An optimized protocol for combined fluorescent lectin/immunohistochemistry to characterize tissue-specific glycan distribution in human or rodent tissues

Summary Lectin histochemical analysis of tissues combined with immunohistochemistry is a valuable tool to characterize and correlate the spatial distribution of glycans with the presence of specific cell types or antigens of interest. The current protocol describes the application of monosaccharide motif specificity of lectin binding to glycan residues to different tissue types. In addition, we describe stereological methods to provide further quantification of the analyzed tissues. For complete details on the use and execution of this protocol, please refer to Mohd Isa et al. (2018), Contessotto et al. (2020), and Samal et al. (2020).

*Correspondence: abhay.pandit@nuigalway.ie https://doi.org/10. 1016/j.xpro.2020.100237 SUMMARY Lectin histochemical analysis of tissues combined with immunohistochemistry is a valuable tool to characterize and correlate the spatial distribution of glycans with the presence of specific cell types or antigens of interest. The current protocol describes the application of monosaccharide motif specificity of lectin binding to glycan residues to different tissue types. In addition, we describe stereological methods to provide further quantification of the analyzed tissues. For complete details on the use and execution of this protocol, please refer to Mohd Isa et al. (2018), Contessotto et al. (2020), and Samal et al. (2020).

BEFORE YOU BEGIN
Experimental design considerations 1. Lectin/antibody staining with fluorescently tagged lectins or antibodies should be ideally performed in frozen (cryostat) tissue sections (either from human or animals), with a thickness that can vary between 5 and 10 mm to enable also high resolution imaging (maximum thickness of 10 mm would prevent super-imposition of histological features). In our experience, paraffin embedded tissue sections increase the non-specific binding of fluorescently labeled lectins, enhancing background staining. The use of permanently positively charged glass slides (Super-Frost Plus) is strongly recommended to avoid the detachment of the tissue sections during this procedure. 2. Throughout the whole staining procedure, tissue sections need to be protected from direct light exposure as fluorescently (e.g., FITC/TRITC-) labeled lectins quench more quickly than Alexa Fluor-labeled probes. Additionally, stained sections should be imaged within 4 days to ensure specific binding. Fluorescently labeled lectins should be stored according to the manufacturer's instructions. Stock solutions (after reconstitution, in case the lectin is purchased as a powder) should be aliquoted and kept at À20 C for long-term storage, or at 4 C for up to 1 month. Avoid freeze-thawing cycles and always protect fluorescently labeled lectins from direct light exposure. 3. An optimization of lectin concentration and incubation time should be done to ensure the optimal concentration will be used in the following analyses (suggested concentrations for the titration: 5 mg/mL, 10 mg/mL, 15 mg/mL, 20 mg/mL). Please note that optimal concentrations will depend on the lectin as well as on the tissue used. The optimal concentration of lectin to be used is achieved when an increase in concentration above the optimal concentration does not increase the degree of staining of the tissue. 4. While the optimal concentration is being determined, the specific negative (inhibitory) control needs to be performed in parallel. Specifically, the lectin should be incubated for 30 min at room temperature with the sugar it is specific for (haptenic sugar) before staining the tissue section (selection of relevant lectins as well as their respective haptenic sugars is described in Table 1). This will enable the binding of the lectin to the sugar, resulting in the reduced binding to the tissue and subsequent decreased signal intensity, confirming the lectin specificity. It is also important to perform a positive control in parallel during the optimization step to guarantee that in case there is no labeling, it is not due to technical errors. Additionally, both negative and positive controls should be included in every batch of lectin labeling to ensure successful and specific labeling. 5. An optimization of primary antibody dilution and incubation conditions should also be carried out. The respective negative control should be done in parallel, where sections are incubated with a non-related primary antibody, to assess if there is non-specific binding of the secondary antibody to the tissue. 6. Bear in mind that phosphate-buffered saline (PBS) cannot be used throughout the whole procedure since phosphate ions bind to metal ions and sequester them from the lectin-binding sites (Brooks and Hall, 2012).

