A Novel Surface Modification and Immobilization Method of Anti-CD25 Antibody on Nonwoven Fabric Filter Removing Regulatory T Cells Selectively

Abstract

Anti-CD25 antibodies were immobilized on polypropylene (PP) nonwoven fabrics to specifically remove mouse regulatory T cells (Tregs) from mouse spleen cells. PP fibers were coated with peptide nanosheets, which were prepared by self-assembling of a mixture of X-poly(sarcosine)-b-(l-Leu-Aib)₆ (X: glycolic acid or a phenylboronic acid) and Y-poly(sarcosine)-b-(d-Leu-Aib)₆ (Y: glycolic acid or diazirine derivative). Anti-CD25 antibodies were immobilized by covalent linking between the sugar moiety of the antibody and the phenylboronic acid group on the peptide nanosheet. The removal rate of mouse Tregs from the mouse spleen cells was more than 95% only by passing the filters, while the nonspecific removal rates of other cells were less than 15%. The coating of peptide nanosheets on PP fibers was very effective to provide a suitable environment for the immobilized antibody to interact with the counterpart cells while the coating suppressed nonspecific adsorption of other cells.

1. Introduction

Regulatory T cells (Tregs) are known to possess suppressive effects on the immune system and play important roles, for example, in the immune tolerance.^1,2 The expression of the transcription factor Foxp3 is an important cell marker on characterizing Tregs.³ Among several Treg subsets, there are accumulating evidences that CD25⁺Foxp3⁺CD4⁺ Tregs prevent the hosts from various diseases such as autoimmune diseases and inflammatory diseases.^2,4 In general, Foxp3 expression is upregulated by the IL-2 signaling, and Tregs are activated by the pathways via CD25, CTLA-4, and glucocorticoid-induced tumor necrosis factor receptor (GITR), resulting in the suppression of the function of antigen-presenting cells (APCs). Consequently, healthy cells and tissues can survive to avoid the attacks of self-responsive antibodies and effector T cells.² On the contrary, the reduction or suppression of Tregs leads to the onset of autoimmune diseases due to the absence of negative feedback. On the other hand, CD25⁺Foxp3⁺CD4⁺ Tregs are involved in the growth of tumors due to their suppressive effects on the immune responses.⁵ Tumor microenvironments are full of various chemokines, which recruit the immune cells including APC and cytotoxic T lymphocyte (CTL).⁶ To the counterpart of it, chemokine CCL22, which is produced from tumor cells, tumor-infiltrating macrophages, myeloid suppressor cells, or dendritic cells, recruits Tregs expressing CCR4 into the tumor region.⁷ The recruited Tregs could suppress the cytotoxicity of APC and CTL via the immunity checkpoint molecule of CTLA-4 expressed on the surface of Tregs and further produce the inhibitory cytokines (TGF-β and IL-10) and cytotoxic substances (perforin and granzyme), resulting in the survival and growth of the cancer cells to lead severe damage on the host. In this respect, the cancer therapy to reduce the number of infiltrating Tregs to the tumor microenvironment has attracted considerable attentions as well as the cancer therapy targeting cytokines and immunity checkpoint molecules. Recently, the antihuman CCR4 monoclonal antibody was reported to reduce the number of Tregs and augmented NY-ESO-1-specific CD8⁺ T-cell responses in an adult T-cell leukemia-lymphoma patient.⁸ An antihuman CCR4 monoclonal antibody, KW-0761, also depleted Tregs from the tumor environment in a patient to evoke the cancer immune response.⁹ Therefore, the removal of Tregs in vivo can improve the response of the cancer therapies of chemotherapy and immunotherapy using cytokines and monoclonal antibodies.

