MicroMagnify: A Multiplexed Expansion Microscopy Method for Pathogens and Infected Tissues

Abstract Super‐resolution optical imaging tools are crucial in microbiology to understand the complex structures and behavior of microorganisms such as bacteria, fungi, and viruses. However, the capabilities of these tools, particularly when it comes to imaging pathogens and infected tissues, remain limited. MicroMagnify (µMagnify) is developed, a nanoscale multiplexed imaging method for pathogens and infected tissues that are derived from an expansion microscopy technique with a universal biomolecular anchor. The combination of heat denaturation and enzyme cocktails essential is found for robust cell wall digestion and expansion of microbial cells and infected tissues without distortion. µMagnify efficiently retains biomolecules suitable for high‐plex fluorescence imaging with nanoscale precision. It demonstrates up to eightfold expansion with µMagnify on a broad range of pathogen‐containing specimens, including bacterial and fungal biofilms, infected culture cells, fungus‐infected mouse tone, and formalin‐fixed paraffin‐embedded human cornea infected by various pathogens. Additionally, an associated virtual reality tool is developed to facilitate the visualization and navigation of complex 3D images generated by this method in an immersive environment allowing collaborative exploration among researchers worldwide. µMagnify is a valuable imaging platform for studying how microbes interact with their host systems and enables the development of new diagnosis strategies against infectious diseases.

washed three times before imaging with Nikon CFI Plan Apo VC 60× C WI (1.2 NA).Samples were then subjected to μMagnify, Chen2021 [1] and Korovesi2022 [2] .Briefly, for Chen2021, C. albicans biofilm were resuspending in 1.2M D-Sorbitol in 0.1M KH2PO4.100 μL of digestion solution containing 10U/mL Lyticase 1M sorbitol, and 50mM Tris buffer was added in the mixture.Incubated for 30min at 30•C with gentle shaking.Before polymerization, the sample was incubated with 0.25% glutaraldehyde (GA) for 10min at room temperature and washed several times in PBS.A monomer solution (8.625% sodium acrylate (w/w), 2M NaCl, 2.5% acrylamide (w/w), 0.15% N, N,-methylenbisacrylamide (w/w) in 1x PBS) was added with 0.2% (w/w) tetramethylethylenediamine (TEMED) 1:50 with 10% stock solution 0.2% (w/w) ammonium persulfate 1:50 with 10% stock solution.Sample was incubated with monomer solution at RT for 5 mins.After 2 hrs polymerization at 37℃, the gel was digested in 8 U/mL proteinase K in digestion buffer (50mM Tris pH 8.0, 0.5% Triton X-100, 1mM EDTA, and 0.8M guanidine HCl) for isotopic expansion, for 1hr at 37℃.Digested gels were completely submerged in fresh ddH20 every 0.5 h 3-5 times unless the gel size did not increase.For Korevesi2022, sample was resuspended in 6 mL of sorbitol buffer (1.2 M sorbitol solution in 0.1 M KH2PO4).Add 0.3 μL of zymolyase (5 U/μL) to 200 μL of fixed cells solution, incubate at 30 °C for 10 minutes. 1 mL of 0.1 mg/mL acryloyl X-SE solution in PBS was added to the sample and allow incubation for 12 hours (overnight) at RT. Cooling the reagents (monomer solution, 99% TEMED, and 50% APS) and the gelation chamber slides on ice for at least 10 minutes prior to gelation.Monomer solution contains: 19 g/100 mL SA, 10 g/100 mL AA, 0.1 g/100 mL Bis in 1x PBS.2.5 uL TEMED stock solution was added to the monomer solution (493ul), followed by adding 5 uL APS stock solution.The mixture was vortexed thoroughly.The slides were kept for 5 more minutes on ice then incubated for 1 hour at 37 °C in a humidified chamber.Coverslip was carefully removed from the slide and the polymerized gel was transferred to a tube. 2 mL of denaturation buffer was added to incubate the gel for 15 minutes at RT.After incubation at 95 °C for 90 minutes, the gel was poured from the Eppendorf tube to a big clean plastic plate to remove excess denaturation buffer.ddH2O was added to Petri dish until the gel is completely submerged.Change ddH2O 2x after 30 minutes and expand the gel in ddH2O overnight at RT.For C. albicans-infected U2OS cells, Chen2021 method was applied the similar way.The staining (Dil in PBST buffer), digestion and GA anchoring were all applied to the cell culture coverslip in the petri dish.For S. aureus-infected U2OS cells, a staining buffer containing DAPI and Dil in PBST (0.1% Tween20) was applied for pre-expansion imaging at RT for 1 hr.For Götz2020, samples were incubated overnight at 37°C in either PBS containing 0.02-2 mg mL−1 lysozyme (ThermoFisher Scientific, Waltham, MA) to digest bacterial cell walls.Sample was treated for 10 min with 0.25% GA at RT and gelated after three washing steps.1ml of the monomer solution containing 0.267 g DMAA (Sigma, 274135) and 0.064 g sodium acrylate (Sigma, 408220) dissolved in 0.57 g ddH2O was degassed for 45 min on ice with nitrogen followed by the addition of 100 μl KPS (0.036 g/ml, Sigma, 379824).After another 15 min of degassing and the addition of 4 μl TEMED per ml monomer solution, gelation was performed for 30 min at RT followed by an incubation of 1.5 h at 37 °C.Hereafter the samples were digested for 3 hrs in digestion buffer (50mM Tris pH 8.0, 1mM EDTA (Sigma, ED2P), 0.5% Triton X-100 (Thermo Fisher, 28314) and 0.8M guanidine HCl (Sigma, 50933)), supplied with 8 U/ml protease K (Thermo Fisher, AM2548).For Kunz2021, the culture cell coverslip was turned upside down on a drop of the monomer solution (8.625% sodium acrylate, 2.5% acrylamide, 0.15% N, N′-methylene bisacrylamide, 2 M NaCl in 1x PBS) containing freshly added 0.2% ammonium persulfate and tetramethyl ethylene diamine for polymerization.The gel was allowed to polymerize for 90 min at room temperature.The polymerized gel was then removed from the glass slides with tweezers and transferred to digestion buffer (50 mM Tris pH 8.0, 1 mM EDTA, 0.5% Triton X-100 and 0.8 M guanidine HCl, containing 5 mg/mL lysozyme and 50 μg/mL.After 20 min, 8 U/ml protease K (Sigma, P4850) was added for another 30 min.Afterwards, gels were washed and expanded in excess of ddH2O.

