 3, 5, 10, 12, 15, or 18 wt.% and up to about 15, 18, 20, 22, 25, 27, or 30 wt.%, relative to the crosslinked polymerization product. Alternatively or additionally, the acrylamide monomer can be present in the hydrogel in an amount of at least about 5, 6, 7, 8, 9, 10, 12, 15, 20, 25, or 30 mol.% and up to about 20, 22, 25, 30, 32, 35, 40, 42, 45, 47, or 50 mol.%, relative to the crosslinked polymerization product. Alternatively or additionally, the acrylamide monomer can be present in an amount of at least about 5, 5.5, 6, 6.5, 7, 8, 9, 10, 12, 15, 17, 20, 22, 25, 27, or 30 eq.% and up to about 22, 25, 27, 30, 32, 35, 37, 40, 42, 45, 47, or 50 eq.%, relative to the crosslinked polymerization product.

[0136] According to the disclosure, the acrylate monomer can be present in the crosslinked hydrogel in an amount in a range of about 10 to 80 wt.%, about 30 to 90 mol.%, or about 30 to 90 eq.% relative to the crosslinked polymerization product (e.g., relative to the non-solvent components of the hydrogel). For example, the acrylate monomer can be present in an amount in a range of at least about 10,15, 20, 22, 25, 27, 30, 32, 35, 37, 40, 42, or 45 wt.% and up to about 37, 40, 42, 45, 47, 50, 52, 55, 57, 60, 62, 65, 67, 70,72, 75, 77, or 80 wt.%, relative to the crosslinked polymerization product. Alternatively or additionally, the acrylate monomer can be present in an amount of at least 30, 33, 35, 37, 40, 42, 45, 50, or 52 mol.% and up to about 45, 47, 50, 52, 55, 57, 60, 62, 65, 70, 72, 75, 77, 80, 82, 85, 87, or 90 mol.%, relative to the crosslinked polymerization product. Alternatively or additionally, the acrylate monomer can be present in an amount of at least about 30, 32, 35, 37, 40, 42, or 45 eq.% and up to about 42, 45, 47, 50, 52, 55, 57, 60, 62, 65, 70, 72, 75, 77, 80, 82, 85, 87, or 90 eq.%, relative to the crosslinked polymerization product.Hydrogel composites

[0137] Hydrogels according to the disclosure are useful in preparing composites to embed, encase or otherwise immobilize a sample, e.g., during imaging and photosectioning operations.

[0138] According to the disclosure and as illustrated in Fig. 2B, Fig. 8A, and Fig. 8B, a hydrogel composite 402 can include a crosslinked hydrogel 101 (as described herein) as a matrix and a sample material 401 immobilized in the matrix (e.g., a biological sample such as cells and / or tissue).

[0139] Sample materials in hydrogel composites 402 according to the disclosure can include mammalian (e.g., human, mouse, or otherwise) tissue or cells such as one or more of brain tissue (e.g., olfactory bulb, hippocampus), kidney tissue, and portions thereof (e.g.,all or less than all of a particular organ or a component of a particular organ can be sampled).

[0140] More generally, the sample material 401 can include any type of biological tissue. For example, the sample material 401 can include insect tissue, plant tissue, microbes, yeast colonies, bacteria (biofilms), archaea, and fungi. The sample can also include molecules, subcellular organelles, single cells, organoids, explants, tissues, and small organisms.

[0141] In some embodiments, the sample material 401 can include one or more non- biological materials. For example, the sample can include one or more of polymers (e.g., gels and plastics) and crystalline materials (e.g., zeolites, metal organic frameworks). In other cases, the hydrogel composite can be used for non-imaging applications such as light- activated, controlled, and targeted release (e.g., location and dose) of gel-linked drugs, chemicals, small molecules, and biomolecules (e.g., vaccine antibodies / nanobodies / mRNA; hormones, growth factors, etc.), where the sample material is the drug, chemical, small molecule, biomolecule, etc. immobilized in the crosslinked hydrogel matrix for subsequent controlled release via photosectioning or other photocleavage mechanism.

[0142] Hydrogel composites 402 according to the disclosure can absorb water so that the sample material contained within can be expanded, preferably linearly expanded, by a factor in a range of about 1.5 to about 10 (e.g., 2-8, 3-7, or 4-5) relative to an original size (or dimensions) of the sample material before being incorporated into the hydrogel composite.

[0143] More generally, the sample material 401 within the composite can be expanded by a factor of at least 1.5, 1.7, 2, 2.5, 3, 3.5, 4, 4.5, or 5 and / or up to 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 7, 8, 9, 10. The expansion factor can represent a linear expansion factor in one dimension that generally would be essentially isotropic in all three dimensions (up to 5% rms error for expansion isotropy can be acceptable) giving an equivalent volumetric expansion values and ranges equal to the foregoing ranges cubed. For example, a sample isotropically expanded by a linear factor of 5 in three mutually orthogonal directions has an equivalent volumetric expansion factor of 125 (i.e. , 5x5x5). The expansion factors in different directions are generally similar but need not be identical. In cases where the linear expansion factors are different in different directions, the volumetric expansion factor can be represented by a product of three linear expansion factors in three mutually orthogonal directions.

[0144] The hydrogel composite 402 can contain a sample material 401 that has been fluorescently stained (e.g., labeled or otherwise bound to one or more fluorophores) to highlight areas of interest marked by the bound fluorophores. Fluorescent stains or labelsand their corresponding fluorophore components can absorb and be excited by incident electromagnetic radiation a particular wavelength (or relatively narrow range of wavelengths around a peak value), whereupon they subsequently emit electromagnetic radiation a different (generally higher) wavelength, and the emitted electromagnetic radiation (e.g., light in the visible spectrum) can be detected, captured, etc. in an imaging process, for example a volumetric optical imaging process according to the disclosure. Fluorophores and corresponding stains or labels are generally known in the art and are commercially available at any of a variety of selectable excitation wavelengths, typically within 400 nm to 800 nm, but more generally within 10 nm to 8000 nm.

[0145] In embodiments, fluorescent staining or labeling can be performed in a two-step process where the sample material 401 is initially bound to a primary antibody, which in turn is bound to a fluorophore-labeled secondary antibody, giving a general structure of tissueprimary antibody-secondary antibody-fluorophore. In other embodiments, fluorescent staining or labeling can be achieved using (1) endogenously expressed fluorescent proteins (FPs) and (2) protein tags (HALOTAG, SNAP-tag, etc.). In other embodiments, the sample can be stained or labeled with known non-fluorescent materials, but such materials can (1) affect the mechanism of readout and / or (2) retention after hydrogel embedding process.

[0146] Hydrogel composites 402 according to the disclosure can contain a fluorophore (e.g., as a component of the fluorescent stain / label) free or bound to the sample. The fluorophore can have one or both of the following characteristics represented by condition (I) and condition (II). For condition (I), the fluorophore has an excitation wavelength at which absorption by the photocleavable crosslinker 104 is not more than 1% or 10% (e.g., fraction of incident light energy absorbed by the photocleavable crosslinker 104 at a specific wavelength or range of wavelengths over which photocleavage can appreciably occur). For condition (II), the fluorophore has an excitation wavelength at least 25 nm or 50 nm different from (e.g., higher or lower than) a cleavage wavelength of the photocleavable crosslinker (or range of wavelengths over which photocleavage can appreciably occur).

[0147] Suitably, the photocleavage wavelengths of the PC are distinct from the excitation wavelengths of any fluorophores used as stains or labels for the sample material. For example, this can be expressed as an absorption threshold (particularly suitable when multiphoton absorption is used for PC cleavage), for example where absorption by the photocleavable crosslinker is not more than 0.0001 , 0.001, 0.01, 0.1, 1, 2, 5, 7.5, or 10% at the excitation wavelengths of the fluorophores used as a fluorescent stain / label (e.g., particularly, considering the relatively short illumination time for fluorophore excitation).Alternatively or additionally, the excitation wavelengths of the fluorophores can be at least 25, 50, 75, 100, 125, 150, 175, or 200 nm different from (e.g., higher or lower than) a cleavage wavelength of the photocleavable crosslinker.

[0148] The absorption threshold and / or wavelength difference between the fluorophore excitation wavelength and the PC cleavage wavelength can equivalently apply to the actual electromagnetic radiation wavelengths used for excitation and cleavage in an imaging method (i.e., as contrasted with characteristic wavelength properties of the components themselves). The absorption threshold is a particularly suitable characteristic when multiphoton absorption is used for PC cleavage. For example, the electromagnetic radiation can be at a wavelength (e.g., peak wavelength or range of wavelengths around a peak) at which absorption by the photocleavable crosslinker 104 is not more than 1% or 10% (or other ranges / values described above). Alternatively or additionally, the electromagnetic radiation can be at a wavelength (e.g., peak wavelength or range of wavelengths around a peak) that is at least 25 nm or 50 nm different from (e.g., higher or lower than; including other ranges / values described above) a cleavage wavelength of the photocleavable crosslinker.

[0149] As illustrated in Fig. 2B, hydrogel composites 402A and corresponding expanded hydrogel composites 402 according to the disclosure can be prepared by first combining an initial (or unexpanded / natural size) sample material 401A with a monomer solution containing the photocleavable crosslinker 104 of the disclosure, and one or more hydrogel monomers 103. The method combining is not particularly limited, but can include one or more of mixing in an aqueous medium, immersing, incubating, etc. The photocleavable crosslinker 104 and the hydrogel monomers 103 are then polymerized, for example via radical polymerization reaction (e.g., when the crosslinker 104 and monomers 103 contain ethylenically unsaturated groups). A resulting crosslinked hydrogel 101A (e.g., unexpanded hydrogel) is formed between the photocleavable crosslinker 104 and the hydrogel monomers 103. The crosslinked hydrogel 101 A serves as a matrix immobilizing the sample material 401A therein, which collectively form the hydrogel composite 402A. The hydrogel composite 402A is then expanded with water or a water-containing (e.g., aqueous) medium, for example by contacting with and / or immersing in water or aqueous buffer for a sufficient time and / or a number of repetitions to swell the hydrogel composite 402A and thereby expand the sample material 401A. As illustrated in Fig. 2B, after swelling and expansion with water, a corresponding expanded hydrogel composite 402 includes an expanded crosslinked hydrogel 101 as a matrix immobilizing the expanded sample material 401 therein. As used herein, references to crosslinked hydrogel, sample material, and hydrogel composite structures can interchangeably refer to an initial state or size / shape of the structure (e.g.,unexpanded or natural size prior to contact with external water or aqueous medium) or a final, expanded state or size / shape of the structure (e.g., after swelling with water or an aqueous medium).

[0150] Fig. 2B also illustrates the expansion factor that is characteristic of the hydrogel composite as well as its crosslinked hydrogel and sample material components, whether as a characteristic of the material itself or a result of its method of making. As shown in Fig. 2B, the initial hydrogel composite 402A and / or crosslinked hydrogel 101 A can have a characteristic linear length dimension Li, which is illustrated as a width, but which also could represent a height, depth / thickness, or other characteristic dimension. Similarly, the initial sample material 401 A can have a characteristic linear length dimension Di , which is illustrated as a diameter of a generally cylindrical or spherical object, but which also could represent a length, height, depth / thickness, or other characteristic dimension. As illustrated after swelling and expansion with water, the expanded hydrogel composite 402 and / or crosslinked hydrogel 101 can have a characteristic linear length dimension L2, and the expanded sample material 401 can have a characteristic linear length dimension D2. A corresponding linear expansion factor between the initial and expanded states can be represented by L2 / L1 (i.e., based on the hydrogel composite or the crosslinked hydrogel expansion) and / or by D2 / D1 (i.e., based on the sample material expansion). In embodiments, the ratio L2 / L1 and / or D2 / D1 can be in a range of 1.5 to 10 or subrange as described above. Similarly, a corresponding volumetric expansion factor can be represented by (L2 / L1)3and / or (D2 / D1)3, for an essentially isotropic expansion. While the hydrogel composite swelling is typically an essentially isotropic expansion, swelling and / or expansion can be controlled to be anisotropic in some embodiments, such as by constraining the expanding hydrogel with one or more solid boundaries, compressing an expanded hydrogel, stretching an expanded hydrogel, etc. For example, swelling in a cylindrical tube, rectangular channel, etc. can induce 1 D expansion, swelling between two flat plates can induce radial or 2D expansion, compression or squeezing (e.g., along a selected axis) of an isotropically expanded gel can induce 2D expansion and 1 D compression, elongation or stretching (e.g., along a selected axis) of an isotropically expanded gel can induce 1 D expansion and 2D compression, etc. Such anisotropic expansion can be performed, for example, to promote more rapid postexpansion immunostaining, selectively increased relative expansion in desired direction(s) (e.g., a 64* volumetric expansion can be represented by a 4x4x4 isotropic expansion or an 8x8x1 anisotropic expansion with greater expansion in a selected 2D plane), etc. In the case of anisotropic expansion, linear expansion factors in three mutually orthogonal directions can be the same or different, for example each being independently in a range of1 to 10, such as at least 1 , 1.2, 1.5, 1.7, 2, 2.5, 3, 3.5, 4, 4.5, or 5 and / or up to 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 7, 8, 9, or 10.

[0151] The initial monomer solution can include one or more polymerization reaction components, such as a radical initiator, a promoter capable of stabilizing the radicals produced by the promoter, and an inhibitor useful to delay polymerization until all reagents diffuse through the sample, for example including APS, TEMED, and 4HT as in the examples herein.

[0152] In some embodiments, the sample material can be incubated in a monomer solution containing the one or more hydrogel monomers and the PC at a low temperature, e.g., 0°C, for at least 20 min, 45 min and up to 3 days before the polymerization reaction begins, to allow the monomer solution to penetrate the sample.