The apheresis therapy is a treatment by removing the cells or proteins causing the disease. The therapy has already been applied for patients suffering from hypercholesterolemia, leukemia, and inflammatory bowel disease (IBD). In these cases, the nonwoven fabric filters were designed suitably for eliminating low-density lipoprotein (LDL) and leukocytes by the extracorporeally circulating process of the patient’s blood. The removal of Tregs from septic patients using polymyxin B-immobilized fiber (PMX-F) has also demonstrated to eliminate about 50% Tregs in peripheral blood mononuclear cells (PBMCs).¹⁰ However, the specificity in Treg elimination was not high so that the nonspecific removal of PBMCs caused the adverse effect of lowering the immune response against foreign microbes.¹¹

To improve the specificity of the removability by filters, antibody usage for the filter has been frequently examined. For example, the selective adhesion of CD25⁺ cells was achieved by the anti-mCD25-IgG-immobilized poly(ethylene) film and their recovery from the film was also successful.¹² Especially in apheresis therapy, there are two ways to achieve specific elimination. One is site-oriented immobilization of antibodies onto the nonwoven fabric fiber surface. Some site-oriented antibody immobilization methods were demonstrated that use phenylboronic acid, which conjugates with the oligosaccharide moiety presented in the Fc region of the antibody or the use of protein A, which specifically binds to the Fc fragment of IgG.^13,14 The other way is the hydrophilic surface modification to suppress nonspecific adsorption of other cells. Rafting the hydrophilic polymer as typified by poly(vinylpyrrolidone) on the nonwoven fabric by the atomic transfer radical polymerization (ATRP) method shows the suppression of bacterial adhesion.¹⁵ However, the high polymerization degree is needed to make the PP fiber surface hydrophilic enough to suppress cell adhesion. It follows that the terminal of the polymer chain may be covered by other hydrophilic polymer chains. Therefore, there are some difficulties to establish a compatible hydrophilic surface introducing site-oriented antibody immobilization to capture target cells efficiently and having high densely hydrophilic polymer chains to suppress nonspecific cell adhesion.

In this study, the removal filter specific for Tregs is developed by the nonwoven fabric immobilized with the anti-CD25 antibody. We propose here a novel and very simple preparative method using peptide nanosheets for the antibody-immobilized nonwoven fabric in an attempt to suppress the nonspecific cell adhesion to the filter (Figure 1). Previously, we have reported that an equimolar mixture of the poly(sarcosine)-b-(l-Leu-Aib)₆ (Aib: 2-aminoisobutyric acid) (SL, Figure 2) and poly(sarcosine)-b-(d-Leu-Aib)₆ (SD) self-assembles into planar nanosheets.¹⁶ The thickness was about 10 nm, indicating the interdigitated monolayer structure comprising the heterochiral helical peptides. The surface density of poly(sarcosine) on the nanosheet was so high that immune cells could not recognize a possible antigen of poly(sarcosine) chains covering the nanosheet.^17,18 Furthermore, the surfaces of molecular assemblies prepared from SL and SD were able to be functionalized with thiol,^19,20 adenine, and oligosaccharide^21,18 by their introductions at the terminal of poly(sarcosine) chains. Accordingly, the peptide nanosheets can be designed to have two faces, one for the phenylboronic acid to connect to antibodies and the other for the photoactivatable radical generator to react with PP fiber. These functionalized peptide nanosheets are attainable for coating of the PP fiber to exhibit compatibly site-oriented antibody immobilization and suppression of nonspecific cell adhesion. Using a free-standing nanosheet is advantageous in view of avoiding the difficulty of the molecular assembly morphology change during the process of the surface modification, which is inevitable for the current liposome-based surface modification method.²² The specific removal ratio of Tregs from mouse spleen cells through the filter was evaluated.

Figure 1.

Schematic illustration of the preparation of the nonwoven fabric immobilized with anti-CD25 antibody. (a) The amphiphilic polypeptides having various functional groups. The red circle indicates the antibody binding group. The green hexagon indicates the cross-linking group to the nonwoven fabric. (b) Molecular assembly taking an interdigitated monolayer structure. (c) Immobilization of the nonwoven fabric with the peptide nanosheet by photochemical reaction. (d) Conjugation of anti-CD25 antibody with the peptide nanosheet on the nonwoven fabric.

Figure 2.

Chemical structures of the amphiphilic polypeptides modified with various functional groups at the terminal of the poly(sarcosine) block.