Supplementary Note 2: Candida albicans infected mouse-tongue sample preparation.
Three days before infection, inoculate a colony of the C. albicans strain SC5314 into 10 ml of YPD broth and incubate it overnight at 30 ℃ with shaking at 200 r.p.m.The next day, transfer 100 μl of the overnight culture to 10 ml of fresh YPD broth and incubate it overnight at the same condition.Repeat this step one more time.The day before the infection, weigh the BALB/c mice (18-25 g; Taconic Farms, cat.no.Balb-M).Use their average weight to calculate the dose of cortisone acetate (Sigma-Aldrich, cat.no.C3130), which should be administered at a concentration of 225 mg/kg in a total volume of 0.2 ml.Use a 1-ml syringe with a 5/8-inch, 25-G needle to inject the animal with 0.2 ml of cortisone acetate in sterile PBS containing 0.05% (v/v) Tween 80 subcutaneously in the dorsum of the neck.Place the isothermal pads in a 60 ℃ water bath overnight.On the day of infection, reduce the water bath temperature to 37 º C before use to avoid overheating the mice.Add 1 ml of the YPD overnight culture to 9 ml of sterile PBS.Centrifuge at 1,000g for 5 min.Decant the supernatant, resuspend the pellet in 10 ml of sterile PBS.Repeat the centrifuge step and resuspend the pellet in 10 ml of sterile HBSS (Sigma, Cat.No. H9269).Dilute the aliquot to make up a suspension of 1x10 6 /ml organisms in 5 ml of sterile HBSS.Warm the C. albicans suspension to 30 ℃ in a dry block and place the calcium alginate swabs in the suspension for 5 min before they are used to inoculate the mice.Remove the isothermal pads from water bath and place two of them on the stainless-steel pan covered with a paper towel.Inject each mouse intraperitoneally with an anesthetic mixture consisting of 10 mg/ml ketamine (Western Medical Supplies, Cat.No. 4165) and 1 mg/ml xylazine (Western Medical Supplies, Cat.No. 5533) in sterile PBS, administering 0.1 ml per 10 grams of body weight.After the mouse is anesthetized in 20-30 min, place the mouse in the supine position on the isothermal pad with a saturated calcium alginate swab placed sublingually for 75 min.At the first and third day after infection, inject the animal with 0.2 ml of cortisone acetate subcutaneously in the dorsum of the neck with the same dose as before.Five days after infection, administer the anesthetic mixture.Once the mouse is anesthetized, euthanize it by cervical dislocation.Excise the tongue and attached oral tissues by dissecting scissors and forceps and place them on a petri dish.Fix the tongue in zinc-buffered formalin for 4 hr at RT.Fixed tissue can be stored in 80% v/v ethanol before processing for histopathology.For paraffinization, the sample was dehydrated in a series of ethanol solution: 80% v/v, 95% v/v, 95% v/v, 100% v/v,100% v/v, 100%v/v each for 1hr.Sample was washed twice with Citrosolv™ (Fisher Scientific, Cat.No. 04355121), 30 min each time.Place the sample in 60 ℃ pre-heated paraffin, repeat this step twice more.Immediately prior to embedding, spray wax mold with Mold Grease (IMS, Cat.No. 105998) and coat the bottom layer of the mold with melted paraffin.Place the sample on top of the base wax.Partially cover the sample with more paraffin.Place pathology cassette on top of the mold and completely fill the mold.Place mold with sample on cooling embedding station until the sample solidify in 10-15 min.The embedded sample was removed from the mold and subjected to sectioning at a thickness of 5 μm.