[0153] The sample can be further combined with a monomer solution containing the one or more ethylenically functional hydrogel monomers, the PC, a radical initiator, a promoter, and an inhibitor in a gelation chamber at a low temperature, e.g., 0°C, for at least 20 min, 45 min and up to 60 min before allowing the radical reaction to occur.

[0154] The polymerization reaction can be promoted by exposing the sample combined with the monomer solution to a temperature of at least about e.g., 20, 25, 30, 35, 37, or about 40°C for a time of at least about 1 , 1.5, 2, 2.5, or 3 hr.

[0155] Optionally, the sample can be treated for protein retention prior to combining with the monomer solution. Protein retention can be performed by incubating the sample in a solution containing a protein retention agent, for example succinimidyl ester of 6- ((Acryloyl)amino)hexanoic acid (i.e., Acryloyl-X or AcX) for a time of about 2 hr. to about 3 days or any time suitable to allow penetration of the protein retention agent.

[0156] The sample material 401 can be immunostained or fluorescently labeled before being combined with the monomer solution and before being treated for protein retention.

[0157] The hydrogel composite 402 can undergo osmotic expansion by being immersed in water or other aqueous medium for a sufficient amount of time or a sufficient number of repetitions for a volume increase up to a selected expansion factor as described herein. The expansion factor can differ between expansion media, for example due to different ionic strength.Method of volumetric imaging by photochemical sectioning

[0158] Hydrogel composites of the disclosure can be advantageously used in volumetric imaging of biological samples.

[0159] Fig. 8A, Fig. 8B, and Fig. 10 illustrate a method for volumetric optical (e.g., fluorescent) imaging 600 of a sample material 401 according to the disclosure. A starting hydrogel composite 402 including the sample material 401 and the crosslinked hydrogel 101 for imaging is provided 610, generally in the expanded or swollen form after contact with water or an aqueous medium (Fig. 2B). An image layer 203 of the hydrogel composite 402 is then optically imaged 620 to obtain a volumetric section image of the portion of the sample material 401 (if any) in the top image layer 203. This can include acquiring and storing in a computer medium a volumetric section image or sub-volume image of any sample material 401 in the image layer 203 situated at the topmost or outermost region of the hydrogel composite 402. A depolymerization layer 202 of the hydrogel composite 402 is then photochemically sectioned 630 to remove the depolymerization layer 202 from the hydrogel composite 402. The imaging 620 and sectioning 630 steps are repeated 640 a plurality of times to provide sequential image layers 203, sequential depolymerization or photodegradation layers 202, and subsequent volumetric section images suitable for forming a composite volumetric image of the sample material 401 in the hydrogel composite 402. The method can further include a stacking step 650 of the sequential image layers to form a composite 3D or volumetric image of the sample material 401. As used herein, “top” as applied to the image layer 203 and the depolymerization layer 202 can reference any outer or external layer of the hydrogel composite 402, whether as initially provided in the imaging process or in its current form after depolymerization and removal of one or more depolymerization layers 202.

[0160] Removal of the depolymerization layer can occur via one or a combination of mechanisms (Fig. 8A, bottom panel): (i) via photodegradation of the layer, where essentially the entire irradiated layer of the hydrogel composite (depolymerization layer) degrades from the block-face inward up to a photodegradation bottom boundary or photosectioning plane 204; and / or (ii) via photoslicing, where a thin plane or layer (e.g., generally having a thickness corresponding to that of the irradiating light) of the hydrogel composite beneath the surface (at the photodegradation bottom boundary 204) is degraded allowing the overlying sample block to be physically removed, detached, or floated away by the surrounding medium 410. For photodegradation, a bulk (remaining) hydrogel composite 402’ includes a new image layer 203’ and a new depolymerization layer 202’ for a subsequent volumetric imaging cycle. A top portion of the hydrogel composite 402’ corresponds to an overlap volume 203” with the previous image layer 203 which allows for subsequent computationalstacking and assembly of a composite image. For photoslicing, the bulk (remaining) hydrogel composite 402’ is essentially the same as that for photodegradation, with the difference being the formation of a cleaved hydrogel composite section 202” essentially corresponding to the previous depolymerization layer 202 (less the thin, degraded photoslicing layer), which can include crosslinked hydrogel, immobilized sample material therein, etc. The photochemical sectioning process generally forms free, water-soluble cleaved PC residue, free non-crosslinked hydrogel polymer chains, and free sample material / biological tissue from the depolymerization layer 202 (if any sample material were present therein) at the point or plane of cleavage / photodegradation bottom boundary 204. In the case of photodegradation, additional free cleaved PC residue, non-crosslinked hydrogel polymer chains, and sample material formed from the bulk of the depolymerization layer 202. In the case of photoslicing, the photochemical sectioning process also provides the cleaved (but otherwise intact) hydrogel composite section 202” that is no longer connected to the bulk (remaining) hydrogel composite 402’. The freed materials and cleaved hydrogel composite section 202” (when formed) then dissipate / diffuse / dissolve into a surrounding aqueous medium, thus exposing a new top or external surface for volumetric imaging that corresponds to the previous photodegradation bottom boundary 204. For example, as illustrated in Fig. 2A (top right), a resulting photodegraded or depolymerized hydrogel 102 includes cleaved PC residues 104C and 104D, and cleaved, non-crosslinked hydrogel polymer chains 103C and 103D. Similarly, as illustrated in Fig. 4, an initial crosslinked hydrogel is present in an aqueous medium (left), the crosslinked hydrogel is then photosliced, i.e., spatially and selectively depolymerized in a thin layer at the “UV” location (middle), and the depolymerized hydrogel components and a cleaved hydrogel composite section fall away into the surrounding aqueous medium (right), thus creating a new “top” or outermost layer of the crosslinked hydrogel for a subsequent imaging step.

[0161] The image layer 203 can have a thickness greater than a thickness of the depolymerization layer 202, e.g., such that sequential image layers have enough overlap 404 (Fig. 8A and Fig. 8B) to facilitate computational assembly of a composite 3D / volumetric image of the entire sample material 401.

[0162] The image layer 203 can have a thickness (or depth) Ti of about 0.1 to 2.6 mm or about 1.2 to 2 mm, for example at least 0.1 , 0.2, 0.4, 0.6, 0.8, 1 , 1.2, 1.4, 1.7, or 2 mm and / or up to 0.5, 0.7, 0.9, 1.3, 1.6, 2, 2.3, or 2.6 mm. The depolymerization layer 202 can have a thickness (or depth) T 1 of about 0.05 to 2.2 mm or about 0.7 to1.4 mm, for example at least 0.05, 0.1, 0.2, 0.4, 0.5, 0.7, 1, 1.2, or 1.5 mm and / or up to 0.5, 0.7, 0.9, 1.2, 1.4, 1.7, 2, or 2.2 mm. The image layer 203 can be thicker than the depolymerization layer 202 by afactor of about 1.1 to 2 (i.e., 10%-100% thicker), such as by a factor of at least about l.1 , 1.2, 1.3, 1.4, 1.5, or 1.7 and / or up to about 1.2, 1.3, 1.4, 1.6, 1.8, or 2 for Ti / T2. Alternatively or additionally, the image layer 203 can be thicker than the depolymerization 202 layer by about 0.2 to 1.4 mm or about 0.4 to 1 mm, such as at least 0.2, 0.4, 0.6, 0.8, or 1 mm and / or up to 0.5, 0.6, 0.7, 0.8, 0.9, 1, 1.2, or 1.4 mm for Ti - T2.

[0163] In some embodiments, the hydrogel composite 402 is divided into multiple imaging sub-volumes 403, each including multiple image sublayers 405 (Fig. 8B). The image sublayers 405 are imaged by varying the focus of the objective before capturing the corresponding image and include an in-between image sublayers overlap 406. The thickness of the photodegradation layer 202 can typically be larger than the thickness of a single image sublayer 405 and smaller than the thickness of a sub-volume 405 to allow for enough overlap 404.

[0164] A volumetric section image or a sub-volume image can have voxels having a length dimension in a range of about 5 nm to 300 nm measured previous to expansion, or in a range of about 5 nm to about 200 nm, or about 5 nm to about 100 nm. One, two or three mutually orthogonal voxel dimensions can be generally in this range and can have the same or different individual values.

[0165] More generally, the voxels can have up to three independent voxel length dimensions (i.e., edge lengths), which can be the same or different, that are at least 5, 10, 20, 30, 40, 50, 60, or 70 nm and / or up to 30, 40, 50, 60, 70, 80, 90, 100, 150, 200, 220, 250, or 300 nm. The voxel dimensions / imaging resolution values can represent effective voxel size accounting for the expansion ratio (e.g., physical imaging voxel size divided by expansion factor). For example, given a (50 nm)3voxel resolution with a hydrogel isotropic linear expansion factor of 5, the physically imaged voxel size is a (250 nm)3volume in the hydrogel composite, which is then size adjusted to (50 nm)3.

[0166] Optically imaging 620 the image layer 203 can be performed by on-block fluorescence imaging using an optical microscope, for example a confocal microscope or a multiphoton microscope, to image a pre-determined volumetric depth (e.g., image layer 403 or image sub-layer 405) from an external top (or block-face) surface. Optical images can be captured or transferred from the microscope to a digital camera (e.g., a CCD camera) and stored in a computer medium for further processing and stitching to form a 3D / volumetric image of the sample within the hydrogel composite.

[0167] Optically imaging 620 the image layer 203 can include exposing the image layer 203 to electromagnetic radiation at one or more excitation wavelengths in a range of about400 nm to 800 nm. Preferably, the one or more excitation wavelengths correspond to the excitation wavelengths of each different fluorophore in the fluorescently stained sample. The electromagnetic radiation can be applied as a single-wavelength light, for example from a laser (e.g., a gas laser, or a solid-state laser), or as broadband illumination (e.g., from LEDs, halogen lights, arc light sources). Photochemically sectioning 630 the depolymerization or photodegradation layer 202 can include exposing at least a portion of the image layer 203 to electromagnetic radiation with a photocleavage wavelength in a range of about 100 nm to about 450 nm. Relative to the excitation wavelength(s) of the electromagnetic radiation, the photocleavable crosslinker can have one or both of the following characteristics represented by condition (I) and condition (II). For condition (I) absorption of the electromagnetic radiation by the photocleavable crosslinker is not more than 1% at the one or more excitation wavelengths. For condition (II), the one or more excitation wavelengths are each at least 50 nm different from (e.g., higher or lower than) the photocleavage wavelength.

[0168] More generally, the excitation wavelengths and the photocleavage wavelengths can be independently selected from wide ranges, such as at least about 10, 100, 270, 280, 315, 380, 400, 410, 420, 450, 485, 500, 565, 590, 625, 750, 800, 1400, or 3000 nm and / or up to about 100, 270, 280, 315, 380, 400, 410, 420, 450, 485, 500, 565, 590, 625, 750, 800, 1400, 3000, or 8000 nm. The foregoing wavelength values can represent a peak intensity value for light at a single wavelength or light distributed across a range of wavelengths, such as from a laser, a lamp, an LED light source, etc.

[0169] Suitably, the photocleavage wavelengths of the PC are distinct from those of excitation wavelengths for any fluorophores used as stains or labels for the sample material. For example, this can be expressed as an absorption threshold (particularly suitable when multi-photon absorption is used for PC cleavage), for example where absorption by the photocleavable crosslinker is not more than 0.0001, 0.001 , 0.01, 0.1 , 1, 2, 5, or 10% at excitation wavelengths of the fluorophores used as a fluorescent stain / label (e.g., also considering the relatively short illumination time for fluorophore excitation). Alternatively or additionally, the excitation wavelengths of the fluorophores can be at least 25, 50, 75, 100, 150, or 200 nm different from (e.g., higher or lower than) a cleavage wavelength of the photocleavable crosslinker.

[0170] Photochemically sectioning 630 the top depolymerization layer 202 can include exposing at least a portion of the image layer 203 or sub-volume 403 to a multi-photon light source (e.g., two-photon, three-photon, etc. illumination for photocleavage via multi-photon absorption (MPA)).

[0171] The electromagnetic radiation and / or multi-photon light source used in photochemical sectioning can be a focused light beam scanned across the depolymerization layer 202 such as in a confocal or a two-photon microscope (Fig. 5). Alternatively, the electromagnetic radiation and / or multi-photon light source can be a lattice light sheet generated by optically expanding light beams from one or more point light sources as described in the examples (Fig. 7) illuminating a top section of the depolymerization layer.

[0172] The depolymerization layer 202 that is depolymerized or photodegraded in a photochemically sectioning step 630 has a thickness smaller than the image layer 203 or sub-volume 403 to maintain sufficient overlap 404 between subsequent image layers 203 and / or sub-volumes 403 to allow for reconstructing the full thickness of the sample material 401.

[0173] According to the disclosure, the hydrogel composite 402 can be maintained in an aqueous environment 410 (Fig. 8B) before and during volumetric imaging, more particularly during the depolymerization step.

[0174] The aqueous environment can be a bath or a reservoir, with a volume typically at least 10x volume relative to the initial or starting hydrogel composite 402 size (e.g., after swelling and expansion). The hydrogel composite can be positioned on or mounted to a bottom support substrate (e.g., glass, a glass slide or other optically transparent material) and further immersed in the reservoir. The aqueous environment suitably can include water, aqueous buffers (e.g., 1x PBS, 10x PBS, 1 mM Tris), and weak aqueous base solutions (e.g., 1 mM NaOH) as an imaging medium. The aqueous environment can change the expansion factor (i.e. , the volume can decrease or increase) depending on the properties of the environment, for example the ionic strength.