2. Results and Disscusion

2.1. Design of Antibody Orientation and Surface Modification

The performance of apheresis filters carrying antibodies, IgGs, to remove specific cells, for example, from peripheral blood mononuclear cells, depends sensitively on the immobilization method of antibodies that should be designed so as to avoid impairing the binding ability of the immobilized antibodies and to suppress nonspecific cell adhesions to filters. In general, the chemical immobilization of antibodies occurs randomly to the functional groups in the antibody due to the multiple available sites in it. Since IgG molecules are glycoproteins having sugar units at the Fc region,²³ the use of phenylboronic acid as a linker chromophore can limit the reaction site in IgG molecules to cis-diol of the sugar units, leading to the orientational control of the immobilized IgG.^13,24 We propose here a new method of coating fibers with molecular assemblies, which prevent nonspecific cell adhesion and introduce the functional group connecting with the antibody, taking a nanosheet morphology as shown in Figure 1.

The photoactivatable 3-phenyl-3-(trifluoromethyl)-3H-diazirine moiety was introduced at the chain terminal of SD (DSD) via the Cu-catalyzed azide alkyne cycloaddition (CuAAC).²⁵ The chain terminal of SL was modified with phenylboronic acid (BSL)^24,26,27 and also with maleimide (MSL) to be linked to the antibody. SL was labeled by fluorescein (FSL) to evaluate the coating rates of the peptide nanosheets onto the nonwoven fabric fibers by a fluorescence microscope observation or fluorescence quantitative measurement.

2.2. Preparation of Peptide Nanosheets

Various mixtures of l- and d-form amphiphilic polypeptides of the ethanol solution were injected into Milli-Q water to form molecular assemblies (Table 1). In most cases, planar nanosheets were observed in accordance with the design of the molecular assembly. However, an equimolar mixture of FSL and DSD generated vesicles as well as planar sheets (entry 1, Figure 3a). The bulky fluorescein groups on one face of nanosheets may provide the sheet with curvature to induce the morphology change from the planar nanosheet to the vesicle. The content of FSL was therefore reduced to 15% of the total amphiphilic polypeptides to generate only planar nanosheets (entry 2, Figure 3b). Then, we check the morphology changes in the increase of ethanol concentration, which let the nanosheets infiltrate easily. The nanosheet sizes increased from approximately 800 nm in the absence of ethanol to approximately 1090 nm with 20% ethanol in the dispersion medium. However, in the case of 20% ethanol, large aggregates were formed, which were not observed in less ethanol content. The planar sheets therefore were prepared in the presence of 10% ethanol. Similarly, an equimolar mixture of heterochiral peptides of MSL and DSD or BSL and DSD self-assembled into planar nanosheets with approximately 400 nm size, which grew into larger sheets with approximately 740 or 1300 nm in the presence of 10% ethanol (Table 1 and Figure 3).

Table 1. Molecular Assemblies Prepared by Combinations of the Heterochiral Amphiphilic Polypeptides^a.

entry	contentsa	molecular ratio (%)	ethanol content (%)	morphology	sizeb (nm)
1	FSL/DSD	50/50	0	vesicle/sheet	730
2	FSL/SL/DSD	15/35/50	0	planar sheet	800
3	FSL/SL/DSD	15/35/50	5	planar sheet	870
4	FSL/SL/DSD	15/35/50	10	planar sheet	930
5	FSL/SL/DSD	15/35/50	20	aggregate	1090
6	MSL/DSD	50/50	0	planar sheet	460
7	MSL/DSD	50/50	10	planar sheet	740
8	BSL/DSD	50/50	0	planar sheet	420
9	BSL/DSD	50/50	10	planar sheet	1250
10	SL/DSD	50/50	0	planar sheet	420
11	SL/DSD	50/50	10	planar sheet	620

^aAn aliquot of ethanol solution of the mixed peptides was injected into MilliQ-water and stirred at 4 °C for 30 min.

^bDetermined by DLS measurements.

Figure 3.

TEM images of molecular assemblies prepared from various heterochiral combinations of amphiphilic polypeptides in Table 1. (a–k) TEM images correspond to the molecular assemblies from entry 1 to 11 in Table 1.