Supplementary Figures
Figure S1: The full workflow for μMagnify.

Figure S2: Comparison of distortion among μMagnify and other published methods for various pathogen expansion. (a)
Candida albicans (C.albicans) biofilm was fixed with 4% PFA and stained with LEL (green and magenta for pre-and post-images).Samples were subjected to different expansion methods: μMagnify, Chen2021 [1] and Korovesi2022 [2] .Deformation field between pre-expansion images (green) taken by Nikon CFI Plan Apo VC 60× C WI (1.2 NA) and post-expansion images (magenta) imaged with the same microscopy setting were overlayed on the registered images in the middle column.Biological scales 2 μm.Right column shows zoom-in views of boxed region with detailed structure of septum ring post-expansion by different methods.μMagnify enables visualization of intact septum structure while the other two methods present cracking at the septum rings (Chen2021 and Korovesi2022) and failed expansion (Chen2021).Scatter plot of the root mean square (RMS) error ratio per measurement length (μ) for each ROI by μMagnify (n=6), Chen2021 (n=5) and Korovesi2022 (n=7).Bars for each group represent the mean of error ratio with error bars showing the standard deviation.ANOVA test indicates a significant difference p< 0.05 among the three groups.Tukey's HSD test show pairwise difference of distortion between: μMagnify and Korovesi2022 (asterisk, p<0.05).(b)Yeast form C. albicans was fixed with 4% PFA and stained with Dil (green and magenta for pre-and post-expansion images).Samples were subjected to different expansion methods: μMagnify and Chen2021.Deformation field between pre-expansion images (green) taken by Nikon CFI Plan Apo VC 60× C WI (1.2 NA) and post-expansion images (magenta) imaged with the same microscopy setting were overlayed on the registered image in the right column.Biological scales 2 μm.Scatter plot of the RMS error ratio per measurement length for each ROI by μMagnify (n=5), Chen2021 (n=5).Bars for each group represent the mean of error ratio with error bar showing the standard deviation.ANOVA test indicates a significant difference p< 0.05 among the three groups.Tukey's HSD test show pairwise difference of distortion between: μMagnify and Chen2021 (asterisk, p<0.05).(c) Staphylococcus aureus (S. aureus)-infected U2OS cells were fixed with 4% PFA and stained with DAPI (green and magenta for pre-and post-expansion images) and Dil (yellow).Samples were subjected to different expansion methods: μMagnify, Götz2020 [3] and Kunz2021 [4] .Deformation field between pre-expansion images (green) taken by Nikon CFI Plan Apo VC 60× C WI (1.2 NA) and post-expansion images (magenta) imaged with the same microscopy setting were overlayed on the registered image in the middle column.Biological scales 2 μm.Boxed region in the left column show the zoom-in views of S. Aureus DNA of pre-and post-expansion images.μMagnify enables consistent DNA expansion between pathogen and host cells while the other two methods present heterogeneous expansion pattern between pathogen and host cells (Götz2020 and Kunz2021).Right column shows lipid retention among different methods, as μMagnify reveals pathogen cell membrane by post-expansion lipid stain (Dil) while others failed.Scatter plot of the RMS error ratio for each ROI by μMagnify (n=17), Götz2020 (n=6) and Kunz2021 (n=6).Bars for each group represent the mean of error ratio with error bar showing the standard deviation.ANOVA test indicates a significant difference p< 0.05 among the three groups.Tukey's HSD test show significant pairwise difference between all three groups (asterisk, p<0.05).