[0175] A composite 3D / volumetric image of the sample material in the hydrogel composite can be formed by stacking the images of the image layers 650. The stacking process can include stitching lateral sections of the same imaging layer to create a larger image in the x-y plane before stacking images along the z-direction. The images can be deskewed, filtered and / or color corrected before or after stitching. The composite 3D / volumetric image can be visualized using methods and software known in the art. The composite 3D / volumetric image can be further processed to isolate or selectively display and / or measure individual structures or selected elements or portions of the sample material.

[0176] According to the disclosure, an imaging kit 500 can include the photocleavable crosslinker 104 of the disclosure and other components 501 useful in the preparation of hydrogel and hydrogel composites (Fig. 9) These components 501 can include: one or moreof hydrogel monomers (e.g., ethylenically or vinyl functional hydrogel monomers), polymerization reagents, fluorescent stain reagents, light sources, gelation chambers, and brushes (e.g., paint brushes or other tools to handle gels or hydrogel composites) used to prepare and / or image hydrogel composites.

[0177] The polymerization reagents can include initiators, promoters, and / or inhibitors of the polymerization reaction. The fluorescent stain reagents can include primary and secondary antibodies that can be further bound to a fluorophore. A plurality of fluorescent stain reagents (e.g., 2, 3, 4, or more) can be included, for example with each reagent having an excitation wavelength distinct from the excitation wavelengths of the other reagents, such as being at least 25, 50, 75, 100, 125, 150, 175, or 200 nm different from (e.g., higher or lower than) each other excitation wavelength. Light sources include sources for photosectioning and for fluorescence excitation; the same light source tuned to different wavelengths can be used. Gelation chambers can include mold or other containers suitable to accommodate different sizes of samples.EXAMPLES

[0178] The following examples are provided for illustration and are not intended to limit the scope of the invention. In general, these examples illustrate the preparation of photodegradable hydrogels, embedding and swelling of biological samples in photodegradable hydrogels and volumetric imaging of selected biological samples using the gels, photochemical sectioning and imaging techniques of the disclosure.

[0179] These examples show the advantages of the disclosed methods in producing high resolution images that are useful in testing biological hypotheses and producing insights that are not attainable by current physical sectioning methods.

[0180] In particular, examples 6 and 7 show volumetric reconstruction of the OB of wild type and NPC1 mice. The axons and myelin sheaths were resolved, demonstrating that methods herein disclosed can capture neural anatomies and pathological abnormalities across whole-mount tissue. In the wild-type mouse, the axon topography analysis revealed that the number of segmented axons traversing the interfaces between adjacent OB layers remained constant throughout the IPL-GCL, EPL / MCL-IPL, and GL-EPL / MCL interfaces. In the NPC1 OB, on the contrary, this number progressively diminished from the IPL-GCL to the EPL / MCL-IPL and then to the GL-EPL / MCL interfaces, highlighting the degeneration in the NPC1 OB vs. wild-type. In a separate image analysis, the axon density and percent myelinated axon analysis revealed that in the NPC1 OB, the GCL axonal density was unexpectedly higher while the proportion of myelinated axons fell, this could indicate newlysprouted axons, which are typically unmyelinated and potentially point to an attempt of the nervous system to restore neural circuit connectivity, a hypothesis that has been proposed but yet to be fully understood. Additional imaging studies, for example, of endosomal and lysosomal pathways, oligodendrocytes, synapses, and to quantify glial activation and synapse density alongside axonal changes could complement the shown VIPS data and refine the current understanding of the interplay between neuronal loss and reactive plasticity.

[0181] Materials: Unless otherwise noted, chemicals and reagents were obtained from Millipore Sigma. Purified water was obtained from a Mili-Q IQ 7000 Ultrapure Water System (Millipore Sigma).

[0182] Dimensions: Unless otherwise specified, all the dimensions are provided at postexpansion scale.Preparation of cells and biological tissuesHuman embryonic kidney 293 cells (HEK 293)

[0183] HEK 293 cells (Thermo Fisher or ATCC) were cultured on a 12-mm round coverslip to a confluency of 80-90%. The cells were then fixed with 4% (w / v) paraformaldehyde (PFA) in 1x PBS for 10 min, washed with 1x PBS twice for 5 min each time, and stored in 1x PBS with 0.03% (w / v) sodium azide. For immunostaining, the cells were permeabilized with 0.1% (w / v) Triton X-100 in 1x PBS for 15 min and incubated in the blocking buffer [5% (v / v) normal goat serum (NGS, Jackson ImmunoResearch) and 0.1% (w / v) Triton X-100 in 1x PBS] for 15 min. The permeabilized and blocked cells were then incubated in the primary antibody (rabbit anti-beta-tubulin antibody, ab6046, Abeam; see Table 1) solution (1:200 dilution with the blocking buffer) at 4°C for 12 hours. Next, the cells were washed with blocking buffer four times for 5 min each time and incubated in the secondary antibody (goat Alexa Fluor 488-conjugated anti-rabbit antibody, A11008, Thermo Fisher or goat ATTO 647N-conjugated anti-rabbit antibody, 40839-1 ML-F, Millipore Sigma) solution (1:200 dilution with the blocking buffer) at room temperature for 2-3 hours. Finally, the cells were washed with 1x PBS four times for 5 min each time and stored in 1x PBS for the subsequent gelation procedure.Mouse brain slices

[0184] 9-week C57BL / 6 mice were anesthetized using ketamine hydrochloride injectable solution (100 mg / mL, Covetrus) and AnaSed Injection (xylazine) sterile solution (20 mg / mL, Akorn, Inc) and transcardially perfused with 4% (w / v) PFA in 1x PBS (10 mL). The brainswere carefully dissected from the skull, post-fixed with 4% (w / v) PFA in 1x PBS at 4°C for 1 day, and stored in sucrose buffer (20% sucrose in 1x PBS) at 4°C. The fixed brains were sectioned using a vibratome (VT 1200S, Leica) or flash-frozen with dry ice and then sectioned using a microtome (Model 860, AO Scientific Instruments) to ~40 pm coronal slices. All the sectioned brain slices were stored in 1x PBS with 0.03% (w / v) sodium azide at 4°C.

[0185] For immunostaining, the ~40 pm brain slices were permeabilized with 0.1% (w / v) Triton X-100 in 1x PBS for 15 min and incubated in the blocking buffer [5% (v / v) NGS and 0.1% (w / v) Triton X-100 in 1x PBS] at room temperature for more than 6 hours. The brain slices were then incubated in the primary antibody (rabbit anti-NF-200, N4142-.2ML, Millipore Sigma and chicken anti-MBP antibody, PA1-10008, Thermo Fisher, or rabbit antiHomer 1, 160003, Synaptic Systems) solution (1:200 dilution with the blocking buffer) at 4°C for 2 days. Next, the brain slices were washed with the blocking buffer four times for 30 min each time, and incubated in the secondary antibody (goat Alexa Fluor 488-conjugated antirabbit antibody, A11008, Thermo Fisher and goat Alexa Fluor 568-conjugated anti-chicken antibody, A11011, Thermo Fisher, or goat ATTO 647N-conjugated anti-rabbit antibody, 40839-1 ML-F, Millipore Sigma, respectively) solution (1:200 dilution with the blocking buffer) at 4°C for 2 days. Finally, the immunostained mouse brain slices were washed with 1x PBS (or the blocking buffer) four times for 30 min each time and stored in 1x PBS for the subsequent gelation procedure.Human Hippocampus

[0186] Postmortem human hippocampus specimens were obtained from the University of Washington BioRepository and Integrated Neuropathology (BRalN) laboratory and the University of Washington Alzheimer’s Disease Research Center (ADRC) Precision Neuropathology Core. Briefly, a fresh piece of posterior hippocampus was dissected at rapid autopsy (postmortem interval <12 hours) and immediately fixed in 4% (w / v) PFA in 1x PBS at room temperature for 48 hours. The fixed tissue block was sectioned to -300-500 pm slices using a vibratome (VT1200S, Leica) and stored in 1x PBS with 0.03% (w / v) sodium azide at 4°C.

[0187] The human hippocampus slices were immunostained using the same protocol as the mouse OBs with minor modifications. After primary and secondary antibody staining, lectin staining was performed at 37°C for 3 days using Lycopersicon esculentum (Tomato) lectin (L-1170-2, Vector Laboratories) conjugated with SeTau-647-NHS (K9-4149, SETA BioMedicals) (1:300 dilution with the staining buffer). The stained slices were washed withthe PTwH buffer twice for 1 hour each time and stored in 1x PBS for the subsequent gelation.Mouse olfactory bulbs (OBs)

[0188] Heterozygous Balb / c Np(lh(Npc1+ / ~) mice were obtained from Jackson Laboratories (RRID: IMSR JAX:003092) and a breeding colony was maintained. 7-week wild-type Balb / c (J\lpc1+ / +) (“WT”) and mutant (Npcr'-) mice (“NPC1”), male, were anesthetized using isoflurane inhalation or an injection of ketamine hydrochloride injectable solution (100 mg / mL, Covetrus) and AnaSed Injection (xylazine) sterile solution (20 mg / mL, Akorn, Inc), and transcardially perfused with 4% (w / v) PFA in 1x PBS (10 mL). The brains were carefully dissected from the skull, post-fixed with 4% (w / v) PFA in 1x PBS at 4°C for 1 day, and stored in a sucrose buffer (20% sucrose in 1x PBS) at 4°C. The olfactory bulbs (OBs) were dissected from the rest of the brain and stored sucrose buffer at 4°C.

[0189] The dissected mouse OBs were immunostained using a modified iDISCO+ protocol as detailed on the continuously-updated website http: / / idisco.info. First, the OBs were washed with 1x PBS four times for 30 min each time, dehydrated using a methanol (Thermo Fisher) / 1x PBS gradient (from 0% methanol, 20% methanol, 40% methanol, 60% methanol, 80% methanol to 100% methanol, 1 hour incubation in each solution), and incubated in methanol at 4°C overnight. The 100% methanol solvent was then replaced with a 66% DCM / 33% methanol solution. Next, the OBs were incubated in the 66% DCM / 33% methanol solution at room temperature overnight with shaking, washed with 100% methanol twice, and rehydrated with a methanol / 1x PBS gradient (from 100% methanol, 80% methanol, 60% methanol, 40% methanol, 20% methanol, to 0% methanol, 1 hour incubation in each solution) at room temperature. Subsequently, the OBs were washed with PTx.2 buffer [0.2% (v / v) Triton X-100 in 1x PBS] at room temperature twice for 1 hour each time.

[0190] The pretreated OBs were then permeabilized with a permeabilization buffer [80% (v / v) PTx.2, 20% (v / v) DMSO, 2.3% (w / v) Glycine] at 37°C for 36 hours, and incubated in a blocking buffer [84% (v / v) PTx.2, 6% (v / v) NGS, 10% (v / v) DMSO] at 37°C for 36 hours. Next, the OBs were incubated in the primary antibody (chicken anti-MBP antibody, PA1- 10008, Thermo Fisher and rabbit anti-NF-200, N4142-.2ML, Millipore Sigma) solution [1:100 dilution with the staining buffer, which consists of 92% (v / v) PTwH buffer, 3% (v / v) NGS, and 5% (v / v) DMSO] at 37°C for 3 days. The PTwH buffer [1x PBS, 0.2% (v / v) Tween-20, 0.001% (w / v) Heparin] was prepared in advance. The primary antibody solution was then replaced with a freshly prepared primary antibody solution, and the OBs were further incubated at 37°C for 3 days. Finally, the OBs were washed with the PTwH buffer four to fivetimes for 1 hour each time (with the last washing being overnight) and incubated in the secondary antibody (goat Alexa Fluor 488-conjugated anti-rabbit antibody, A11008, Thermo Fisher and goat Alexa Fluor 568-conjugated anti-chicken antibody, A11011, Thermo Fisher) solution (1:100 dilution with the staining buffer) at 37°C for 3 days. The secondary antibody solution was replaced with a freshly prepared secondary antibody solution, and the OBs were further incubated at 37°C for 3 days. The immunostained OBs were washed with the PTwH buffer four times for 1 hour each time and stored in 1x PBS for subsequent gelation.Basic gelation, digestion and expansion protocol

[0191] Unless otherwise noted, photodegradable samples were prepared using a custom protein-retention expansion microscopy protocol with the photocleavable crosslinker (PC). Briefly, fixed and immunostained tissue samples were incubated in a diluted AcX solution (0.1 mg / mL in 1x PBS) at room temperature overnight. The samples were then washed with 1x PBS twice for 15 min each time and incubated in the monomer solution at 4°C overnight. The monomer solution contained: 1x PBS, 2 M NaCI, 8.625% (w / v) sodium acrylate (Pfaltz&Bauer), 2.5% (w / v) acrylamide, 4.980% (w / v) PC.

[0192] Concentrated stock solutions of ammonium persulfate (APS) [10% (w / v)], tetramethylethylenediamine (TEMED) [10% (w / v)], and 4-hydroxy-2, 2,6,6- tetramethylpiperidin-1-oxyl (4HT) [0.5% (w / v)] were added to the monomer solution to a final concentration of 0.2% (w / v), 0.2% (w / v), and 0.01% (w / v), respectively, to yield the gelling solution.

[0193] Samples in the monomer solution were immediately incubated in the mixed gelling solution at 0°C for 45 min, transferred to a gelation chamber of various shapes and sizes, and gelled in a humidified 37°C incubator for 2 hours (“PC-gel”). The gelled samples were trimmed and immersed in a digestion buffer [1 mM EDTA, 0.5% Triton X-100, 1 M NaCI, 8 units / mL Proteinase K (proK, New England Biolabs) in 1x PBS, no Tris] at room temperature overnight. The digested samples were stored in 1x PBS at 4°C.

[0194] The stored samples were expanded in purified water three times for 20 min each time before imaging and photodegradation.