Periodate treatment of bovine serum albumin
Timing: 3 days Note: Treatment of bovine serum albumin (BSA) is necessary to remove glycosylated contaminants (e.g., serum glycoproteins) that might promote lectin binding directly to BSA, interfering with the results of the assay (Glass, Briggs and Hnilica, 1981). It is recommended to use 99% purity BSA or greater to reduce any potential contaminants present as well.
CRITICAL: Periodic acid is an oxidizing agent and a severe respiratory tract irritant. Therefore, while handling periodic acid, lab coat, gloves, powder particulate face mask, and safety goggles should be worn. Preparation of periodic acid solution as well as BSA dissolution should be done with caution in a fume hood, and it is crucial that the pH is kept at 4.5.
Note: pH should be adjusted using glacial acetic acid and NaOH, if needed.
8. Dissolve high purity (99% purity or greater) BSA (Sigma, A7638) in freshly made periodic acid solution (recommended concentration: 5% BSA in freshly prepared periodic acid solution). 9. Incubate at room temperature (RT) (~18 C-23 C) for 6 h under mild stirring using a magnetic stirrer on a plate. CRITICAL: Always visually inspect the sample to ensure that the BSA is completely dissolved. Do not vortex the solution as it creates bubbles.
10. Dialyze solution against deionized water for 48 h in a 5 liter beaker with stirring ( Figure 1). 11. Keep dialysis set up at 4 C and change the water a total of four times (recommended to do it 3 h, 12 h, 24 h, and 36 h after starting the dialysis). 12. Freeze the solution at À80 C overnight (~16 h). It is recommended to freeze in aliquots of 20 mL inside 50 mL centrifuge tubes at an angle to facilitate the freezing of the inner core of the aliquots and to improve lyophilization. 13. Lyophilize it on the next day in a freeze drier until completely dry (for a volume of 20 mL it should take 2-3 days). 14. Store lyophilized periodate-treated BSA at 4 C in a sealed container (to avoid moisture). The color of the final powder should be between ivory white and white, and the texture should be fluffy ( Figure 2). Figure 2. Final appearance of periodate-treated BSA Its color can range from ivory white to white, and it usually dries with a fluffy texture. If the resulting treated BSA is pink, orange, or brown, then the BSA is oxidized, and should be discarded as there will be solubility issues.

MATERIALS AND EQUIPMENT
Alternatives: In this protocol, a fluorescence confocal microscope is used. However, any fluorescence microscope could be used to image the staining performed as long as it has the adequate filters/lasers needed to image the fluorescent tag on the lectins/antibodies.
CRITICAL: Presence of CaCl 2 and MgCl 2 is crucial for lectin function. Store autoclaved or filter-sterilized solution at RT (~18 C-23 C) for a maximum of 3 months.
CRITICAL: Adjust pH to 7.2 with concentrated HCl or 1 M NaOH, depending if Tris base or Tris-HCl are used initially, respectively. HCl is corrosive and irritant to skin and eyes, so adjustment of pH should be done inside a fume hood. While handling HCl, lab coat, gloves, powder particulate face mask, and safety goggles should be worn.
Note: To prepare TBS-T, add 0.05% of either Tritonä X-100 or Tweenâ 20 to TBS. Whichever detergent is selected, it should be consistently used as these are not interchangeable. Tweenâ 20 is more soluble than Tritonä X-100, so the Tritonä X-100 solution should be gently mixed after addition of the detergent.
Note: Do not use phosphate-buffered saline (PBS) in this protocol since phosphate ions bind to metal ions and sequester them from the lectin-binding sites (Brooks and Hall, 2012).

ll OPEN ACCESS
Note: Paraformaldehyde is toxic, flammable, carcinogenic, corrosive, and irritant to skin, eyes, and lungs airways. Therefore, while handling PFA, lab coat, gloves, powder particulate face mask, and safety goggles should be worn. All steps should be carried out inside a fume hood. PFA waste should be disposed of safely, in a specific container.
CRITICAL: PFA should be dissolved in 800 mL of 13 TBS at 60 C with continuous stirring, inside a fume hood. Add concentrated NaOH until solution is clear and allow it to cool. Filter the solution, adjust pH to 6.9 with concentrated HCl and complete to 1 L with 13 TBS. The final 4% PFA solution can be kept for up to 1 month at 4 C or aliquoted and stored long term at À20 C.