2.3. Coating Nonwoven Fabric with Peptide Nanosheets

The nonwoven fabric was made hydrophilic by UV/O₃ treatment. The peptide nanosheets were prepared from a mixture of FSL and DSD in the presence of 10% ethanol (entry 4 in Table 1) and incubated with nonwoven fabrics under irradiation of a Xe lamp. After washing the fabric with Milli-Q water, fluorescence microscopy was carried out to confirm the surface modification by peptide nanosheets. Fluorescently modified fibers were observed, suggesting the fabrics were successfully covered with peptide nanosheets (Figure 4). The surface of the nonwoven fabric fiber was further analyzed with atomic force microscopy (AFM). Ten nanometer height sheet structures were observed frequently with the fabrics treated with peptide nanosheets (Figure 5c,d,f), while mechanically smooth surfaces were observed with untreated nonwoven fabric and fabric treated with UV/O₃ (Figure 5a,b,e). These imbricated surfaces (Figure 5c,d) that indicate peptide nanosheets were accumulated on the fabric surface while maintaining sheet morphologies. Simultaneously, DMT-modulus mapping images are also consistent with topological images (Figure 5g–j). Taken together, the peptide nanosheets densely covered the fiber by the dense poly(sarcosine) layer. The amount of the peptide nanosheets on the PP fibers was evaluated by measuring dissociated FSL with dimethylformamide, which collapse the interaction in self-assembly and make FSL drain off from fiber. The coating ratios, which were calculated by dividing the surface area of the peptide nanosheets on the fibers by the total surface area of the fibers, drastically increased with the addition of ethanol as we consider (Figure 6). In the presence of ethanol, the peptide nanosheets became larger sizes and easier to infiltrate into the nonwoven PP fabrics. The coating ratio was as high as about 250% with the addition of 10% ethanol to Milli-Q water, which did not change so much with repetitive photoreactions (Figure 6).

Figure 4.

Fluorescence microscope image of the nonwoven fabric immobilized with the nanosheet prepared from FSL/SL/DSD (entry 4 in Table 1).

Figure 5.

(a–d) 2D-topological and (g–h) DMT-modulus images of (a, g) the untreated nonwoven fabric, (b, h) the nonwoven fabric treated with UV/O₃, (c, i) one time immobilized with the nanosheet prepared from BSL/DSD (entry 9 in Table 1) and (d, j) three times immobilized with the nanosheet. (e, f) The cross section height profile of (b) and (c) (red line).

Figure 6.

Coating ratio of the nanosheet on the nonwoven fabric in Milli-Q water (light green), 5% ethanol in Milli-Q water (yellow-green), and 10% ethanol in Milli-Q water (green).

2.4. Immobilization of the Antibodies to the Peptide Nanosheets on the Nonwoven Fiber

The PP nonwoven fabrics were coated with peptide nanosheets containing MSL or BSL in the presence of 10% ethanol. In the former case, the anti-CD25 antibody was processed by tris(2-carboxyethyl)phosphine (TCEP) to generate thiol groups to react with maleimide groups on the nanosheets. In the latter case, the fabrics with nanosheets were only incubated with the anti-CD25 antibody to conjugate with phenylboronic acid on the nanosheets. The immobilized amounts of anti-CD25 antibodies were evaluated by ELISA using the anti-Fab antibody to be 0.9 and 3.6 μg per one filter coated by MSL and BSL nanosheets, respectively. In the latter case, the surface density of the anti-CD25 antibody is calculated to be 1.9 mg/m², which corresponds roughly to immobilization of one antibody in 130 nm² area. As a reference experiment, the filter coated with nanosheets (SL) without MSL and BSL was resulted in nearly no antibody immobilization (Figure 7). The amounts of the immobilized antibody on the nonwoven fabrics did not increase even when the immobilization reactions were repeated (Figure 7).

Figure 7.

Amount of anti-CD25 antibody on the nonwoven fabric via the functional group on the nanosheet prepared from BSL/DSD (entry 9 in Table 1, BSL, blue bar), MSL/DSD (entry 7 in Table 1, MSL, red bar), and SL/DSD (entry 11 in Table 1, SL, gray bar) after the repetitive reactions. n = 3.