[0195] Non-photodegradable samples were prepared using the standard proteinretention expansion microscopy protocol as described above, but replacing the photocleavable crosslinker bis-acrylamide as the crosslinker (Bis-gels).Fluorescently labeled, photodegradable blank gel

[0196] Fluorescently labeled, photodegradable blank gel (blank PC-gel) was prepared using a modified basic protocol. The monomer solution from the basic protocol was mixed and incubated with an AcX and an Alexa Fluor 488 amine (218C0, Lumiprobe) solution to final concentrations of 2.1 and 5 mg / mL, respectively, at room temperature for more than 1 hour under agitation. Stock solutions of APS and TEMED were added to the mixed monomer solution to final concentrations of 0.2% (w / v) and 0.2% (w / v), respectively, to yield the gelling solution. The gelling solution was then transferred to gelation chambers of various shapes and sizes and gelled in a humidified 37°C incubator for 2 hours. Finally, the synthesized gels were stored in 1x PBS at 4°C for subsequent use.Sample preparation for light-sheet imaging and photodegradation

[0197] The following procedures were used to prepare the samples imaged in Examples 5-7 below.

[0198] Photodegradable mouse OB and human hippocampus samples were expanded in 1x PBS and purified water, respectively, trimmed, and mounted on a cleaned 25 mm coverslip using the superglue mounting method with minor modifications (48). The 25 mm coverslips were soaked in 1M KOH in water for 30 min, rinsed with purified water three times, stored in 30% (v / v) ethanol in water, and air-dried before the sample mounting.

[0199] A modified superglue mounting method was used to immobilize the photodegradable mouse OB and human hippocampus samples to the 25 mm coverslip as it provided stronger and long-lasting adhesion. Briefly, a thin layer of superglue was applied to a small area of the cleaned 25 mm coverslip. Excessive liquid around the expanded sample was wicked away before the samples were placed on the applied superglue. After the gel was glued to the coverslip, it was submerged under 1x PBS or purified water for several hours for further curation of the superglue and for soaking away the impurities from the superglue. After curing, an opaque interface was formed between the gel and the superglue.

[0200] A modified poly-L-lysine mounting method was used to immobilize other samples (e.g., -100-300 pm thick) to the 25 mm coverslip (48). Briefly, a few droplets of 0.1% (w / v) poly-L-lysine aqueous solution were applied to the top surface of the cleaned 25 mm coverslip for 20 min. The poly-L-lysine modified surface was rinsed with purified water three times and air-dried for 1 hour in a clean environment. Excessive liquid around the expanded samples was wicked away before the samples were placed on the poly-L-lysine modified surface of a 25 mm coverslip. After 20-30 s, a few droplets of 1x PBS or purified water were added to keep the sample hydrated.

[0201] Following sample immobilization onto the 25 mm coverslip, a 100 pL volume of 1 :1000 diluted fluorescent bead solution [amine-terminated FluoSpheres, 0.2 pm, green fluorescent (488 / 515), Thermo Fisher] was applied to the sample surface for lattice lightsheet autofocusing. The coverslip was subsequently secured to the lattice light-sheet microscope sample holder using either superglue or metal clips. Finally, the sample holder was transferred to a sample chamber containing either 1x PBS or purified water.Lattice light sheet setup

[0202] Sequential on-block volumetric lattice light-sheet imaging and light-sheet photochemical sectioning was performed using a lattice light-sheet microscope schematically shown in Fig. 7. Briefly, three lasers 309B, 309C, 309D were expanded to 1 / e2diameter of 2.0 mm, combined into one path, these included a 488 nm laser (500 mW, 2RU- VFL-P-500-488-B1R, MPB Communications Inc.), a 560 nm laser (1000 mW, 2RU-VFL-P- 1000-560-B1 R, MPB Communications Inc.), and a 642 nm (2000 mW, 2RU-VFL-P-2000- 642-B1R, MPB Communications Inc.) laser. The expanded light was passed onto an acousto-optic tunable filter 301 (AOTF; ACTFnC-400.650-CPCh-TN, Quanta-Tech, AA Opto Electronic). The collimated beam was fanned out to uniformly expand in the xopticai axis using a Powell lens (LQCP-8.9R20-2.0, Laserline Optics Canada). The zopticai axis was expanded using a pair of 50- and 250-mm cylindrical lenses (25 mm diameter; ACY254-050, LJ1267RM-A, Thorlabs). The expanded beam illuminated a horizontal stripe on a grayscale spatial light modulator 303 (SLM; AVR17-0105, Meadowlark Optics, AVR Optics). The light diffracted by the SLM 303 was focused onto a mask containing user-selected annuli of numerous sizes (Thorlabs Imaging) to block unwanted DC and higher diffraction orders. The light passing through the mask was reflected off a pair of galvanometer mirrors (6SD11226 and 6SD11587, Cambridge Technology, Novanta Photonics), which were conjugated to the back pupil of the excitation objective 305 (TL20X-MPL, Thorlabs) and used to scan along the xopticai and zopticai axes. This lead to a lattice light sheet of -200 pm wide along the image y- axis. The fluorescence generated by the specimen was collected through the detection objective 306 (20x, 1.00 NA, 1.8 mm working distance, 421452-9800-000, Zeiss), projected onto a pupil-conjugate deformable mirror 307 (DM; DM69, ALPAO) that corrects system aberrations and simultaneously imaged onto two sCMOS cameras 308A and 308B (ORCA Fusion with 2304 x 2304 pixels, Hamamatsu Photonics). Appropriate dichroic (T600dcrb, Chroma) and emission filters (FF03-525 / 50-25 and FF01 -538 / 685, Semrock) were used to separate fluorescent signals into the two cameras. An EM drive (MLS-3252 Electromagnetic Direct-Drive, SmarAct) was used to ensure a smooth and continuous movement of thesample holder along the x direction for the “sample scan” imaging acquisition. The y and z movements were respectively controlled by two linear stages (SLS-5252, SmarAct).

[0203] For light-sheet photochemical sectioning, a 405 nm light-sheet was introduced to the imaging region horizontally to the sample coverslip via a separate optical path. The 405 nm laser beam 309A (100 mW, iBeam-Smart-405-S-BZ-1, Toptica) was expanded to a 1 / e2diameter of 2.0 mm, laterally fanned out to uniformly expand using a Powell lens 304 (LOCP-8.9R20-2.0, Laserline Optics Canada), and axially compressed using 50- and 250- mm cylindrical lenses 302A and 302B. This resulted in the formation of a light sheet that had a thickness ranging from ~70 to 100 pm over an area of ~18 x 18 mm, which was sufficient to cover the region of the sample that needed to be photodegraded while maintaining a desirable thickness for precise photochemical sectioning.Light-sheet imaging and photodegradation method

[0204] Lattice light-sheet imaging of PC-gel embedded samples was performed in the sample scan mode where the sample was translated continuously in the plane of the coverslip to allow for increased field of view and long-range fast scanning. To cover a large sample volume, a rectangular parallelepiped was defined as the hard limits for the tiled volume. The microscope software covered the volume with a 3D matrix of rectangular tiles with a desired tile overlap of 10 pm in the sample Y axis and 3 pm in the sample Z axis.

[0205] Before imaging, autofocus was performed on a 0.2 pm diameter fluorescent bead located on the sample surface. During imaging, autofocus was performed on a puncta fluorophore in the sample every ~8 hours to account for small system drifts. During each autofocus measurement, the bead or the puncta was precisely located using a normal imaging volume sweep. The light-sheet was statically held at the bead, while the sample stage was swept along the axis of the detection objective. The fluorescence intensity as function of a piezo position was fitted with a Gaussian curve and the peak center gave the correct piezo offset to use.

[0206] The lattice light-sheet applied in this study was a dithered HexRect lattice with a center numerical aperture of 0.25 and a Gaussian bounding factor of 0.08. This resulted in an effective field-of-view of -200 pm in the image y direction and -70 pm in the image x direction. The microscope resolution for signals emitted at a wavelength of 500 nm was -250 x 250 x 520 nm.

[0207] When possible, an additional sample mounting procedure was implemented to increase the imaging depth per cycle and minimize the deformation caused by scatteredphotodegradation light to the unimaged volume. Upon mounting the sample-laden coverslip onto the sample holder, the narrow axis of the sample was aligned parallel to the scan direction. This alignment situated the sample between the two objectives, thereby permitting a larger imaging depth as opposed to positioning the sample with its long axis parallel to the scanning direction.

[0208] The effective voxel size for lattice light-sheet imaging is given as the effective sampling voxel size at the pre-expansion scale, obtained by dividing the camera pixel pitch by the optical magnification and then by the expansion factor.

[0209] In the case of mouse olfactory bulb (OB) samples, which had an expansion factor of -2 and a lateral scan range of ~6 mm, it was possible to reach an imaging depth of -800 pm without the gel contacting the objectives. In practice, the imaging depth was limited to -600 pm from the gel block-face. After imaging every sub-volume, light-sheet photochemical sectioning with the 405 nm laser was performed up to -400 pm above the bottom plane of the imaged sub-volume. This -400 pm distance was selected to minimize the impact of the scattered photodegradation light on the underlying sample while maintaining a reasonable thickness for each sub-volume. The light-sheet illumination was applied for -2-3 hours to complete the photodegradation. Subsequent imaging of the next sample sub-volume commenced from -150-200 pm above the bottom of the previously imaged one, ensuring sufficient overlap between the imaged sub-volumes, and continued until the imaging reached a total thickness of -600 pm. This overlap served as a reference for the computational stitching pipeline. In summary, each imaging-photochemical sectioning cycle encompassed a volume with a total thickness of -600 pm, including a -150-200 pm thick overlap with the preceding sub-volume and -400-450 pm from the current sub-volume.

[0210] A human hippocampus sample was imaged in purified water with an expansion factor of -4.4 using similar experimental settings as the mouse OB samples. Briefly, each imaged sub-volume extended across -250 x 230 x 300 pm, including a -100 pm overlap with the preceding one. A total of 8 sub-volumes were imaged, resulting in a total volume of -250 x 230 x 1630 pm (-57 x 52 x 370 pm at pre-expansion scale). Between each round of imaging, the same light-sheet illumination with a 405 nm laser was applied for -20 minutes to complete the photodegradation.Image analysis and visualization methodsFlat-field correction, deconvolution, deskewing / rotation and stitching

[0211] First, image tiles within an imaged sub-volume were stitched together into a subvolume using PetaKit5D software with flat-field correction applied during this step. The background was estimated for each sub-volume by averaging the blank frames from all tiles within the sub-volume. The flat-field image was then estimated using the xy maximum intensity projections (MIPs) from all tiles along with the estimated background image through the BaSiC software.

[0212] After initial stitching, masks for the xy, xz, yz MIPs were generated from the stitched sub-volume MIPs using user-defined thresholds to delineate object boundaries. Then, each sub-volume was deconvolved with experimentally measured point-spread- functions (PSFs), followed by deskewing / rotation. Both processes used the large-scale processing strategies in PetaKit5D, employing the MIP masks to skip empty regions. The deskewing / rotation results were output with a Nyquist voxel size of 127. 5 x 127.5 x 168.5 nm for the mouse OB datasets.

[0213] After deskewing / rotation, the intensity profiles of the sub-volumes were adjusted to correct uneven intensities resulting from photobleaching and antibody gradient effects both within and across sub-volumes. To adjust the intensity profiles within each sub-volume, the MIPs were used to estimate the intensity profiles along each axis. The xz MIP was initially smoothed using a Gaussian filter with a sigma of 2. Subsequently, the median value of all nonzero elements was calculated. The median values along the z-axis were calculated and normalized by dividing them by the overall median value. These normalized median values were then used for linear fitting. Correction factors along the z-axis were derived from this linear fitting model and capped to user-defined values (i.e. , [0.5, 2] or [0.4, 2.5]) to avoid excessively extreme corrections. The correction factors along the x-axis were calculated using the median values along the x-axis. Similarly, the correction factors along the y-axis were determined using the yz MIP. Furthermore, an intensity correction factor for each subvolume was calculated by dividing a constant value (i.e., 5000) by the median intensity of the xy MIP. Intensity correction was applied using both axis-specific correction factors and an overall correction factor across sub-volumes. These factors were multiplied and then limited to user-defined ranges (e.g., [0.4, 2.5] or [1Zs, 3]). The intensity correction was performed using batch processing with a batch size of [1024, 1024, 1024], applying the correction factors to the corresponding batches.

[0214] After intensity correction, the scales on the x and y axes of the sub-volumes were adjusted to account for the sample’s varying expansion during acquisition or gel degradation. The xz and yz MIPs were used to determine the optimal resampling factors betweenadjacent sub-volumes. Ultimately, these pairwise resampling factors were unified across all sub-volumes. The resampling tool in PetaKit5D was used to rescale the sub-volumes with the unified resampling factors.

[0215] After resizing each sub-volume along the x and y axes, the next step was to stitch them together along the z direction. Initially, sub-volume coordinates were estimated using the overlapped volumes identified from the MIPs. Rigid 3D stitching was performed using the large-scale strategy described in PetaKit5D with cross-correlation registration and feather blending. Here, bounding boxes to define dominant regions were applied during feather blending to assign higher weights to the overlapped volumes from the top sub-volumes (which were imaged earlier) to better preserve the image quality, as the overlapped volumes from the bottom sub-volumes were subject to an increased photobleaching effect. To stitch the WT and MT mouse OB datasets in particular, the registration information from the fullresolution and 5 x 5 x 5 downsampled data were used, respectively.

[0216] The human hippocampus samples were processed similarly to the mouse OB datasets, excluding the intensity correction and xy resampling. Steps included flat-field correction, stitching of each sub-volume, deconvolution, deskewing / rotation, stitching across the sub-volumes, and exported at a voxel size of 98.0 x 98.0 x 90.0 nm. All three channels were used for stitching within each sub-volume, but only the axon (NF-200) and myelin sheath (MBP) channels were used for stitching across the sub-volumes.