STEP-BY-STEP METHOD DETAILS
Lectin staining of fixed-frozen tissue sections

Timing: 4 h
In this section, lectin histochemistry of tissues is performed to investigate specifically the spatial distribution of glycans in the tissue, which can be eventually combined with antibody staining. Tissuedependent modifications are outlined in the relevant steps.
Note: This protocol is a combination and optimization of other methods previously published by (Brooks and Hall, 2012;Mohd Isa et al., 2018;Kilcoyne et al., 2019;Contessotto et al., 2020).
CRITICAL: Once the staining procedure starts (after the first wash and throughout the whole procedure), it is crucial to always keep the tissue wet and not allow it to dry.
1. Leave tissue frozen (cryostat) sections on a slide warmer at room temperature (~18 C-23 C) between 30 min to 1 h (depending on the size of the tissue) before starting the staining to enable evaporation of moisture and to improve tissue adhesion to the slide. a. If the tissue used was not fixed before being frozen, incubate sections with 4% PFA solution on the slide for 2 min (inside fume hood) 2. Circle/frame each tissue section with a PAP pen to avoid leakage of the lectin solution away from the tissue section. The PAP pen creates a hydrophobic barrier, allowing the use of less reagents.
Note: If the hydrophobic barrier vanishes throughout the following steps, carefully dry the borders of the glass close to the tissue section and repeat this step.
3. Wash sections with TBS-T three times, for 3 min each at RT (~18 C-23 C) in a glass Coplin staining jar placed on a shaking platform at a low speed (60 -80 rotations per minute (rpm)). a. Monitor the sections to be sure they do not detach from the slide. If this happens, check Troubleshooting 1. 4. Lay out the slides with the fixed-frozen sections horizontally, with the tissue section facing up, in an opaque (blackout) StainTray slide staining system.
Alternatives: If a StainTray slide staining system is not available, the slides might be placed on glass rods in an opaque sandwich box lined with damp filter paper or other similar alternatives. Note: Blocking solution should be prepared fresh just before the start of this procedure.
6. Wash sections with TBS three times, for 3 min each at RT (~18 C-23 C) in a glass Coplin staining jar, with gentle shaking.
CRITICAL: From this point, do not expose your samples to direct light for an extended period of time. It is suggested to carry out the remaining steps in darkness or under a red light to prevent the quenching of fluorescence.
7. Incubate sections with fluorescently labeled lectins diluted in TBS-T for 1 h at RT (~18 C-23 C) in a dark chamber (e.g., StainTray slide staining system or an alternative opaque staining equipment). a. We recommend performing a titration experiment to identify the optimal concentration of lectin to be used, as mentioned above. See Table 1 for binding specificity of a selection of relevant lectins and Figure 3 for suggested experimental design.
CRITICAL: Working solutions of lectins should be prepared fresh 30 min before each staining and discarded at the end of the experiment.
Note: The fluorescent label on the lectins should be selected according to the lasers of fluorescence available in the microscope. However, fluorescein isothiocyanate (FITC) and tetramethylrhodamine isothiocyanate (TRITC) are the most commonly commercially available and used fluorescent tags for lectins. b. In parallel, perform haptenic sugar inhibition controls by pre-and co-incubating each lectin with 100 mM of the appropriate sugar in TBS-T at RT (~18 C-23 C) (Akimoto and Kawakami, 2014). See Table 1 for list of haptenic sugars to use with a selection of relevant lectins. Lectins should be prepared in the relevant sugar in TBS-T and pre-incubated for at least 30 min before adding to the tissue and continuing the sugar-lectin co-incubation on the tissue. A positive control should also be performed in parallel to ensure successful labeling. This could be done by using a tissue section that has been reported to display the glycan residues the lectin is specific to (Akimoto and Kawakami, 2014).
Note: The specificity of a lectin is characterized by the sugar residue that inhibits the interaction between the lectin-binding site(s) and its respective glycan. This sugar (haptenic sugar) displays a competitive binding to the lectin-binding sites in relation to the carbohydrate motifs present in the tissue sample (Marks and Rodnight, 1978;Gerlach et al., 2012). Decreased binding in the presence of the haptenic sugar indicates the specificity of the lectin for the sugar. Complete inhibition (negative labeling) might also occur in some cases; however, this does not always happen and it is more common to expect a marked diminution in labeling intensity. No change in binding intensity indicates that lectin binding to the tissue was nonspecific or mediated by a non-carbohydrate binding site on the lectin.
8. Wash sections with TBS-T twice, for 3 min each, and then once with TBS* at RT (~18 C-23 C). We suggest performing drop washes while keeping the sections horizontal in the StainTray slide staining system (or alternative opaque staining equipment). The final wash in TBS without detergent avoids the retention of minimal bubbles on the tissue section. a. *If staining brain tissue: wash twice with TBS-T for 10 min each and twice with TBS for 10 min each. This is recommended for brain tissue due to its high content in lipids and in auto fluorescent lipofuscin, which increases the background signal. Longer and increased number of washes can help to reduce non-specific signal.
Optional: Starting from this point, lectin histochemistry can be combined with immunohistochemistry. For more details, see next step ''Antibody staining following lectin incubation.'' 9. Counterstain either with Hoechst 33342, diluted 1:2,000 (final concentration at 0.01 mM) in TBS for 5 to 10 min, at RT (~18 C-23 C), or with ProLongä Gold Antifade Mountant with DAPI (a single drop is sufficient for an approx. 0.5 cm 2 -sized tissue section). 10. If Hoechst 33342 is used, wash sections with TBS three times, for 3 min each* at RT (~18 C-23 C). a. *If staining brain tissue (or other tissues rich in auto fluorescent lipofuscin): after these washes, incubate with 0.1% Sudan black B solution in 70% ethanol (1 mg/mL) in a glass Coplin staining jar for 5 min at RT (~18 C-23 C). Wash then with TBS twice at RT (~18 C-23 C), for 4 min each.
Note: Sudan black B solution can be re-used up to three times. Sudan black B binds to lipofuscin (granules rich in lipids, and abundant in the adult brain and other tissues), reducing/ eliminating tissue auto fluorescence and background that usually arises from these granules (Oliveira et al., 2010).
11. After washing excess Hoechst 33342, apply one or two drops of Fluoromountä aqueous mounting media and cover the slide with a glass coverslip (a single drop is sufficient for an approx. 0.5 cm 2 -sized tissue section). If ProLongä Gold Antifade Mountant was used previously, this step would not be necessary. 12. Leave slides to cure overnight (~16 h) at RT (~18 C-23 C, in the dark) and seal around the edges of the coverslip with clear nail polish to prevent the coverslip from sliding and the mountant from drying out. 13. Keep slides at 4 C until imaging. We recommend imaging within 4 days for best results. For precious samples, stained slides can be stored at À20 C for up to 1 month. 14. Image stained slides by using a fluorescence microscope. Confocal microscopy enables the acquisition of images from several planes within the tissue section, in a Z-stack, providing a better resolution.
CRITICAL: Image stained slides within 4 days. For unexpected outcomes, check Troubleshooting 1.