Figure S7: Demonstration of signal unmixing algorithm with three rounds of staining. (a)
With all biomolecules of target anchored onto the hydrogel, antibodies and fluorescent labels were accumulatively added to the sample in separate round of staining along with one reference stain.Numbers of stain for each round is determined by the microscope channel capacity.Raw image (image2) captured at the second round consists of the true signal R2 and R1 times a coefficient alpha.Likewise, raw image i captured at round i consists of Ri, R1 times a coefficient alpha, R2 times a coefficient beta, et.al.The true signal of each round (Ri) equals to image i subtracted image i-1 times a coefficient.Enumeration of possible coefficient allows you to resolve an optimal coefficient that minimizes the mutual information between Ri and image i-1.(b) For demonstration, S. aureus infected samples were stained for three rounds, with DAPI as reference channel for image registration.Registered raw images of three rounds were shown in the raw image column, with raw signal of each channel is assigned a type of color.Using unmixing algorithm, true signal of each channel of each round could be unmixed, as shown in GFP, vimentin, ConA, NEDD4, alpha-tubulin, Dil, CD63, WGA, and NHS images in the middle column.Pre-unmixing and post-unmixing comparison was shown on the left column with each channel re-assigned an individual color.Table S1: Fixations and homogenization and expansion factors for different pathogen samples * Heat denaturation buffer (1% w/v SDS, 8M Urea, 25 mM EDTA, 2× PBS, pH 7.5 at RT) ** Digestion cocktail consists of 500U/mL Mutanolysin, 500U/mL Lysostaphin, 50U/mL Zymolyase, 1kU/mL Collagenase (optional, C2799, Sigma).*** Expansion factors were measured in H2O, except for specially specified.Expansion factor error in terms of s.e.m. over N biological replicates, n technical replicates.

Figure S3 :
Figure S3: Examples of morphological details revealed by μMagnify comparing to preexpansion confocal images.Both pre-and post-expansion images were taken at 60x (water).(a) Comparison between pre-(up) and post-expansion (bottom) images of S.P. D39 strain stained with DAPI (blue) and NHS-AAtto647(magenta).(b) Comparison between pre-(up) and postexpansion (bottom) images of S.A. that was stained with DAPI (blue, only post-expansion) and Dil (red).(c) Comparison between pre-(left) and post-expansion (right) images of E.C. that was stained with DAPI (blue) and Dil (red).(d) Comparison between pre-(up) and post-expansion (bottom) images of yeast form C.A. that was stained with BPDIPY (green).(e) Comparison between pre-(up) and post-expansion (bottom) images of C.A. hyphae cells that was stained with DAPI (blue), Dil (red, only post-expansion) and LEL (green, only post-expansion).(f) Comparison between pre-(left) and post-expansion (right) images of SA infected U2OS cells that was stained with DAPI (blue), Dil (red).(g-h)Comparison between pre-(g) and postexpansion (h) images of U2OS (LITAF-GFP) cells that was stained with DAPI (blue), anti-GFP (green), anti-NEDD4 (red), and anti-CD63 (magenta).On the left is the overlay image with four markers.On the right is montage image of four markers with half-size.All images were presented in physical scales: 10μm.

