Tetrameric Fluorescent Antigen Arrays for Single-Step Identification of Antigen-Specific B Cells

Summary

This protocol details the customizable production and use of fluorescent probes for labeling antigen-specific B cells.

Abstract

Fluorescent antigen production is a critical step in the identification of antigen-specific B cells. Here, we detailed the preparation, purification, and the use of four-arm, fluorescent PEG-antigen conjugates to selectively identify antigen-specific B cells through avid engagement with cognate B cell receptors. Using modular click chemistry and commercially available fluorophore kit chemistries, we demonstrated the versatility of preparing customized fluorescent PEG-conjugates by creating distinct arrays for proteolipid protein (PLP_139-151) and insulin, which are important autoantigens in murine models of multiple sclerosis and type 1 diabetes, respectively. Assays were developed for each fluorescent conjugate in its respective disease model using flow cytometry. Antigen arrays were compared to monovalent autoantigen to quantify the benefit of multimerization onto PEG backbones. Finally, we illustrated the utility of this platform by isolating and assessing anti-insulin B cell responses after antigen stimulation ex vivo. Labeling insulin-specific B cells enabled the amplified detection of changes to co-stimulation (CD86) that were otherwise dampened in aggregate B cell analysis. Together, this report enables the production and use of fluorescent antigen arrays as a robust tool for probing B cell populations.

Introduction

The adaptive immune system plays a critical role in the progression or regression of many disease states, including autoimmunity, cancer, and infectious diseases¹. For broad applications including the study of immunopathology or the development of new precision treatments, it is often critical to assess antigen-specific B and T cell responses underlying disease progression^2,3,4. Major histocompatibility complex (MHC) tetramers are widely and commercially available for identifying antigen-specific T cell clones⁵. These fluorophore-labeled constructs present quadrivalent peptide-MHC complexes to avidly engage with cognate T cell receptors for labeling applications such as microscopy and flow cytometry.

Antigens for B cell interrogation can present highly varied molecular weights, charges, and solubilities^6,7. When using monomeric antigen as soluble B cell probes, physicochemical antigen properties may not be stabilized through complexing with the much larger, water soluble streptavidin molecule, or could present solubility issues as monomeric reagents prior to conjugation. Thus, some proteins present bioconjugation difficulties and unexpected results in practice^7,8. Direct fluorescent dye conjugation can sometimes render constructs water insoluble and lipophilic. These direct dye-antigen compounds are susceptible to nonspecific embedding within cell membranes, confounding antigen-specific analyses^7,8,9. Some strategies have overcome solubility challenges by coupling antigen and fluorophores with other functional groups. Cambier et al., for example, employed biotinylated insulin in its native form to engage with insulin-specific B cell receptors (BCRs) before adding fluorophore-labeled streptavidin in a stepwise fashion¹⁰. While this approach enabled the assessment of B cells that bind to monomeric insulin with high resolution, two labeling steps were required. A generalizable protocol for the preparation of ready-to-use polymer-based B cell probes that is readily integrated with common fluorescent antibody labeling procedures would be of benefit for furthering the study of B cells in disease.