Antibody staining following lectin incubation
Timing: up to 2 days In this step, immunohistochemistry can be carried out after lectin staining to complement the previous analysis. This procedure allows for the co-localization of glycan distribution with certain cell types or specific antigens present in the tissue. A general protocol is given, and tissue-dependent modifications are outlined in the appropriate steps.
Pause Point: Tissue sections already lectin stained can be kept overnight covered with TBS in a humidified chamber before starting the immunostaining procedure if the operator cannot continue immediately after performing the lectin histochemistry. CRITICAL: The whole immunostaining step should be carried out in darkness or in a redlight room since the tissue was already incubated with fluorescently labeled lectins and it is crucial to prevent the quenching of the fluorescence. Additionally, the whole immunostaining procedure should be done by placing the slides in a humidified chamber to prevent them from drying.
Note: A prior optimization for the immunostaining should be performed to detect the optimal primary antibody dilution and incubation conditions (time and temperature -usually the primary antibody should be incubated overnight at 4 C or for 2 h at RT). In addition, the usage of either TBS or TBS-T should be tested for the specific antibody to avoid interference with an optimal positive signal intensity (e.g., monoclonal vs polyclonal). The respective negative control should be done in parallel, where sections are incubated with a non-related or isotype control antibody, to assess if there is non-specific binding of the secondary antibody to the tissue. It may be appropriate to block with normal serum to avoid FC-mediated false-positive staining. However, this may affect previous lectin binding as serum contains glycoproteins, so appropriate controls should be employed during staining optimization.
CRITICAL: The whole immunostaining step should be carried out continuing to use TBS and not switching to PBS.