2.5. Treg Removal from Mouse Spleen Cells

The nonwoven fabrics with a 6.8 mm diameter coated with MSL/DSD, BSL/DSD, and SL/DSD nanosheets (entries 7, 9, and 11 in Table 1, respectively) were incubated and immobilized with anti-CD25 antibodies. Six nonwoven fabrics were packed in a 1 mL Mobicols (MoBiTec) column (six filters coated with MSL nanosheets (M6), BSL nanosheets (B6), and SL nanosheets (S6) columns). The numbers of CD25⁺Foxp3⁺CD4⁺ Tcells and whole T cells except Tregs, which were obtained from the mouse spleen, were counted and measured by a flow cytometer before and after passage through the columns, and the removal ratios were calculated (Figure 8a,b). The total number of T cells in the whole spleen cells was 5.8 × 10⁶ including 1.9 × 10⁵ of Tregs. The B6 column removed 90% Tregs while the M6 column showed 65% removal. Addition to the difference of the immobilized antibody amount, the removal ratio difference may be attributable to the binding ability of the immobilized antibody. In the case of the B6 column, the whole antibody was connected to the surface, but the M6 column possessed the antibody of the half size due to the reductive reaction by TCEP. The SL column reflecting the nonspecific adsorption of T cells and Tregs showed the removal rate less than 10%. Notably, nonspecific removals of the whole spleen cells were less than 50% of the highest, and the B6 column was the best as low as 15%. The B6 column therefore specifically removed Tregs.

Figure 8.

(a) Flowcytometry of before cell flow and after B18, B9, B6, M6, or S6 column. (b) The removal ratios of Tregs (red bar) and T cells (gray bar) through the column equipped with the nonwoven fabric immobilizing anti-CD25 antibody (column B18, B9, B6, and column M6) and without the antibody (column S6). Columns B18, B9, or B6 have 18, 9, or 6 nonwoven fabrics conjugated with an antibody via phenylboronic acid, and column M6 have six nonwoven fabrics conjugated with an antibody via the maleimide functional group. (n = 3 except B18; n = 1). The statistical significance of differences of the removal ratio was evaluated by the two-tailed Student t-test. p values: *p < 0.05, **p < 0.01, n.s. (nonsignificant), p ≥ 0.05. (c) The SEM pictures of nonwoven fabrics before cell flow (pre) and first to ninth fabric of after flow of the B9 column.

The removal rates of Tregs were evaluated by increasing the numbers of the PP nonwoven fabrics in the column from 6 (column B6) to 9 (column B9) and 18 (column B18). With the increase of the fabric number, the removal rate raised up to 95% (Figure 8b). Notably, the nonspecific removal of the T cells remained as low as approximately 15% irrespective of the fabric number. It may indicate that nonspecific adsorption occurs at flow channels other than the nonwoven fabric. Each of the nine fabrics in the column B9 after passage of the spleen cells was subjected to the scanning electron microscopy (SEM) observation. The spleen cells were obviously trapped by first to sixth nonwoven fabrics but were scarcely found in the SEM images of seventh to ninth (Figure 8c). This observation also indicates that nonspecific removal of the spleen cells was significantly suppressed by the surface modification of the PP fibers with the peptide nanosheets having anti-CD25 antibodies.

3. Conclusions

The specific removal of Tregs from the mouse spleen cells was demonstrated by using the nonwoven fabrics, which were immobilized with the anti-CD25 antibody via peptide nanosheets. With the surface coating by the peptide nanosheets, the hydrophilic polymer brush suppresses nonspecific cell adhesion and ready for antibody immobilization under the control of the orientation. The removal rate of Tregs was as high as 95% while keeping the nonspecific removal rate of the whole spleen cells less than 15%. Since the nonspecific removal rate did not increase with the fabric number equipped in the column, the specific removal rate owing to the immobilized antibody is expected to satisfy the demanded performance with increasing fabric number in the column without losing the other cells. The highly specific removal of the target cell is an essential demand for the sort of the apheresis equipment to suppress severe adverse effects caused by undesired removal of essential cells for patients.