In this protocol, we detail the production and use of fluorescent antigen arrays (FAAs) for the generalizable and single-step labeling of antigen-specific B cells for microscopy and flow cytometry experiments (Figure 1). Soluble antigen arrays (SAgAs) have been employed over the past decade as B cell-targeted antigen-specific immunotherapies against autoimmune diseases^{11,12,13,14,15,16,17,18,19,20,21,22}. SAgAs leverage multivalent antigen display on flexible, polymeric backbones to avidly engage B cell receptors and elicit immunomodulatory effects, though their antigen-specificity provides another opportunity for probing B cells of interest when coupled to a fluorophore²³. The polymeric backbones constituting SAgAs confer water solubility to the overall biomacromolecule and can dampen the sometimes extreme antigen characteristics that confound probe generation and staining specificity^6,24. We have grafted numerous antigens ranging in size and complexity onto SAgA platform using modular click chemistry, which is conducive to the use of small peptide epitopes and full proteins^14,18. Here, we demonstrate FAAs as robust antigen-specific B cell labeling tools that can be used in parallel with typical fluorescent antibody labeling. We prepared and evaluated FAAs consisting of human insulin for labeling B cells in a transgenic mouse model of Type 1 Diabetes (VH125), as well as FAAs that incorporated proteolipid protein 139-151 (PLP), a peptide epitope for experimental autoimmune encephalomyelitis (EAE), the mouse model of Multiple Sclerosis. Our intention in employing these disease models was to demonstrate the versatility of this platform, both for the modular substitution of antigens used, as well as the viability of use with peptide epitopes (PLP) and full proteins (insulin) alike. This protocol is presented with the purpose of accessibility, without extensive bioconjugation expertise required. The reagents, as well as synthesis and purification methods, are designed to be versatile and readily implemented at most research labs focused in chemistry, molecular biology, or immunology.

Protocol

All animal procedures represented in this work were approved by the Institutional Animal Care and Use Committee at the University of Kansas.

1. Antigen array synthesis (4–6 days)

1. Functionalize unmodified antigen with an alkyne handle (1 h). Add 1 equivalent insulin (100 mg, 17.4 μmol) to a 20 mL glass vial with a stir bar and dissolve in anhydrous dimethylsulfoxide (DMSO) (2 mL) with gentle heating to 40–50 °C using a heat-gun or water bath.
   1. Add 1,1,3,3-tetramethylguanidine (40.2 mg, 348.8 μmol) and 1.35 equivalents freshly prepared 78.5 mM propargyl N-hydroxysuccinimide ester (NHS-ester) stock solution in anhydrous DMSO (0.3 mL, 5.3 mg, 23.6 μmol).
   2. Stir for 30 min at room temperature then quench the reaction with 0.05% HCl (12 mL).
2. Purify the singly modified insulin-alkyne by reverse-phase liquid chromatography (4 h). Use a preparative C18 column (19 mm x 250 mm, 300 A pore size, 5 μm particle size) with a 10 min 30–40% B gradient (A, water with 0.05% trifluoroacetic acid (TFA); B, acetonitrile with 0.05% TFA) and a flow rate of 14 mL/min. The desired product elutes immediately after unmodified insulin.
   1. Evaporate acetonitrile and TFA by nitrogen gas stream or rotary evaporation under reduced pressure (650 Pascals) and rotating at 150 rpm. Freeze the aqueous solution and lyophilize to dryness. Store functionalized insulin at -20 °C under a dry atmosphere.
      NOTE: The functionalized insulin is stable for up to a year under these storage conditions.
3. Synthesize the antigen array by copper-catalyzed, azide-alkyne cycloaddition (CuAAC) (2.5 h)²². Add 6 equivalents insulin-alkyne (38 mg, 6.3 μmol) to a 10 mL glass vial with a stir bar and dissolve in DMSO (1.2 mL) with gentle heating.
   1. Add 50 mM sodium phosphate buffer (1.8 mL) pH 7.4, 1 equivalent 20 kDa 4-arm PEG azide (21 mg, 1.05 μmol), copper (II) sulfate pentahydrate (3.15 mg, 12.6 μmol), Tris(3-hydroxypropyltriazolylmethyl)amine (27.37 mg, 63 μmol), and sodium ascorbate (50.34 mg, 254 μmol).
   2. Stir for 2 h at room temperature then add DMSO (3 mL) to solubilize any precipitates and acidify the solution with 0.05% HCl (4 mL).
4. Purify the antigen array by reverse-phase liquid chromatography (3 h). Use a preparative C18 column (19 mm x 250 mm, 300 A pore size, 5 µm particle size) with a 10 min 20–60% B gradient (A, water with 0.05% TFA; B, acetonitrile with 0.05% TFA) and a flow rate of 14 ml/min. The desired product elutes immediately after insulin-alkyne.
   1. Evaporate acetonitrile by nitrogen gas stream or rotary evaporation under reduced pressure. Freeze the aqueous solution and lyophilize to dryness. Store at -20 °C under a dry atmosphere.
      NOTE: The insulin antigen array is highly hygroscopic. The lyophilized fibers may coalesce into a dense pellet after repeated exposure to the atmosphere. FAA properties may differ depending on the application-specific antigen used. The insulin antigen array is stable for several months under these storage conditions.