Keep slides in the StainTray slide staining system (or alternative opaque staining equipment)
and block with 3% periodate-treated BSA in TBS, for 1 h at RT (~18 C-23 C).
Note: This step (2 nd blocking) is not crucial and might be skipped. This will depend on antibody staining, as nuclear antibody stains may require serum-based blocking. While this may have an effect on previous lectin staining, this must be optimized by the end-user using appropriate controls to determine suitability of each counterstain, ensuring specificity and minimal off-target binding. Immunohistochemical counterstaining should be reconsidered if interference occurs.
16. Incubate sections with primary antibody at the concentration, time and temperature optimized previously (usually the primary antibody should be incubated overnight (~16 h) at 4 C or for 2 h at RT (~18 C-23 C), diluted in either TBS or TBS-T). 17. Wash sections twice with TBS-T for 10 min each and once with TBS for 10 min, at RT (~18 C-23 C). These should be drop washes while keeping the sections horizontal in the StainTray slide staining system (or alternative opaque staining equipment)*. a. *If staining brain tissue (or other tissues with potential auto fluorescence): vertical washes are recommended using a glass Coplin staining jar. 18. Incubate sections with fluorescently labeled secondary antibody diluted in TBS (at optimized dilution) for 1 h at RT (~18 C-23 C).
CRITICAL: The secondary antibody and lectin used must be labeled with fluorophores with a narrow absorption wavelength without spectral overlap with each other. For instance, if the lectin is FITC-labeled (l ex = 488 nm, l em = 525 nm), then the secondary antibody cannot be Alexa Fluor 532 (l ex = 532 nm, l em = 554 nm) as they would emit at an overlapping wavelength spectrum, causing significant bleed through. A more appropriate combination would be FITC (l ex = 488 nm, l em = 525 nm) and Alexa Fluor 647 (l ex = 650 nm, l em = 671 nm), for example.
19. Wash sections with TBS three times, for 3 min each* at RT (~18 C-23 C). These should be drop washes while keeping the sections horizontal in the StainTray slide staining system (or alternative opaque staining equipment).

OPEN ACCESS
a. *If staining brain/cardiac tissue (or other tissues rich in auto fluorescent lipofuscin): wash once with TBS-T for 10 min each and three times with TBS for 10 min each; vertical washes are recommended using a glass Coplin staining jar. 20. Counterstain either with Hoechst 33342, diluted 1:2,000 (final concentration at 0.01 mM) in TBS for 5 to 10 min at RT (~18 C-23 C) or with ProLong Gold Antifade Mountant with DAPI (single drop is sufficient for an approx. 0.5 cm 2 -sized tissue section). 21. If Hoechst 33342 is used, wash sections with TBS three times, for 3 min each* a. *If staining brain tissue (or other tissues rich in auto fluorescent lipofuscin): after these washes, incubate with 0.1% Sudan black B solution in 70% ethanol (1 mg/mL) in a glass Coplin staining jar for 5 min at room temperature. Wash then with TBS twice, for 4 min each, at RT (~18 C-23 C). 22. After washing excess Hoechst 33342, apply one or two drops of Fluoromountä aqueous mounting media and cover slide with a glass coverslip (a single drop is sufficient for an approx. 0.5 cm 2 -sized tissue section). If ProLongä Gold Antifade Mountant was used previously, this step would not be necessary. 23. Leave slides to cure overnight (~16 h) at RT (~18 C-23 C, in the dark) and seal around the edges of the coverslip with clear nail polish to prevent the coverslip from sliding and the mountant from drying out. 24. Keep slides at 4 C until imaging. For precious samples, stained slides can be stored at À20 C for up to 1 month. 25. Image stained slides by using a fluorescence microscope. Confocal microscopy enables the acquisition of images from several planes within the tissue section, in a Z-stack, enhancing image resolution. Example of an expected outcome is described in Figure 5.
CRITICAL: We recommend imaging within 3 days for best results.