Figure S4 :
Figure S4: RNA in situ hybridization in 4% PFA fixed mNeon expressing E. coli.Test group shows images of E. coli with different expression levels of mNeon proteins (green), mNeon RNAs (red) and 16s rRNA (blue).Ctrl1 shows the same sample that stained with mNeon rRNAs (red) and PEG-10 RNA probes (blue).Ctrl2 shows the same sample that stained with only imager probes (the one for mNeon RNA, in red) and mScarlet RNA probes (blue).Ctrl3 shows an uninduced E. coli, that carrys an mNeon expressing plasmid.Sample was stained with mNeon RNAs (red) and 16s rRNA (blue).Scale bar: 5 μm.

Figure S6 :
Figure S6: Representative bright field images of different types of microbial keratitis in comparison with μMagnify generated fluorescence images.(a) LEFT: PAS image of Candida keratitis of cornea taken at 40x.RIGHT: 3D reconstruction of μMagnify image for dense Candidiasis in cormea, with sample stained by DAPI (cyan), WGA (yellow), Dil (red), and NHS-Atto647N (magenta).Scales: 10μm.(b) LEFT: H&E image of eyeball sample with Staphylococcus epidermidis keratitis taken at 4x, scale: 250μm.RIGHT: 40x Zoon-in view of green arrow-pointed region on the left.Blue dot reveals the microbes, scale: 10μm.(c) Gram stain image of Pseudomonas keratitis cornea, scale: 100μm.Representative views of (i) heavily infected spot (outlined and pointed with green arrow), (ii) mild infection of keratocytes that cannot be revealed by bright field image versus (iii) 3D reconstruction of single cell level infection revealed by μMagnify, with sample stained by DAPI (cyan), WGA (yellow), and NHS-Atto647N (magenta).(i-iii) scale: 10μm.(d) LEFT: AFB image of cornea tissue with Atypical mycobacterial keratitis taken at 4x, scale: 100μm.RIGHT: a zoom-in view of green arrow pointed region on the left.Magenta rod-shape objects reveal the bacteria, scale: 10μm.(e) H&E image of Acanthamoeba keratitis cornea, with representative views of acanthamoeba infections circled by green outline, scale 100μm.(i) zoom-in view of cell wall (green arrow pointed) of dead acanthamoeba.(ii-iv) zoom-in view of acanthamoeba infection.(i-iv) scale: 10μm.(f) Representative H&E images (taken at 40x) of cornea tissue containing: 1. squamous epithelium, 2. polygonal cell, 3. basal epithelium, 4. Bowman's layer, 5. Keratocytes, 6. Collagen fibrils.

Figure S9 :
Figure S9: Correlation and colocalization test on pre-expansion confocal images.(a) 4channel correlation matrix for pre-expansion confocal images capturing SA-infected wildtype U2OS cells (n=13).(b) 4-channel correlation matrix for pre-expansion confocal images capturing SA-infected mutant U2OS cells (n=7).(c) Delta matrix between matrix in (b) and (a) showing the differences in channel correlations between mutant and wildtype U2OS cells.(d) 4channel colocalization matrix for pre-expansion confocal images capturing SA-infected wildtype U2OS cells (n=13).(e) 4-channel colocalization matrix for pre-expansion confocal images capturing SA-infected wildtype U2OS cells (n=7).(f) Delta matrix showing the differences in channel colocalization between mutant and wildtype U2OS cells.

Table S2 :
Polymer synthesis for different pathogen samples

Table S3 :
RNA FISH probes design

Table S4 :
P-values of ANOVA test for colocalization matrices between mut and wt