2. Fluorophore conjugation (1–2 days)

1. Conjugate the fluorophore to the antigen array (2.5 h). Add 1 equivalent tetravalent insulin (21.7 mg, 0.5 μmol) to a 20 mL glass vial with stir bar and dissolve in DMSO (1.75 mL) with gentle heating. Add freshly prepared 100 mM carbonate buffer (8 mL) pH 9.0 and 5 equivalents of fluorescein isothiocyanate in a freshly prepared 10 mM stock solution in DMSO (0.25 ml, 0.97 mg, 2.5 μmol). Stir for 2 h in the dark at room temperature.
2. Purify the product by dialysis (24 h). Use 3.5 kDa molecular weight cutoff dialysis tubing in a stirred 5 L bucket with distilled water in the dark at room temperature. Dialyze for 24 h and change the dialysis solution every 6–12 h.
3. Freeze the dialyzed solution and lyophilize to dryness to yield the FAA. Store in the dark under a dry atmosphere at -20 °C.
   NOTE: The FAA is stable for several months under these storage conditions.

3. FAA characterization (2–4 h)

1. Analyze the products of steps 1 and 2 by sodium dodecyl sulfate-polyacrylamide gel electrophoresis according to Laemmli²⁵ (2 h). Prepare samples containing 5 μg of purified monovalent antigen, 20 kDa 4-arm PEG azide, antigen array, and FAA.
   1. Visualize fluorophore labeling in the FAA samples by fluorescence imaging in a gel-imager.
   2. Perform Coomassie blue staining²⁶ to visualize the antigen.
   3. Perform iodine staining²⁷ to visualize the PEG backbone. Rinse the gel in distilled water (3 x 2 min), incubate in 5% barium chloride solution (20 mL) for 10 min, rinse in distilled water (3 x 2 min), incubate in 0.1 N iodine solution (20 mL) for 1 min, then rinse in distilled water (3 x 2 min) to remove background staining before visualization.
      NOTE: To make 0.1 N iodine solution, add potassium iodide (2.0 g, 12 mmol) and iodine (1.27 g, 5 mmol) to 100 mL distilled water.
2. Calculate the degree of dye-labeling by UV-Vis spectroscopy (1 h). Dissolve a small amount of FAA at 0.1 mg/mL and record the absorbance at 280 nm and the peak absorption wavelength for the dye. Obtain the molar extinction coefficients for the antigen and the dye.
   NOTE: Some dyes are pH sensitive. Make sure to use an appropriately buffered solution like100 mM carbonate buffer with pH 8.3 when measuring the FAA absorbance to ensure the reported dye extinction coefficient is accurate.
   1. Perform the following calculation²⁸ where A_{Dye, maxλ} is the FAA absorbance at the peak absorption wavelength for the dye, E_{Dye, maxλ} is the molar extinction coefficient at the peak absorption wavelength for the dye, and E_{antigen, 280 nm} is the molar extinction coefficient for monovalent antigen at 280 nm.:
      Correction Factor (CF) = A_{Dye, 280 nm} / A_{Dye, maxλ} 0.25 * (A_{Dye, maxλ} / E_{Dye, maxλ}) * (E_{antigen, 280 nm} / (A_{280 nm} – (A_{Dye, maxλ} * CF))) = dye/FAA
3. Analyze the FAA for free dye or potential degradation products by RP-HPLC (1 h).
   1. Dissolve a small amount of FAA at 1.0 mg/mL and analyze with an analytical C18 column (4.6 mm x 150 mm, 300 A pore size, 2.5 μm particle size) using a 10 min 5–95% B gradient (A, water with 0.05% trifluoroacetic acid: B, acetonitrile with 0.05% trifluoroacetic acid) with a flow rate of 1 mL/min. Set the UV-Vis detector to monitor the peak absorbance wavelength of the dye (437 nm for FITC).
      NOTE: For pH sensitive dyes, HPLC analysis in acidified solvents may require shifting the monitored wavelength from the dye’s peak absorption wavelength in neutral, aqueous conditions. In 0.05% TFA, the pH is approximately 3 and the max absorption for FITC has shifted.