[0217] The stitching of sequential two-photon photochemical sectioning was performed in a similar manner. The tile coordinates were first estimated by manually inspecting intensity profiles across adjacent tiles and saved in an image list CSV file. Then the tiles were stitched together using the stitcher in PetaKit5D with the image list CSV file.Filtering non-specific antibody signals

[0218] This step was performed by removing small objects using variable background segmentation, via two steps: estimating local background and segmentation. The local background within a volume of 128 x 128 x 128 voxels in size was estimated using batch processing with this batch size and an additional 64 voxel size border buffer on both sides. For each batch, the image was first smoothed with a 3D Gaussian filter with a user-defined sigma value (4 for the datasets). After smoothing the image, the xy MIP was computed. The candidate threshold value (T) was calculated based on the non-zero elements of the smoothed 3D data that fell below the 99.9th percentile in the xy MIP. Next, the xy MIP was used to determine another candidate threshold value (T_mask) by masking objects in the MIP. The mask for objects was generated by thresholding the background image, computedusing a Gaussian filter with sigma 80 applied to the MIP. When the masked area was found to cover more than 90% of the image, the multiplier applied to the background image was increased. Next, the background mask was generated by combining the object mask’s complement after a 5-pixel erosion, with the MIP thresholder using user-defined minimum and maximum thresholds. This was followed by image opening and an additional 5-pixel erosion. After obtaining the background mask, the threshold T_mask was determined using the mode of voxels within the mask in the MIP. When this threshold was lower than the 25th percentile, the median value was used instead. T_mask was then capped to [T_base, T_max], The background value for the batch was defined as 90% of the mask threshold T_mask, plus 10% of the maximum value between the Otsu-based threshold (T) and T_max. After processing all batches, background values were collected at their corresponding locations within the 128 x 128 x 128 downsampled data.

[0219] In the second step, the processing was split into small batches of 1024 x 1024 x 1024 voxels, each with an additional border buffer of 100 voxels on both sides of all axes. For each batch, the local background region was loaded and up-sampled to match the target batch region using linear interpolation. Then the batch region was Gaussian smoothed with a sigma of 4 and binarized using the up-sampled background multiplied with a user-defined factor (i.e. 1.8). Objects in the binarized mask smaller than a user-defined volume threshold (e.g., 5000 for myelin and 2000 for axon channels) were discarded. The cleaned mask was applied to the raw region to eliminate small objects and backgrounds. After cropping out the border buffers, the processed data was saved. The entire process was parallelized using the generic computing framework in PetaKit5D.Dataset export

[0220] To balance fluorescence intensities across the datasets, adjustments were made using manually annotated masks defining five anatomical layers of mouse olfactory bulbs [generated with ITK-SNAP (ver. 4.0.1)]: olfactory never layer (ONL, including the surrounding tissue), glomerular layer (GL), external plexiform layer (EPL) and mitral cell layer (MCL), internal plexiform layer (I PL), and granule cell layer (GCL). For each layer, the upper bound for normalization was determined by the 99.99th percentile of non-zero voxel intensities in the 20 x 20 x 20 downsampled data, with 0 as the lower bound. The data at full resolution was normalized using these upper and lower bounds, then rescaled to a range of 0 to 65535 for each layer. These normalization processes were performed using batch processing with the generic computing framework in PetaKit5D. The data was exported tor visualization with IMARIS software. The 5 x 5 x 5 downsampled data were normalized asdescribed above and converted to an IMARIS file using the IMARIS File converter in PetaKit5D.Visualization

[0221] Unless otherwise noted, all the 3D-rendered datasets (composite 3D / volumetric images) were flat-field corrected, deconvolved, stitched, filtered to remove non-specific antibody signals, and gamma adjusted for visualization as described herein. The mouse OB datasets were visualized and 3D-rendered at 5 x 5 x 5 downsampled resolution using IMARIS (ver. 10.0.0, Oxford Instruments). The human hippocampus dataset was visualized and 3D-rendered at full resolution using IMARIS (ver. 9.3.1, Oxford Instruments). The 10 traced axons were displayed using the “Cone” style with a scale of 1, and the starting points and terminal points were indicated by blue and green balls, respectively. Additional image processing and visualization parameters are shown in Table 2:Table 2.

[0222] Unless otherwise noted, all the 2D cross-sectional and maximum intensity projection (MIP) views of the imaged volumes were generated using Imaged distribution Fiji (ver. 1.54f).Segmentation and skeletonization of axons and myelin sheaths

[0223] To segment the axons and myelin sheaths, a procedure akin to filtering nonspecific antibody signals was applied to the final stitched data (before removing the nonspecific signals). Specifically, a threshold multiplier of 2.0 was used for local background and Gaussian smoothing with a sigma of 2.5 was applied.

[0224] After segmenting the data, skeletonization was performed using batch processing with a size of 1024 x 1024 x 1024 voxels and borders of 100 x 100 x 100 voxels on the segmentation using the MATLAB function bwskel applied to each batch. A minimum branch threshold of 50 voxels was selected to eliminate small branches. Following this, the skeletonized blobs were pruned by removing clusters with 5 or more neighboring voxels, along with their associated neighboring voxels.Axon and myelination analysis

[0225] The axon skeletonization and myelin mask were used to assess myelination of axons. An axon voxel was considered myelinated if it intersected with the myelin mask or fell within a user-defined threshold (e.g., 1 pm) of the nearest myelin voxel. Batch processing with a batch size of 1024 x 1024 x 1024 voxels was used for the analysis. For each batch, the images were first up-sampled to isotropic voxel size (i.e. , 127.5 x 127.5 x 127.5 nm) via nearest neighbor interpolation. Next, myelinated voxels were identified, and total and myelinated axon lengths were computed for each layer. After processing all batches, the total and myelinated axon lengths were summarized over the results of all batches for each layer. The myelination ratios were calculated by dividing the myelinated axon lengths by the corresponding total axon lengths.

[0226] To determine the axonized myelin, that is, the myelin with axons inside, a similar process was conducted using the myelin skeletonization and axon mask.

[0227] As described, ONL were excluded from the OB analysis and quantification because this layer contained additional tissue from the OB surface or the surrounding structures and had elevated level of immunostaining background. Hence, the masked region did not provide an accurate representation of neuronal structures within the layer.Multi-region of interest (ROI) analysis

[0228] 50 cropped ROIs of 1000 x 1000 x 1000 voxels (~66 x 66 x 87 pm at preexpansion scale) from each layer were generated from an evenly spaced grid for the centers across the entire volume of the dataset. The number of grid points in each dimension was approximately proportional to the dimensions of the volume. Briefly, an initial set of grid point numbers was used to generate the grid coordinates for the centers throughout the entirevolume, with the largest possible grid distances to cover the entire volume and the grid centered. Then, each ROI centered at a grid coordinate was checked to see if it was fully contained within the layer, as defined by the manually annotated masks. For the I PL layer, the constraint was relaxed to include an ROI with at least 75% of the volume within the layer, as this layer was thin. The valid ROIs following the above criteria were counted. If their number was between 50 and 55, the process stopped and the center coordinates for the valid ones were used to crop the ROIs. Otherwise, the grid was adjusted by changing the number of points to get the valid regions within the 50 to 55 range. This process was repeated for all the layers.

[0229] After identifying the centers of valid ROIs for all layers, the ROIs were cropped, and the XY MIPs of each ROI were generated for visualization at full resolution. The same method used for the analysis of the whole dataset was applied to the analysis of axons and myelin sheaths for each ROI.Tracing

[0230] Semi-automatic tracing of axons was performed using IMARIS Filament Tracer (ver. 9.3.1, Oxford Instruments) with the “AutoPath” method.

[0231] Additional image processing methods can be applied to images of individual layers or a reconstructed volume to, for example, selectively visualize a single imaging channel, or a specific biological structure or cell type. Some of these processes include: (1) applying Gaussian filters to smooth potential sharp intensity transitions introduced by stitching or deconvolution procedures, (2) automatic counting and mapping of features of interest using batch processing to identify voxel pairs useful in tracking features that span different layers, (3) resolution analysis and comparison vs. conventional imaging and sectioning methods, (4) expansion isotropy and image deformation analysis by comparing pre- and post-expansion (or post-photo slicing) images and calculating the root mean square error (r.m.s) for all the point-to-point measurements, and (5) photobleaching analysis.Example 1 : Synthesis of a photocleavable cross-linker

[0232] To enhance the usability and accessibility of VIPS, the PC was designed according to the following principles: (i) a photocleavage wavelength distinct from popular fluorophores; (ii) bio-orthogonal and water-soluble; (iii) low molecular weight for enhanced sample permeability; and (iv) synthesizable in a minimal number of steps with common wetlab equipment.cheme 2).Scheme 2

[0234] An ethyl formate (40 mL) solution of polyethylene glycol bis(2-aminoethyl) ether (1, Mn -2000 g / mol, 1 g, -0.5 mmol, Zhengzhou Alfa Chemical Co., Ltd) was stirred at reflux (80°C) for 48 hours, and dried under reduced pressure using a rotary evaporator (N-1300, EYELA) (with a water bath temperature set at 42°C, here and after, unless otherwise noted). The product was then dissolved in anhydrous tetra hydrofuran (THF, 40 mL) before triethylamine (EtsN, 2.8 mL) was added. Next, the solution was chilled and stirred at -70°C, and a THF (2 mL) solution of phosphoryl chloride (POCI3, 0.4 mL, 4.3 mmol) was added to the reaction mixture dropwise over a period of 20 min. The solution was stirred overnight in an ice bath, poured into a saturated sodium carbonate (Themo Fisher) aqueous solution (150 mL, chilled at 4°C), and further stirred at < 20°C for one hour. The THF layer was then evaporated under reduced pressure and the remaining aqueous layer was extracted five times with 200 mL of dichloromethane (DCM, Themo Fisher) each time. Finally, the combined DCM layer was dried under reduced pressure before ethyl acetate (EA, 10 mL) was added to dissolve the crude product. An excess of petroleum ether (Thermo Fisher) or hexane (Themo Fisher) was added to the EA solution, and the mixture was placed in an ice bath for 30 min. The precipitates were filtered, washed with cold petroleum ether or hexane, and dried under vacuum overnight to yield the solid product (2).

[0235] b) Synthesis of the photocleavable cross-linker (PC, 3) (Scheme 3)acrylic acid 2-nitrobenzaldehyde

[0236] The photocleavable crosslinker (PC, 3) was synthesized using a Passerini reaction. First, compound 2 from the previous synthesis step (Mn -2000 g / mol, 0.95 g,-0.475 mmol), 2-nitrobenzaldehyde (250 mg, 1.50 mmol), and acrylic acid (120 mg, 1.67 mmol) were dissolved in THF (20 mL). The reaction mixture was then stirred for 24 hours at room temperature in the dark. Next, THF was evaporated from the reaction mixture under reduced pressure, and the raw product was dissolved in EA (10 mL) for reprecipitation. Briefly, an excess of petroleum ether or hexane was added to the EA solution, and the mixture was placed in an ice bath for 30 min. The precipitates were filtered, washed with cold petroleum ether or hexane, and dried under vacuum overnight. Finally, the crude product was dissolved in double-distilled water to a concentration of -25 mg / mL, dialyzed (Spectra / Por 6 Dialysis Membranes 1kD, Repligen) for 12 hours, dried under reduced pressure, and reprecipitated once again using EA and petroleum ether / hexane to yield the final product (PC, 3).1H NMR (500 MHz, Chloroform-d, 5): 6.74 (s, 2H; (CO)-CH- (Ar)O-), 6.90 (s, 2H; -NH-), 6.51 (d, 2H; CH2= CH-CO), 6.23 (m, 2H; CO-CH = CH'H), 5.97 (m, 2H; CO-CH = CH'H).Example 2: Effect of PC concentration in swelling and degradation of photodegradable hydrogels (PC-gel)

[0237] The molecule of Example 1, which had a molecular weight of about 2 kDa, a polyethylene glycol (PEG) backbone and two photocleavable o-nitrobenzyl moieties was used to prepare polyacrylamide / sodium polyacrylate photodegradable hydrogels via free- radical chain-growth polymerization. The resulting hydrogels (PC-gels) were optically transparent and mechanically elastic.

[0238] Non-photodegradable gels were prepared by cross-linking polyacrylamide / sodium polyacrylate with bis-acrylamide (Bis-gel). Bis-gel was prepared with the standard bisacrylamide concentration (1.5 g / L) to have similar swelling properties as those for the PC- gel.

[0239] The PC-gels and the Bis-gels underwent osmotic swelling by immersing them in purified water three times for 20 min each time. Consistent with the Bis-gel expansion factor, the PC-gel expansion increased as the PC concentration decreased (Fig. 3A). A 4 - 5-fold expanded sample was found to be the most convenient to handle, as the PC-gel maintained mechanical integrity at these expansion levels. Unless otherwise noted, an expansion factor between 4 - 5 was used in the preparation of biological samples.

[0240] Global expansion isotropy of the PC-gel was measured, a root-mean-square (r.m.s) error of -1 -5% was found, which is comparable to Bis-gel.

[0241] Without intending to be bound by theory, it is believed that a PC cross-linker does not change transparency, mechanical, or swelling properties of polyacrylamide / sodium polyacrylate gels vs. an analogous hydrogel cross-linked with bis-acrylamide. Further, the PC cross-linker amount can be selected to control the resulting expansion factor in the expanded crosslinked hydrogel.

[0242] The influence of PC concentration in photodegradation time of PC-gels was studied by preparing and photodegrading a series of PC-gels. HEK 293 cells were embedded in PC-gels with different PC concentrations [1.992% (w / v), 2.490% (w / v), 3.735% (w / v), 4.358% (w / v), 4.980% (w / v), and 6.225% (w / v)] and subject to the same 405 nm illumination. The degradation times were recorded and shown in Fig. 3B.