EXPECTED OUTCOMES
After performing the protocol for combined lectin and immunohistochemistry, it is possible to colocalize the expression of antigens of interest with glycan distribution. For example, in Figure 5, mouse intervertebral disc was stained with a marker for chondrocytic phenotype (SOX9, SRY-related box 9 transcription factor) and DSA lectin (Datura stramonium; specific for GlcNAc/GalNAc residues). Since lectin histochemistry can be challenging and the outcomes can be very distinct depending on the type of tissue used, in Figure 6 there is a compilation of expected outcomes of SNA I (Sambucus nigra agglutinin I) staining on different tissue types (heart, brain, and intervertebral disc tissues). In the same figure, there are also examples of commonly seen errors, which will be helpful to unexperienced users to interpret their results and to predict what could be considered an expected result or a suboptimal outcome.

QUANTIFICATION AND STATISTICAL ANALYSIS
Quantifying area of tissue fluorescence with ImageJ/FIJI Stereology quantification is one method that may be adopted to calculate the percentage (%) volume fraction of detectable lectin binding in each tissue sample. Lectin histochemical binding fluorescence should be obtained from eight to ten microscopic views of each slide with three technical and at least three biological replicates using FIJI/ ImageJ software version 1.51 (NIH, Bethesda, MD, USA). This is the recommended number of fields to acquire if magnification is 203; however, if magnification used increases, then the number of fields acquired should increase as well. The use of z-stack image capture is recommended to reduce error in image focusing and perceived image intensity. Volume fraction (Vv) is calculated by quantifying the area fraction of the positively stained component divided by the total area of interest and converting into a percentage (%) as below: Percentage Volume Fraction ð%VvÞ = Area Fraction Total Area x 100% Note: Prior to image processing, we recommend using .tif or .oib file types for future batch processing.
3. Set measurement scale by drawing a line over the scale bar, then: Analyze / Set scale. 6. Calculate the area of fluorescent tissue by surrounding the fluorescent tissue with the rectangular selection tool, and then: Analyze / Analyze particles. 7. Enter the minimum particle size according to the magnification used and tissue stained, toggle Show Outlines, check Display Results, and click OK. Outline of the analyzed area will be drawn. 8. Calculate the entire area of the tissue by performing the threshold of the new image using manual setting: Image / Adjust / Threshold (Adjust sliders to include complete area of the tissue and click Apply). 9. Surround the entire tissue with the rectangular selection tool, and then: Analyze / Analyze particles. 10. Calculate %Vv as mentioned above.
Alternatives: If the tissue section covers the entire image, follow steps 1 to 4 as stated above, and then employ a simple method of fluorescent area calculation as follows: Figure 6. Examples of expected outcomes from lectin histochemistry using different types of tissues All tissues were stained with SNA I (Sambucus nigra agglutinin I), which is specific for Neu5Ac-a(2,6)-Gal/GalNAc residues. Examples of commonly seen errors are also provided, as well as potential solutions.