4. Assay development by FAA titration (3–5 h)

NOTE: FAA use for flow cytometry is presented, but the same steps can be modified for use in other formats such as immunohistochemistry or fluorescent microscopy. When applying FAAs for a new format, a new optimization assay must be completed. Mixed or isolated cell populations may be obtained through blood or lymphoid organ processing methods^23,29. Harvest through enzymatic digestion of tissue is not recommended, as cells surface markers may be shed.

1. Suspend FAA stock to 1 mg/mL in FACS buffer (1x PBS + 5% fetal bovine serum and 0.1% sodium azide). Up to 10% DMSO can be incorporated to accelerate dissolution.
2. Obtain cells positive for antigen-specific B cells of interest and dispense into microcentrifuge tubes. Approximately 1 x 10⁷ total cells will be required.
   NOTE: Splenocytes were harvested from VH125 and EAE mice according to IACUC-approved protocols at the University of Kansas for the present demonstration. Researchers may employ the methods described herein for cell samples specific to their own application and FAA preparation.
   1. Dispense 5 x 10⁵ cells for titration labeling replicates (at least 3 replicates per titration).
   2. Dispense 1 x 10⁵ cells for single-stain controls as well as an unstained control.
      NOTE: Antibody isotypes and fluorescence minus one control may also be used to fully validate the assay.
3. Wash cells by adding 1 mL of FACS buffer to each tube. Centrifuge at 200 x g for 5 min.
4. Aspirate the supernatant and resuspend cell pellets in 50 µL of FACS buffer.
5. Add CD3 and CD19 fluorescent antibodies to each sample at manufacturer recommended concentrations, as well as titrated FAA doses.
   NOTE: Fluorescent antibody working concentrations may be titrated independently to confirm the lot integrity. Appropriate FAA dose ranges will vary by application, but a good starting point is 50, 25, 10, 5, and 1 μg per sample. 1 µL of stock FAA solution equals 1 µg of dose.
6. Mix each sample well and incubate covered from light, on ice for 30 min.
7. After the incubation, wash cells by adding 1 mL of FACS buffer to each tube. Centrifuge the samples at 200 x g for 5 min.
8. Aspirate the supernatant and repeat the wash by adding 1 mL of FACS buffer to each tube. Centrifuge the samples at 200 x g for 5 min.
9. Aspirate the supernatant and resuspend the samples in 200 µL of fresh FACS buffer. Place the samples on ice and head to the cytometer.
10. Run the samples on the cytometer and collect at least 50,000 events.

5. FAA titration analysis and labeling dose optimization (1–2 h)

1. Using flow cytometry analysis software, gate for single cells.
2. On single cells, place gates for CD19+ (B cells) and CD3+ (T cells).
3. Within CD19+ and CD3+ parent populations, gate for FAA+ events in the relevant fluorochrome channel.
4. Record the percentage of FAA+ events within the CD19+ and CD3+ parent populations.
5. Calculate a specificity ratio by dividing the proportion of FAA+ events in the CD19+ population (specific) by the proportion of FAA+ events in the CD3+ population (nonspecific). Specificity ratio should be maximized for successful FAA use.
6. Employ the labeling dose corresponding to the highest specificity ratio for future experiments.