[0243] Without intending to be bound by theory, it is believed that the degradation time of PC-gels increases substantially linearly with the concentration of photodegradable crosslinker.Example 3: Spatially precise photodegradation of PC-gels

[0244] Blank PC-gels (i.e. , gels not containing any tissue or cells samples within) were prepared and expanded in purified water using the basic gelation protocol described herein. The expanded gels were then transferred to a UV chamber where they were exposed to a continuous illumination of a 365 nm light (365 nm, 60 W, PC-60-DJ, Phrozen Tech Co. LTD), the bulk PC-gel underwent complete photodegradation, transforming into a disintegrated liquid state. Fig. 2A schematically shows the effect of UV light on the photodegradation of the cross linker and consequently on the PC-gel, degrading an initially crosslinked hydrogel 101 into photodegraded hydrogel 102 with cleaved PC residues 104C and 104D, and cleaved, non-crosslinked hydrogel polymer chains 103C and 103D.

[0245] PC photocleavage occurs most efficiently around 270-350 nm, however PC also absorbs at larger wavelengths, for example 405 nm. Since biological specimens are commonly imaged by confocal or multiphoton microscopy using an illumination light with a wavelength of 405 nm, a test was performed at this wavelength to verify that photodegradation can be performed in a spatially precise manner at commonly used wavelengths. A piece of PC-gel was illuminated from the side with a thin sheet of 405 nm laser. Fig. 4 shows the PC-gel before, during, and after illumination at 405 nm, it can be observed that the UV light beam effectively and precisely cut the PC-gel.Example 4: Volumetric fluorescence imaging via confocal illumination (ConfocalVIPS)

[0246] Once blank PC-gels were successfully and accurately sectioned using a 405 nm light beam tests were run using embedded cells using a confocal microscope for imaging and photo-sectioning. Three sets of HEK cells were embedded in PC and Bis -gels for single-photon widefield photodegradation testing across a defined region-of-interest (ROI) using a confocal microscope. HEK 293 cells were fluorescently labeled for p-tubulin according to the method described herein. The cells were embedded in PC-gels using the basic protocol described herein modified as follows: the cell samples were directly incubated in a modified gelling solution [1x PBS, 2 M NaCI, 8.625% (w / v) sodium acrylate, 2.5% (w / v) acrylamide, 6.225% (w / v) PC, 0.2% (w / v) APS, 0.2% (w / v) TEMED], and immediately transferred to the gelation chamber in a humidified 37°C incubator for gelation. For the gelling solution, 4HT was replaced with purified water to ensure swift gelation. In addition, the gelation was performed for 1 hour instead of 2 hours. The gelled and digested cell samples were incubated in a SiR-DNA (CY-SC007, Cytoskeleton, Inc) solution (1:300 dilution with 1x PBS) for 1 hour prior to expansion and imaging.

[0247] HEK 293 cell samples embedded in photodegradable and Bis gels were expanded in purified water and immobilized onto a 6-well glass bottom well plate using a poly-L-lysine mounting method described in the art; the corresponding expansion factors are specified for each gel in Table 3. The expanded samples were first imaged using the 488 nm, 561 nm, and 640 nm channels with a Z-step of 0.4-1 pm across a 2-by-2 or 3-by-3 tiled field-of-view (FOV, 15% overlap) on a confocal microscope (Yokogawa CSLI-X1 , Nikon). For photodegradation, the 405 nm channel was used to illuminate the central FOV for ~6-8 min at 100 % power (5.17 x 10'5mW / pm2). After illumination, the samples were imaged again using the 488 nm, 561 nm, and 640 nm channels with a Z-step of 0.4-1 pm across the same 2-by-2 or 3-by-3 tiled FOV. The non-photodegradable controls were expanded and subject to the same 405 nm illumination for >20 min and imaged across a 3- by-3 tiled FOV before and after the illumination. A summary of the imaging and photodegradation times are shown in Table 3.Table 3.

[0248] Fig. 6A includes confocal microscope images of the labeled HEK cells embedded and expanded in PC-gel. Fig. 6A shows the imaged cells before (left) and after (right) photodegradation with a dotted line indicating the limit of the region exposed to the UV laser (ROI). In the pre-exposure image (left), the fluorescent signal highlights areas of the cell where p-Tubulin is present. An image of the same region after exposure (right) shows no signal in the ROI (i.e. , the area of the sample and PC-gel exposed to the UV laser). In contrast, the cells embedded in Bis-gel (Fig. 6B) show the fluorescently labeled p-tubulin regions in both, pre and post, exposure conditions indicating that the signal loss within the photodegraded ROI resulted from the diffusion / detachment of freed, originally gel-anchored fluorophores, rather than from photobleaching.Example 5: Volumetric fluorescence imaging via two-photon photochemical sectioning (2P VIPS)

[0249] In order to spatially confine PC-gel photodegradation, a sequential volumetric fluorescence imaging and two photon (2P) photochemical sectioning method was tested (2P VIPS). In a 2P VIPS method 200 illustrated in Fig. 5, the hydrogel composite 402 sample is first volumetrically imaged in the image layer 203, and then raster-scanned with a 2P- focused spot to photodegrade the top portion or photodegradation layer 202 of the sample before iterating this process to reach an imaging depth beyond the objective working distance (Fig. 5). During this process, the degraded volume or photodegradation layer 202 was smaller than the imaged volume or image layer 203 to maintain sufficient overlaps between the imaged sub-volumes (Fig. 8A). In these experiments, an upright Leica SP8 DIVE multi-photon microscope with a HC IRAPO L 25x (1.00 NA) W motCORR waterdipping objective (11507704, Leica; WD: 2.6 mm) was used to image and photodegrade the samples. For single-photon confocal imaging, green (488 nm), red (552 nm), and far-red (638 nm) channels were used to excite the fluorophores, and the Hybrid Detector (HyD) was used to detect the fluorescence signals. For two-photon photodegradation, an adjustable Spectra Physics Mai Tai laser (690-1040 nm) light source 201 was used to illuminate the sample at a wavelength of 740 nm, the photodegradation power for each of two samples imaged in this example are specified in Table 4.

[0250] Two 40 pm mouse brain slices were gel embedded using the basic protocol. The photodegradable mouse brain slices corresponding to a C57BL / 6 genotype were expanded by an expansion factor or 5.4 in purified water and immobilized on a no.1.5 coverslip glued to a 100 mm petri dish using the poly-L-lysine mounting method. The petri-dish containing the samples was then mounted on a multi-photon microscope (SP8 DIVE, Leica).

[0251] To test the spatial control of 2P photochemical sectioning, a 740 nm 2P laser was used to photodegrade a 3D ROI of the first PC-gel-expanded mouse tissue sample densely labeled for synaptic proteins prepared according to methods described herein.

[0252] The first sample was imaged using the microscope’s single-photon confocal mode with green (488 nm), red (552 nm), and / or far-red (638 nm) channels at a Z-step size of 2 pm and a zoom factor of 1. Next, the middle volume (~60 pm thick) of the sample was photodegraded using the multi-photon mode at a wavelength of 740 nm at a Z-step size of 1 pm, a laser intensity of 30%, and a zoom factor of 2 for ~11 min. The sample was imaged again in the single-photon confocal mode using the same imaging conditions. After photodegradation, the illuminated region showed both lateral and axial confinements that matched well with the defined 3D ROI. Photodegradation and imaging times and voxel sizes are listed in Table 4.

[0253] After this first validation, the 2P VIPS sequential two-photon photochemical sectioning was applied to a second PC-gel-expanded mouse brain slice to overcome the working distance limit of a commercially available objective. After incubation, the mouse brain slice was gel embedded using a modified basic protocol as follows: for gelation chamber spacer, glass slides of -1.0-1.2 mm in thickness were used instead of the coverslips. Additionally, after expanding the sample in purified water, the gel block was trimmed into a cube with a side length of ~4-5 mm. The gel block was vertically rotated and immobilized on a no. 1.5 coverslip glued to a 100 mm petri dish using the poly-L-lysine mounting method. Single-photon confocal imaging and two-photon photodegradation were then performed iteratively. Briefly, part of the top ~1.5 mm of the gel block containing the tissue was imaged in the single-photon confocal mode with XY tiling (two FOVs, 10% overlap) at a Z-step size of 5 pm and a zoom factor of 0.75. Next, the top -1.0 mm of the gel block was photodegraded in the multi-photon mode at a wavelength of 740 nm with XY tiling (42 FOVs, 10% overlap), a Z-step size of 5 pm, a laser intensity of 100%, and a zoom factor of 0.75. After the two-photon photodegradation, no fluorescence signals were observed in the top -1.0 mm of the gel block. Next, the plane lying -1.0 mm below the initial gel surface was defined as the new surface of the gel, and another -1.5 mm volume of the sample was imaged in the single-photon confocal mode with XY tiling (two FOVs, 10% overlap), a Z-step size of 5 pm, and a zoom factor of 0.75. The same process of two-photon photodegradation, redefinition of the gel surface, and single-photon confocal imaging (with the last imaging step covering a -1.8 mm thick volume) was repeated once more until the total imaging depth from the initial gel surface reached -3.8 mm. Three rounds of on-block volumetric confocal imaging were performed in total with depths of -1.5 mm, -1.5 mm, and -1.8 mm, respectively, with each imaging round, except for the last, followed by a -1.0 mm 2P photodegradation step across the entire sample block-face. As a result, the three stitched sub-volumes spanned a -3.8 mm thick volume from the original sample surface, far beyond the objective’s working distance of -2.6 mm. Photodegradation and imaging times and voxel sizes are listed in Table 4.Table 4.Example 6: Volumetric fluorescence imaging via light-sheet photochemical sectioning (light-sheet VIPS)

[0254] Light-sheet photo sectioning was tested to increase speed and reduce susceptibility to photobleaching in large-volume imaging. The test aimed at extending the imaging capacity to thick samples, beyond the microscope’s working distance.

[0255] To achieve spatially precise photodegradation parallel to the sample surface, a separate UV light-sheet illumination path (Fig. 7) as described herein was employed. This implementation generated a thin UV light-sheet compatible with both single and dual objective configurations since it bypasses the objectives and arrives parallel to the sample surface. Photodegradation using a 100 mW, 405 nm laser revealed that a complete photodegradation of the PC-gel-embedded samples could be completed in ~20 minutes for ~4-5-fold expansion in water and in ~2 hours for ~2-fold expansion in 1x PBS.

[0256] This light-sheet VIPS configuration was applied to a ~4.4-fold expanded human hippocampus tissue sample labeled for axons and myelin sheaths as described herein using the antibodies described in Table 5. The labeled human hippocampus slices were gelembedded using a modified basic protocol. Briefly, the stained brain slices were incubated in the AcX solution (0.1 mg / mL in 1x PBS) for 2 days, which was replaced with a freshly prepared AcX solution every 12 hours. The brain slices were then incubated in a modified monomer solution [1x PBS, 2 M NaCI, 8.625% (w / v) sodium acrylate, 2.5% (w / v) acrylamide, 7.470% (w / v) PC] at 4°C for 1 day and in the same gelling solution at 0°C for 60 min before gelled in a humidified 37°C incubator for 2 hours. For the gelation chamber spacer, a stack of three no. 1.5 glass coverslips were used. After gelation, the samples were digested with proK for 2-3 days in the digestion buffer, which was replaced with a freshly prepared digestion buffer every day. Finally, the digested samples were incubated in 4% (w / v) SDS in 1x PBS at 37°C for 6-7 hours and washed with 1x PBS five times over two days. The samples were stored in 1x PBS at 4°C previous to imaging.Table 5.

[0257] To image the entire sample, eight rounds of sequential volumetric lattice lightsheet imaging were performed followed by light-sheet photochemical sectioning across a ~1.6 mm sample thickness at post-expansion scale — five times the imaging depth achievable without VIPS (Fig. 8B). The total imaging time was about 7 hr. with a total of 2hr. photodegradation time. The imaged sub-volumes were then registered and stitched using PetaKit5D with a pre-expansion Nyquist-sampled voxel size of 22 by 22 by 59 nm with a post-expansion voxel size of 98 by 98 by 260 nm. The final dataset corresponded to an imaged volume of 57 by 57 by 370 pm (250 x 230 x 1630 pm post-expansion) and allowed the tracing and reconstruction of individual axons with diameters as small as -150 nm withinthe imaged volume, confirming that the pipeline preserved the connectivity and traceability of anatomical features across multiple rounds of imaging and photochemical sectioning. These results demonstrated light-sheet VIPS scalability for high-throughput, super-resolution imaging of whole-mount expanded samples.Example 7: Light-sheet VIPS of the Olfactory Bulb (OB) of a wild-type mouse

[0258] To showcase the scalability of light-sheet VIPS, a two-color imaging of an entire 7-week wild-type mouse (genotype BLAB / c(N pc 7+ / +)) OB was performed. These type of studies of the anatomical and molecular architecture of the OB can provide important insights into the anatomy and function of the mammalian sensory system.

[0259] For this experiment, a ~2-fold expansion with a pre-expansion Nyquist-sampled voxel size of 56 by 56 by 135 nm was selected, as this resolution is sufficient to trace and reconstruct most axons in the mouse brain. The post-expansion voxel size was 108 by 108 by 260 nm. Higher expansion factors would increase the data volume to tens of petabytes for the raw, intermediate, and fully processed volumes, complicating data handling and requiring additional computation capabilities.