OPEN ACCESS
STAR Protocols 2, 100237, March 19, 2021 11. Select Analyze / Set Measurements, Check Limit to Threshold, then convert to binary image and select Analyze / Measure. 12. Calculate %Vv as mentioned above While these methods can accurately calculate area fraction of fluorescence, the authors recommend that if the tissue covers the entire background of the image, then only step one should be employed (assuming the entire area of the tissue = the area of the entire image). This way, batch processing can be performed in an entirely objective quantification protocol. Additionally, this whole protocol can be included as a Macro, which is a program that automates all these commands, allowing the quantification of the stained area to be done in one step.
Quantifying intensity of tissue fluorescence with ImageJ/FIJI Besides quantifying the area of tissue stained and subsequent volume fraction, another type of quantification that can be performed relates to the measurement of staining intensity in the tissue.
For that, a microarray grid should be used and the preliminary settings should be calibrated according to the distribution of the staining, which would be tissue-dependent as well.
13. Open the micrograph to be analyzed in ImageJ/FIJI. 14. Convert the image to an 8-bit image (grayscale): Image / Type / 8-bit. 15. Set measurement scale by drawing a line over the scale bar, then: Analyze / Set scale. 16. In the Set Scale window enter the actual distance in Known Distance box, confirm unit of measurement is appropriate and check Global. 17. Apply the grid over the image: Analyze / Tools / Grid, and select Circles in Grid type. Define Area per point according to the tissue and staining used. 18. Calculate the mean intensity of fluorescent staining inside the grid points by selecting: Image / Overlay / To ROI Manager. 19. In the ROI Manager, select GRID1 and check Measure. A new window will pop up, and the intensity measurement will be described as Mean Grey Value.
As an example, see Figure 4, where a circular-fitted microarray grid was set to quantify the fluorescence intensity staining of capillaries and cardiomyocytes membranes in light of the homogenous distribution of these morphological components in left ventricular cardiac tissue.
Alternatives: If the ''Grid'' tool is not available, another similar plugin might be used after step 16.
20. Select Plugins / Microarray profile (which is available online for free download as ImageJ jar) /Change ROI (Region of interest). Set the shape according to the distribution of the region of interest (if the distribution of the ROI is homogenous, a circled one is suitable as a sampling unit) /Measure RT. 21. A new window will pop up and the relative fluorescence intensity will be described as Mean Grey Value in a results table.

LIMITATIONS
Although lectin histochemistry is an easy, quick, and straightforward protocol to establish across research groups focused on different fields, it has some limitations associated to it. The most important one would be that even though lectins can be used to detect carbohydrates in both glycoproteins and glycolipids, they can only discriminate between the identity of the actual carbohydrate residues present and not the nature of the glycosylated molecule. Additionally, lectins detect mainly terminal residues (Varki et al., 2009), and cannot elucidate the sequence of the entire oligosaccharide chain. For further information on full glycan structures present or the nature of the glycosylation molecule, the sample may be enriched for the membrane or extracellular matrix portions for further analyses on the glycans (e.g., lectin microarrays and liquid chromatography-tandem mass spectrometry (LC-ESI-MS/MS)). While highly informative in expert facilities, these investigative techniques remain highly challenging, with difficulty in interpretation and often require tissue-specific optimization.
Furthermore, although the use of fluorescently labeled lectins provides sharp and detailed images, and allows the co-staining with antibodies, it requires the imaging to be carried out within a short period of time post-staining, as their fluorescence fades within a few days.
The broad binding specificity of some lectins poses another significant limitation, since it might promote the binding of those lectins to structures that are not the ones of interest.
Moreover, the mostly qualitative (and only semi-quantitative) nature of this method presents some hurdles to an accurate quantification of the expression of each sugar residue in the tissue sample. This is due to the multivalent binding of lectins. Nonetheless, relevant stereological approaches can be further integrated in the analysis of the structures of interest (e.g., cellular membranes, extracellular matrix proteins). Quantification of lectin histochemistry images may be further validated using semi-quantitative western/lectin blotting techniques.

TROUBLESHOOTING Problem 1
Sections detaching from charged glass slide (step 3).

Potential solutions
When performing vertical washes (in glass Coplin staining jars), it is possible that tissue sections start to detach. To prevent this, numerous approaches can be taken into consideration: Solution 1: While cutting the sections in the cryostat and placing them onto the glass slides, they should be kept at room temperature for 30 min to 1 h to promote complete adherence of the tissue section to the glass. Solution 2: Slides such as APES (aminopropyltriethoxysilane) treated glass slides may be used (instead of SuperFrostä Plus slides) to increase tissue adherence (https://theolb.readthedocs. io/en/latest/molecular-biology/apes-treatment-of-slides.html). Solution 3: Before starting the staining, enough time should be given for the slides to be at room temperature, to ensure that the sections are completely attached to the glass slides. Solution 4: If tissue is not already fixed, fix with PFA on slide to promote further adhesion. Solution 5: While performing the staining, gently replace the buffer between washes and, when using shaking, select a low speed. Solution 6: If the sections continue to detach, carry out all washes horizontally in a drop-wise fashion.

Problem 2
High fluorescence signal is detected in sections that were incubated with the haptenic sugar (inhibitory controls) (step 14 or 25).