Results

The purified yield of insulin-alkyne (Figure 2, upper panel), determined by weight, typically varied from 50–65%. Yields of less than 40% were likely caused by water contamination in the anhydrous DMSO and or hydrolysis of the propargyl NHS-ester. For antigen multimerization (Figure 1B), the purified yield of the insulin antigen array (Figure 2, middle panel) varied from 60–75% and SDS-PAGE analysis confirmed the major p...

Discussion

We developed a protocol (Figure 1) to construct customized FAAs for identifying antigen-specific B cells, simplifying the generation of B cell probes for difficult antigen targets. We selected 4-arm PEG polymers with terminal azide groups as a facile substrate for building FAAs, as PEG confers water solubility while the functional azide handles enable simple click conjugation reactions³³. The defined number of functional handles (4 arms) is conducive to simplified che...

Materials

Name	Company	Catalog Number	Comments
1,1,3,3-tetramethylguanidine	Alfa Aesar	AAA12314AC
12 M hydrochloric acid	Fisher Chemical	A508-4
12% Mini-PROTEAN TGX Precast Protein Gels, 10-well, 30 ul	Bio-Rad	4561043
20 kDa 4-arm PEG azide	Jenkem USA	A7185-1
3.5 kDa MWCO dialysis tubing (regenerated cellulose)	Fisher Scientific	25-152-10
Acetonitrile, HPLC grade	Fisher Chemical	A998-4
Alexa Fluor 647 anti-mouse CD19 Antibody	BioLegend	115522
anhydrous dimethylsulfoxide	ACROS Organics	AC610420010
barium chloride	ACROS Organics	203135000
Brilliant Blue G-250 Dye	Fisher BioReagents	BP100-50
Cell Prime r-insulin	EMD Millipore	4512-01	Recombinant human insulin for insulin FAA synthesis
Copper (II) sulfate pentahydrate	ACROS Organics	AC197722500
dimethylsulfoxide	Fisher Chemical	S67496
fluorescein isothiocyanate (FITC) isomer 1	Sigma-Aldrich	F7250-1G
Glacial acetic acid	Fisher Chemical	A38-212
Glycerol	ACROS Organics	15892-0010
Glycine	Fisher Chemical	G46-500
homopropargyl-PLP	Biomatik	NA	Custom synthesis (sequence: homopropargyl-HSLGKWLGHPDKF; purity: 97.29%)
iodine	Sigma-Aldrich	207772-100G
Methanol, HPLC grade	Fisher Chemical	A452-4
NHS-Rhodamine (5/6-carboxy-tetramethyl-rhodamine succimidyl ester), mixed isomer	Thermo Scientific	46406	A commercially available analog of the Rhodamine-B NHS ester used in the paper
PE/Cyanine7 anti-mouse CD3 Antibody	BioLegend	100220
potassium iodide	Sigma-Aldrich	30315
propargyl N-hydroxysuccinimide ester	Sigma-Aldrich	764221
sodium ascorbate	Sigma-Aldrich	A7631
sodium biocarbonate	Sigma-Aldrich	S5761-1KG
sodium dodecyl sulfate	Fisher BioReagents	BP166-100
Sodium phosphate monobasic monohydrate	Fisher Chemical	S468-500
trifluoroacetic acid	Sigma-Aldrich	302031-10x1mL
Tris base	Fisher BioReagents	BP152-500
Tris(3-hydroxypropyltriazolylmethyl)amine	ClickChemTools	1010-1000
xBridge BEH C18 3.5 um, 4.6 x 150 mm column	Waters	186003034
xBridge BEH C18 5um OBD Prep Column, 19 x 250 mm	Waters	186004021