[0260] The OB tissue was stained according to methods described herein using the antibodies and immunostaining targets indicated in Table 6, then gel embedded using a modified basic protocol as follows: after immunostaining, the OBs were incubated in the AcX solution (0.1 mg / mL in 1x PBS) for 2 days, which was replaced with a freshly prepared AcX solution every 12 hours. The OBs were then incubated in the monomer solution (as in the basic protocol) at 4°C for 1-2 days and in the gelling solution (as in the basic protocol) at 0°C for 60 min before gelled in a humidified 37°C incubator for 2 hours. For the gelation chamber spacer, double-stacked glass slides (-2.0-2.4 mm in thickness) were used. After gelation, the samples were digested with proK for 2-3 days in the digestion buffer, which was replaced with a freshly prepared digestion buffer every day. The digested samples were stored in 1x PBS at 4°C until imaging and photodegradation.

[0261] After mounting the expanded OB, the entire sample volume (~ 3.3 by 4.8 by 2.2 mm pre-expansion I -9.4 by 6.4 by 4.3 mm at post-expansion scale) was imaged with the peak acquisition rates approaching 2 teravoxels per hour. Imaging was performed using the 488 nm and 560 nm illumination channels; and photodegradation using a 405 nm wavelength at 100 mW power. See Table 6 for imaging exposure times and power.

[0262] The OB imaging was completed in 10 days, utilizing nine rounds of interleaved on-block volumetric lattice light-sheet imaging and light-sheet photochemical sectioning,totaling -142 hours of imaging and -20 hours of photodegradation. This process yielded 17,280 image tiles, constituting -0.5 petabytes of raw data, which were computationally combined to reconstruct the entire mouse OB. To process this petabyte-scale image dataset with academic-scale computing resources pipelines were built as described herein, using PetaKit5D, a high-performance computing framework designed for efficient image reading, writing, geometric transformations, stitching, and deconvolution. This pipeline was used to deskew, rotate, deconvolve, and stitch the entire imaged volume with drastically reduced computation time. Advantageously, the gentle and spatially precise nature of photochemical sectioning was crucial for achieving lossless overlap when stitching together the imaged sub-volumes. The total volume imaged was about 3320 by 4870 by 2220 pm (6410 by 9401 by 4290 pm post-expansion).Table 6.

[0263] The reconstructed wild-type mouse OB dataset revealed distinct axonal projections and myelination patterns across all its major anatomical layers. Detailed 3D visualization highlighted unique axonal and myelination patterns within individual glomeruli of the glomerular layer (GL). The vertical axonal projections of tufted cells traversed through the external plexiform layer (EPL) and mitral cell layer (MCL), while the dense lateral axonal projections spanned the internal plexiform layer (I PL). The axonal bundles from the tufted and mitral cells were observed passing through the granule cell layer (GCL), exiting the olfactory bulb, and projecting through the lateral olfactory tract to the olfactory cortex. Furthermore, the nanoscale resolution of the dataset enabled segmentation, skeletonization, and reconstruction of individual myelinated and unmyelinated axon fibers, allowing for precise measurements such as calculating their total lengths throughout the entirety of mouse OB.Example 8: Light-sheet VIPS of the Olfactory Bulb (OB) of a NPC1-type mouse

[0264] Light-sheet VIPS was used to study axon degeneration and de- / dysmyelination, believed to be hallmarks of neurodegeneration and aging, in an NPC1 brain. In the neurodegenerative lysosomal storage disorder Niemann-Pick type C1 (NPC1) disease, mutation in the NPC1 genes cause early-stage dysmyelination. Anatomical and physiological studies have identified pathophysiological changes in the NPC1 animal’s olfactory system as an early indicator of disease progression, which is also characterized by the reduced myelin basic protein (MBP) expression in the OB at the neonate and adolescent stage. However, the precise mechanism behind de- / dysmyelination remains unclear, particularly regarding whether the reduced expression of MBP in the OB persists into adulthood and results in anatomical loss at the single axon / myelin sheath level.

[0265] For this study, the OB of a 7-week NPC1 (Npc1~ / ~') mouse OB (~2-fold expanded and extending a volume of ~11.2 by 6.0 by 3.6 mm) was prepared and imaged according to methods described herein using the antibodies and targets described in Table 7. A voxel size 56 by 56 by 135 nm (108 by 108 by 260 nm post-expansion) was used throughout. Imaging was performed using the 488 nm and 560 nm channels using the exposure times and powers indicated in Table 7, and photodegradation performed using a 405 nm laser wavelength at a 100 mW power.Table 7.

[0266] The resulting multi-scale OB datasets from this and the previous experiments, allowed visualizing the axon and myelin sheath morphologies in the corresponding regions across the entire OB. Close comparison of the wild-type (WT) and NPC1-mouse datasets revealed distinct anatomical changes, including decreased axon and myelin sheath densitiesin the GL, EPL / MCL, and IPL of the NPC1 OB, suggesting potential axon degeneration and de- / dysmyelination in these layers.

[0267] OB-wide analysis of axon fibers showed reduced axon density in the NPC1 OB, consistent with previous studies showing early-onset axonal degeneration across all major parts of the NPC1 brain. Unexpectedly, the GCL exhibited a higher axon density for the NPC1 sample, while all other layers showed consistent decreases. Moreover, the severity of axon degeneration varied spatially across the GL, EPL / MCL, and IPL, with substantial reduction in the GL and the least in the IPL. These results suggest that local activity, connectivity, and the extracellular environment within each OB layer are more likely to be responsible for axonal integrity and survival, given that axonal damage often precedes cell body loss.

[0268] Additionally, the myelination profile of both the WT and NPC1 mouse OBs was compared. A reduction in the density of total myelin sheaths and myelinated axons was evident throughout the NPC1 OB affirming that early-stage dysmyelination persists as the animal maturates. Interestingly, the ratio of myelinated axons to all axons (the myelination ratio) varied across the layers, with decreases in IPL and GCL, but increases in GL and EPL / MCL in the NPC1 OB. This suggested that the effects of dysmyelination may differentially impact axon degeneration across different OB layers. Such spatial heterogeneity also exists locally within the same layer as the myelination ratio and other axonal / myelination parameters varied drastically across the spatially sampled ROIs within each layer.Example 9: Olfactory bulb-wide topography and fractography

[0269] Capitalizing on the nanoscale resolution and whole-volume continuity provided in the VIPS images of examples 7 and 8, a topography and fractography analysis was completed to evaluate axonal connectivity across the OB of the WT and NPC1 mice.

[0270] Axon fractography was generated from skeletonized axon volumes in batches (2048 x 2048 x 2048 voxels with a border buffer of 32 x 32 x 32 voxels). For each batch, the connected components were identified, those with fewer than 200 voxels were excluded. For each remaining component, the endpoints and branching points were detected, initiating tracking from both. At each branching point, tracks were merged by selecting the pair of branches with the most similar directions. For every point along a track, up to 50 voxels in both directions (fewer if near the border) were retrieved and applied Gaussian smoothing with a window size of 51. The resulting smoothed path was then resampled to 1,000 points using spline interpolation. Angular orientation at each point was computed using thegradients of the fitted curve, yielding two angles: the polar angle 0 and the azimuthal angle cp. Angles were mapped to the upper hemisphere (0 > 0). When 0 < 0, cp was adjusted by adding TT. Finally, azimuthal angles cp were wrapped to the interval [0, 2TT) to ensure continuity and positivity. To identify and exclude abnormal tracks, continuous segments longer than 40 voxels where 0 was either 0 or approximately 0.755 (within a tolerance of 0.01) were flagged. Tracks containing such segments were excluded from subsequent angle analysis.

[0271] The myelinated fractography was derived from the initial fractography, retaining only those voxels classified as myelinated, based on the same criteria outlined herein.

[0272] In the WT OB, the number of segmented axons traversing between adjacent layers remained constant throughout the bulb. In the NPC1 OB, however, progressively fewer segmented axons traveled from the IPL-GCL to the EPL / MCL-IPL and then to the GL- EPL / MCL interfaces, indicating that degeneration begins in the outer layers and advances inward. Single-axon fractography further enabled visualization and quantitative analysis of the 3D organization of the segmented axons across the entire OB, as well as within specific OB layers and localized regions. Within each layer, axon orientations systematically shifted from a ventral-dorsal axis in the outer layer to an anterior-posterior axis in the deeper layer (e.g., GCL), reflecting the shifting paths of information flow. Compared to the WT OB, the NPC1 OB showed a more prominent reduction in axons projecting along the ventral-dorsal axis within the I PL, suggesting that the degenerative process can be selective for axons with specific orientations. Within this overall architecture, however, there existed substantial local heterogeneity, with axons projecting in many directions, even if the average orientations in each region followed an overall trend. Finally, in the WT OB, the myelinated axon orientations mirrored those of the total axon population, while in the I PL and GCL of the NPC1 OB, the myelinated axons were strongly biased along the anterior-posterior axis. This suggests that the remaining axons in these layers, mostly myelinated, project primarily to or from the olfactory cortex via the lateral olfactory tract.Example 10: Comparison of photochemical vs physical sectioning

[0273] The photochemical sectioning step is a key element of the VIPS method. To compare its performance with traditional physical sectioning methods, the following were conducted: (a) optimization of mechanical sectioning conditions for expanded tissue using a standard vibratome, (b) examination the mechanical sectioning performance under the optimized best-practice sectioning conditions, and (c) evaluation of the advantages and disadvantages of photochemical sectioning versus mechanical sectioning.

[0274] For comparison with photochemical sectioning, all mechanical sectioning tests were performed using a standardized expansion microscopy protocol and widely accessible sectioning equipment, like a vibratome. Noteworthily, the performance of vibratomes varies with sectioning parameters such as the vibration amplitude, blade advance speed, and sample buffer. To obtain the best-practice sectioning conditions, sectioning tests were first performed at different conditions and their sectioning success rates were compared. Briefly, mouse brain chunks were expanded with the commonly used protein-retention expansion microscopy (proExM) protocol as it is known in the art, and subsequently sectioned with a standard vibratome (VT1200S, Leica). Each sectioning condition was repeated approximately 20 times with the sectioning thickness matching the nominal working distance of the lattice light-sheet microscope setup (-300 pm).

[0275] A “failed sectioning” was defined as a sectioning attempt with either no sectioning or substantial damage to both the sectioned gel and the block-face, clearly visible to the naked eye. This included: (i) macroscopic breaking or cracking of the sectioned gel which likely indicates damage to the block-face; (ii) scraping of the block-face with severe deformation and little to no sectioning; (iii) macroscopic damage to the block-face such as breaking, cracking, or severe roughness. All other outcomes were classified as “successful sectioning”. However, it was noted that while sectioning may appear successful at the macroscopic level, the true success rate is likely significantly lower due to microscopic damage.

[0276] Gel submergence: two submergence conditions for gel sectioning were tested: partial submergence, where a few millimeters of the gel remained above the liquid surface, and full submergence (Table 8). “No submergence” was an excluded condition, as it led to global gel deformation and drying. Interestingly, partial submergence consistently produced more successful trials than full submersion. This disparity is likely due to the water bath ripples generated by the vibrating blade and its arms during full submersion, leading to fragmentation of both the block-face and the sectioned slice. In addition, full submersion often reduces gel visibility, making sample tracking and collection difficult. Based on these findings, partial submergence was determined the best condition for gel sectioning and handling. All subsequent experiments were conducted under this condition unless otherwise noted.Table 8.

[0277] Vibration amplitude: For the Leica-VT1200S vibratome, the amplitude ranges from 0.0 to 3.0 mm. Five conditions were tested (Table 9) and found that a vibration amplitude of 1.1 mm provided the highest success rate. Based on these experiments, high amplitudes (e.g., 2.9 mm) often caused fragmentation of the gel block-face, while low amplitudes (e.g., 0.5 mm) led to inadequate sectioning (e.g., gel block-face getting scraped instead of sectioned).Table 9.

[0278] Blade advance speed: For the Leica- T1200S vibratome, the blade advance speed ranges from 0.01 to 1.5 mm / s. Five conditions were tested, and found that 0.6 mm / s provided the highest success rate, with speeds of 0.3 mm / s, 0.9 mm / s, and 1.2 mm / s also achieving high success rates (>70%) (Table 10). For sectioning, 0.3 mm / s was eliminated from the optimal condition because it took too long (> 1 min) to complete each section. Therefore, an advance speed of 0.6-1.2 mm / s was found to be the best condition for sectioning.Table 10.

[0279] Based on these results, the optimal sectioning parameters for expanded samples were determined as: “Amplitude = 1.1 mm, advance speed = 0.6-1.2 mm / s, with the gel partially submerged in water”. These conditions were then tested in 1x PBS, repeated -10 times per condition. For thinner gels (< ~2.5 mm), the macroscopic success rate remained at -100% with the optimal conditions similar to water. For thicker gels (> -2.5 mm), however, the optimal conditions required higher vibration amplitudes (2.3 mm), with a success rate of -70 %. This reduced success rate for thicker gel is likely due to the increased viscoelasticity and the reduced expansion factor of the gel in 1x PBS, which led to increased “sticking” of the blade when the cutting plane became more unsteady.

[0280] Next, both macroscopic and microscopic morphology of the mechanically sectioned samples was examined. As described, the macroscopic sectioning success rates hardly achieved 100% in water or 1x PBS despite the extended optimization process. Macroscopic damage was observed, including substantial sample distortion and tearing, for macroscopically failed sectioning attempts. For macroscopically successful sectioning, sample deformation and surface roughness in the order of hundreds of microns were further observed across the sample or part of the sample. Due to the limitation in the imaging setup and throughput, the sectioned surfaces of all the successfully sectioned samples could not be analyzed. However, for the samples checked, microscopic damage was commonly observed, which compromises the downstream nanoscale volumetric imaging and analysis.