Potential solution
If lectin binding is carbohydrate-mediated, by incubating a lectin with its haptenic sugar, the haptenic sugar will occupy the lectin-binding site and prevent the lectin from binding to the glycans in the tissue, leading to significantly decreased signal in those sections (complete abolition should not always be expected). Consult manufacturer's instructions to confirm the most appropriate sugar to use. Additionally, it is worth noting that some lectins can only be inhibited by complex sugars and not monosaccharides, in which cases appropriately glycosylated glycoproteins should be employed (Brooks and Hall, 2012). If lectin binding is not significantly decreased in the presence of the haptenic sugar, lectin binding is either non-specific or mediated by the non-carbohydrate binding sites present in the lectin (Gerlach, et al., 2012). A different lectin with similar desired specificity should be selected in the latter case.

Potential solution
If a high background staining is detected, there are several steps that can be tried to reduce it. This can be due to several reasons: Reason 1: Insufficient blocking of non-specific binding sites, which can be overcome by increasing the concentration of periodate-treated BSA or by extending the incubation period of the blocking step. Reason 2: Excessive lectin concentration, incubation temperature, or incubation time. In these cases, it is advised to reduce the concentration of lectin and/or to incubate sections at 4 C instead of at room temperature and/or for a reduced period of time. Also, the lectin used could be potentially diluted in a buffer containing the blocking agent (e.g., 0.1% periodatetreated BSA in TBS-T or TBS). The use of other blocking agents (such as normal serum, which is frequently used in immunohistochemistry) is not appropriate as it possesses multiple glycosylated molecules, which might inhibit specific binding of lectins to the tissue or cause nonspecific binding. Reason 3: Number and duration of washes are too short. These can be increased to ensure that any lectin that is non-specifically bound is removed. Reason 4: The tissue might have physiological structures/molecules that can cause auto fluorescence due to endogenous fluorophores (e.g., lipofuscin) (Croce and Bottiroli, 2014). To solve this problem, it is crucial to know the features and composition of the tissue and try to address them by finding a specific compound that can bind to these structures and reduce their auto fluorescence.

Problem 4
No fluorescence signal is detected in the stained tissue sections (step 14 or 25).

Potential solution
Lack of fluorescence signal in the samples might be caused by multiple factors: Factor 1: Not enough lectin is bound to glycans of interest, for which the recommended course of action would be to use a higher concentration of lectin, to incubate it during a longer period (e.g., 2 h or even 3 h) or to incubate it at a higher temperature (e.g., at 37 C). Factor 2: The lectin might have been improperly stored (e.g., kept at room temperature for long periods or in an environment not protected from light), which could lead to quenching of its signal or loss of its function. To ensure that the lectin has kept its function and labeling, a positive control should be run in parallel in every staining experiment. This positive control should be a tissue known to express the glycosylation motif specific to that lectin as characterized by the literature or based on previous in-house experiments. Factor 3: The specific glycosylation motif might not be present in the tissue sample. To confirm this, a positive control using a sample where this glycosylation motif has been reported to be found should always be run in parallel. Factor 4: The fixation procedure may modify the glycan structure that is necessary for recognition by the lectin. To solve this issue, a shorter fixation period or a different fixation method are recommended.

OPEN ACCESS
Factor 5: TBS buffer might be contaminated with bacteria, which might alter its properties and compromise the staining and functions of the lectin. For this, it is advised to only use freshly prepared sterile-filtered or autoclaved TBS.

Problem 5
The positive signal associated with the glycan distribution is not homogenous across the tissue section (step 14 or 25).

Potential solution
Consider the histological nature of the region of interest where the lectin binding is localized. Suppose the region of interest has an extended and multi-layered morphology (e.g., artery), suitable confocal high magnification imaging needs to be implemented to detect the different layers of these structures. Then, only the relative areas of interest in ImageJ should be selected for fluorescence intensity measurement.

RESOURCE AVAILABILITY
Lead contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Prof. Abhay Pandit (abhay.pandit@nuigalway.ie).

Materials availability
This study did not generate new unique reagents.

Data and code availability
This study did not generate/analyze any datasets/code.