[0281] Unlike mechanical sectioning, the performance of photochemical sectioning is independent of the sample’s mechanical properties or mounting configurations. Therefore, it remains effective on samples of any shape, angle, and aspect ratio, even those with challenging geometries. Imaging whole-mount tissue samples typically require long imaging time and repeated sectioning. Photochemical sectioning avoids physical contact with the sample, thereby substantially reducing risks of sample detachment during the sectioning process. Photochemical sectioning also circumvents issues potentially arising from mechanical mismatches between the tissue and surrounding gel matrix, thus preventing sample cracking and delamination at the tissue-matrix interface. Additionally, photochemical sectioning can be integrated with existing microscopy setups with minor (e.g., LLSM VIPS)or no (e.g., 2P VIPS) modifications. This enables continuous sectioning and imaging on the same microscopes with minimal sample transfer and disturbance, further reducing the risks of sample loss, breaking, and distortion.

[0282] Because other modifications and changes varied to fit particular operating requirements and environments will be apparent to those skilled in the art, the disclosure is not considered limited to the example chosen for purposes of illustration, and covers all changes and modifications which do not constitute departures from the true spirit and scope of this disclosure.

[0283] Accordingly, the foregoing description is given for clearness of understanding only, and no unnecessary limitations should be understood therefrom, as modifications within the scope of the disclosure may be apparent to those having ordinary skill in the art.

[0284] All patents, patent applications, government publications, government regulations, and literature references cited in this specification are hereby incorporated herein by reference in their entirety. In case of conflict, the present description, including definitions, will control.

[0285] Throughout the specification, where the compositions, processes, kits, or apparatus are described as including components, steps, or materials, it is contemplated that the compositions, processes, or apparatus can also comprise, consist essentially of, or consist of, any combination of the recited components or materials, unless described otherwise. Component concentrations can be expressed in terms of weight concentrations, unless specifically indicated otherwise. Combinations of components are contemplated to include homogeneous and / or heterogeneous mixtures, as would be understood by a person of ordinary skill in the art in view of the foregoing disclosure.LIST OF COMPONENTS101 Crosslinked hydrogel102 Photodegraded hydrogel103 Monomer / Polymer chain104 Photocleavable crosslinker (PC)200 Optical setup for two photon (2P) volume imaging201 Light source202 Depolymerization layer203 Image layer204 Photosectioning plane I photodegradation bottom boundary300 Lattice light sheet illumination setup301 Acousto-optic tunable filter302 Cylindrical lens (A, B for first, second)303 Spatial light modulator304 Powell lens305 Excitation Objective306 Detection objective307 Deformable mirror308 CMOS camera (A, B for first, second)309 Laser (A, B, C... for first, second, etc.)401 Sample material402 Hydrogel composite403 Sub-volume (A, B, C... for first, second, etc.)404 Sub-volume overlap405 Image sublayer406 Image sublayers overlapAqueous environment Kit Ethylene or vinyl functional hydrogel monomers, polymerization reagents, fluorescent stain reagents, light sources, gelation chambers, and brushes Method of volumetric imaging by photochemical sectioning Provide the hydrogel composite Optically image a top image layer of the hydrogel composite Photochemically section a top degradation layer Repeat steps 620 and 630 a plurality of times Stack and / or stitch the image layers to form a composite 3D / volumetric image of the sample material

Claims

CLAIMS1. A photocleavable crosslinker (PC) comprising: a water-soluble hydrocarbon linker (L) comprising a plurality of ethylene oxide repeat units; two ethylenically unsaturated groups (U) each comprising an acrylate group; and two photocleavable groups (P) each comprising an o-nitrobenzyl group; wherein the photocleavable crosslinker (PC) has a structure U-P-L-P-U.

2. The photocleavable crosslinker of claim 1 , wherein: the photocleavable crosslinker has a structure according to formula (I):n is in a range of 10 to 100.

3. A crosslinked hydrogel comprising: a crosslinked polymerization product between (i) the photocleavable crosslinker of claim 1 , (ii) an acrylamide monomer, and (iii) an acrylate monomer; wherein: the photocleavable crosslinker is present in the crosslinked hydrogel in an amount in a range of 5 to 60 wt.%; the acrylamide monomer is present in the crosslinked hydrogel in an amount in a range of 3 to 30 wt.%; and the acrylate monomer is present in the crosslinked hydrogel in an amount in a range of 10 to 80 wt.%.

4. A hydrogel composite comprising: the crosslinked hydrogel of claim 3 as a matrix; and a sample material immobilized in the matrix, the sample material comprising mammalian tissue and a fluorophore having at least one of condition (I) and condition (II): (I) an excitation wavelength at which absorption by the photocleavable crosslinker is not more than 1%, and (II) an excitation wavelength at least 50 nm different from a cleavage wavelength of the photocleavable crosslinker.

5. A method for volumetric optical imaging of a sample material, the method comprising: (a) providing the hydrogel composite of claim 4, wherein the hydrogel composite is inan aqueous environment;(b) optically imaging an image layer of the hydrogel composite to obtain a volumetric section image of any sample material in the image layer, wherein optically imaging the image layer comprises exposing the image layer to electromagnetic radiation at one or more excitation wavelengths in a range of 400 nm to 800 nm;(c) photochemically sectioning a depolymerization layer of the hydrogel composite, wherein photochemically sectioning the depolymerization layer comprises exposing the image layer to electromagnetic radiation at a photocleavage wavelength in a range of 100 nm to 450 nm;(d) repeating steps (b) and (c) a plurality of times for sequential image layers and depolymerization layers; and(e) stacking the sequential image layers to form a composite volumetric image comprising voxels having a length dimension in a range of 5 nm to 100 nm.

6. A photocleavable crosslinker (PC) comprising: one or more water-soluble hydrocarbon linkers (L); two or more ethylenically unsaturated groups (II) or other reactive functional groups (F) each bound directly or indirectly to one or more water-soluble hydrocarbon linkers (L); and one or more photocleavable groups (P) each bound directly or indirectly to at least one of the one or more water-soluble hydrocarbon linkers (L) or other reactive functional groups (F), and to at least one of the two or more ethylenically unsaturated groups (II).

7. The photocleavable crosslinker of claim 6, wherein the photocleavable crosslinker (PC) comprises two or more photocleavable groups (P).

8. The photocleavable crosslinker of claim 6, wherein the photocleavable crosslinker (PC) has a structure U-P-L-P-U.

9. The photocleavable crosslinker of claim 6, wherein: the photocleavable crosslinker (PC) has a structure C-[(L-P)n-U]2 where n = 1 , 2, 3, or 4; or the photocleavable crosslinker (PC) has a structure C-[L-P-U]nwhere n = 2, 3, or 4; andC is a hydrocarbon core.

10. The photocleavable crosslinker of claim 6, wherein the one or more water-soluble hydrocarbon linkers (L) comprise a plurality of ethylene oxide repeat units.

11. The photocleavable crosslinker of claim 6, wherein: photocleavable crosslinker (PC) comprises the two or more ethylenically unsaturated groups (II); and the two or more ethylenically unsaturated groups (II) each comprise a (meth)acrylate group.

12. The photocleavable crosslinker of claim 6, wherein: photocleavable crosslinker (PC) comprises the reactive functional groups (F); and the reactive functional groups (F) are each independently selected from the group consisting of alkyne groups, hydroxyl groups, amine groups, N-hydroxysuccinimide (NHS)- esters, azide groups, and maleimide groups.

13. The photocleavable crosslinker of claim 6, wherein the one or more photocleavable groups (P) each comprise an o-nitrobenzyl group.

14. The photocleavable crosslinker of claim 6, wherein the one or more photocleavable groups (P) each comprise a photolabile group selected from the group consisting of nitrobenzyl groups, phenacyl groups, benzyl groups, and combinations thereof.

15. The photocleavable crosslinker of claim 6, wherein the one or more photocleavable groups (P) is cleavable upon exposure to electromagnetic radiation having a wavelength in a range of 100 nm to 450 nm.

16. The photocleavable crosslinker of claim 6, wherein the photocleavable crosslinker and photocleaved residues thereof are water soluble.

17. The photocleavable crosslinker of claim 6, wherein the photocleavable crosslinker has a molecular weight in a range of 200 g / mol to 10000 g / mol.

18. The photocleavable crosslinker of claim 6, wherein: the photocleavable crosslinker has a structure according to formula (I):n is in a range of 10 to 1000.

19. A crosslinked hydrogel comprising: a crosslinked polymerization product between (i) the photocleavable crosslinker of claim 6 and (ii) one or more hydrogel monomers.

20. The crosslinked hydrogel of claim 19, wherein the photocleavable crosslinker is present in the crosslinked hydrogel in an amount in a range of 5 to 60 wt.%, 0.1 to 5 mol.%, or 0.2 to 10 eq.% relative to the crosslinked polymerization product.

21. The crosslinked hydrogel of claim 19, wherein the one or more ethylenically functional hydrogel monomers comprise acrylamide monomer and acrylate monomer.

22. The crosslinked hydrogel of claim 21 , wherein: the acrylamide monomer is present in the crosslinked hydrogel in an amount in a range of 3 to 30 wt.%, 5 to 50 mol.%, or 5 to 50 eq.% relative to the crosslinked polymerization product; and the acrylate monomer is present in the crosslinked hydrogel in an amount in a range of 10 to 80 wt.%, 30 to 90 mol.%, or 30 to 90 eq.% relative to the crosslinked polymerization product.

23. The crosslinked hydrogel of claim 19, wherein the one or more ethylenically functional hydrogel monomers comprise one or more of (i) N,N-dimethylacrylamide (DMAA) monomer and acrylate monomer, and (ii) thiolated polyethylene glycol (PEG-SH) monomer and polyethylene glycol-diacrylamide (PEG-diacrylamide) monomer.

24. A hydrogel composite comprising: the crosslinked hydrogel of claim 19 as a matrix; and a sample material immobilized in the matrix.

25. The hydrogel composite of claim 24, wherein the sample material comprises biological tissue.

26. The hydrogel composite of claim 24, wherein the sample material comprises mammalian tissue or cells selected from the group consisting of brain tissue, kidney tissue, and portions thereof.

27. The hydrogel composite of claim 24, wherein the sample material is linearly expanded by a factor in a range of 1.5 to 10 relative to an original size of the sample material before being incorporated into the hydrogel composite.

28. The hydrogel composite of claim 24, wherein the sample material is fluorescently stained.

29. The hydrogel composite of claim 28, wherein the sample material comprises a fluorophore having at least one of condition (I) and condition (II):(I) an excitation wavelength at which absorption by the photocleavable crosslinker isnot more than 1%; and(II) an excitation wavelength at least 50 nm different from a cleavage wavelength of the photocleavable crosslinker.

30. A method for forming a hydrogel composite, the method comprising: combining a sample material with a monomer solution comprising (i) the photocleavable crosslinker of claim 6 and (ii) one or more hydrogel monomers; polymerizing the photocleavable crosslinker with the one or more hydrogel monomers to form (i) a crosslinked hydrogel from the photocleavable crosslinker and the one or more ethylenically functional hydrogel monomers, and (ii) a hydrogel composite comprising the crosslinked hydrogel as a matrix and the sample material immobilized in the matrix; and expanding the hydrogel composite with water.

31. The method of claim 30, wherein the sample material is fluorescently stained.

32. A method for volumetric optical imaging of a sample material, the method comprising:(a) providing the hydrogel composite of claim 24;(b) optically imaging an image layer of the hydrogel composite;(c) photochemically sectioning a depolymerization layer of the hydrogel composite; and(d) repeating steps (b) and (c) a plurality of times for sequential image layers and depolymerization layers.

33. The method of claim 32, wherein: optically imaging the image layer of the hydrogel composite is performed to obtain a volumetric section image of any sample material in the image layer; and the method further comprises (e) stacking the sequential image layers to form a composite volumetric image.

34. The method of claim 32, wherein the image layer has a thickness greater than a thickness of the depolymerization layer.

35. The method of claim 32, wherein a volumetric image of the sample material in the top image layer comprises voxels having a length dimension in a range of 5 nm to 100 nm.

36. The method of claim 32, wherein a volumetric image of the sample material in the top image layer comprises voxels having a length dimension in a range of 5 nm to 300 nm.

37. The method of claim 32, wherein optically imaging the image layer comprises performing on-block fluorescence imaging.

38. The method of claim 32, wherein: optically imaging the image layer comprises exposing the image layer to electromagnetic radiation at one or more excitation wavelengths in a range of 400 nm to 800 nm; photochemically sectioning the depolymerization layer comprises exposing the image layer to electromagnetic radiation at a photocleavage wavelength in a range of 100 nm to 450 nm; and at least one of condition (I) and condition (II) is satisfied:(I) absorption by the photocleavable crosslinker is not more than 1% at the one or more excitation wavelengths; and(II) the one or more excitation wavelengths are each at least 50 nm different from the photocleavage wavelength.

39. The method of claim 38, wherein photochemically sectioning the depolymerization layer comprises exposing a portion of the image layer to electromagnetic radiation at a photocleavage wavelength in a range of 100 nm to 450 nm.

40. The method of claim 32, wherein photochemically sectioning the depolymerization layer comprises exposing the image layer to a multi-photon light source.

41. The method of claim 32, wherein: the hydrogel composite is in an aqueous environment.

42. The method of claim 32, wherein photochemically sectioning the depolymerization layer comprises photodegrading the depolymerization layer to form a remaining bulk portion of the hydrogel composite.

43. The method of claim 32, wherein photochemically sectioning the depolymerization layer comprises photodegrading a portion of the depolymerization layer at an interface between the depolymerization layer and the image layer, thereby releasing a cleaved hydrogel composite portion from a remaining bulk portion of the hydrogel composite.

44. A kit comprising: the photocleavable crosslinker of claim 6; and one or more of ethylene or vinyl functional hydrogel monomers, polymerization reagents, fluorescent stain reagents, light sources, gelation chambers, and brushes.

Images