4. Experimental Section

4.1. General Procedure

Anhydrous CH₂Cl₂, MeOH, and THF were purchased from Wako Pure Chemical Industries, Ltd. Other chemicals were purchased from commercial sources and used without further purification. Silica gel for column chromatography was purchased from Nacalai Tesque (silica gel 60, spherical, neutrality). Transmission electron micrographs (TEM) images were taken using JEM-2000EX II or JEM-1400 (JEOL Ltd., Japan). Dynamic light scattering (DLS) measurements were taken using an DLS-8000KS (Photal Otsuka Electronics). Atomic force microscopy (AFM) images were obtained using with MultiMode 8-HR (Bruker). TLC was carried out on silica gel 60 F254 (Merck). NMR spectra were recorded on a Bruker DPX400 spectrometer. HRMS (ESI-MS analysis) were obtained on an Exactive Plus spectrometer (Thermo Fischer Scientific). MALDI-TOF mass spectra were recorded on an Autoflex III plus (Bruker) spectrometer using super-DHB or α-cyano-4-hydroxycinnamic acid (CHCA) (Sigma-Aldrich) as the matrix. Purification by silica gel column chromatography was carried out by elution of a column with a stepwise gradient elution procedure or using a CombiFlash Rf 75 system with a linear gradient elution procedure on standard conditions. Chemical reactions were monitored by TLC. The TLC plates were visualized by immersion in an appropriate stain (5% w/w ninhydrin in ethanol) followed by heating. The nonwoven fabric manufactured from PP (Asahi Kasei Corporation, Japan) has 0.58 mm in thickness, 5.52 μm in average diameter, and 0.81 m²/g in specific surface area.

4.2. Preparation of Peptide Assemblies

Each amphiphilic polypeptide in ethanol (1 mg/20 μL) was prepared as stock solution. The appropriate mixture of the polypeptides solution was injected to Milli-Q water, Milli-Q water containing 5% v/v ethanol, Milli-Q water containing 10% v/v ethanol, or Milli-Q water containing 20% v/v ethanol and then kept stirring at 4 °C for 30 min.

4.3. Transmission Electron Microscopy (TEM)

The dimension of the molecular assemblies was analyzed by TEM. A drop of dispersion was mounted on a carbon-coated Cu grid and stained negatively with 2% uranyl acetate, followed by suction of the excess fluid with a filter paper. TEM images were obtained at an accelerating voltage of 100 kV.

4.4. Dynamic Light Scattering (DLS)

The hydrodynamic diameters of molecular assemblies were measured by DLS-8000KS (Photal Otsuka Electronics) using the He–Ne laser at 25 °C for DLS measurements.

4.5. Immobilization of the Nonwoven Fabric with Peptide Nanosheet

PP nonwoven fabric was punched out to a diameter of 6.8 mm and was treated with a UV/O₃ hydrophilization treatment apparatus for 8 min. The prepared peptide nanosheet dispersion and the three hydrophilized nonwoven fabrics were placed in a 1.5 mL glass vial tube and stirred under reduced pressure with a vacuum pump over 25 min. After degassing treatment, the mixture was irradiated with ultraviolet light for 10 min with a xenon lamp light source under stirring. After staying the immobilized nonwoven fabric for 20 min, they were equipped with a Mobicol column (Mobitec) and washed with 6 mL of Milli-Q water with a peristaltic pump over 10 min. For repetitive immobilization of the peptide nanosheet, the three immobilized nonwoven fabrics were treated from the degassing process again.

4.6. Fluorescence Microscope Observation of the Nonwoven Fabric Immobilized with the Peptide Nanosheet Bearing with Fluorescein

After the immobilization of the nonwoven fabric with the peptide nanosheet prepared from FSL/SL/DSD in the Milli-Q water containing 10% ethanol, the prepared nonwoven fabric was observed with the fluorescence microscope (Olympus BX51, Olympus). The untreated nonwoven fabric was also observed as a control sample.

4.7. Atomoic Force Microscopy (AFM) Observation of the Immobilized Nonwoven Fabric with the Peptide Nanosheet

After a fiber was pulled out the immobilized nonwoven fabric with the peptide nanosheet, topological and DMT-modulus images of the nonwoven fiber immobilized with peptide nanosheets were measured using with a Multimode V controlled by NanoScope 8 (Bruker) in a peak force tapping mode using the gold-coated silicon tip on the nitride lever (SCANASYST-Air-HR, k = 0.4 N/m, Bruker).

4.8. Determination of the Amount of Peptide Nanosheet Immobilized the Nonwoven Fabric

One nonwoven fabric having photochemically immobilized peptide nanosheets was placed in a 1.5 mL glass vial. DMF (1.0 mL) was added to the vial bottle. After gently stirring for 30 min, the solution was diluted to five times by DMF and measured the fluorescence emission at 445 nm for excited wavelength. Furthermore, for preparing the calibration curves, the emission of FSL in DMF with appropriate concentration was measured. After performing fluorescence measurements and preparing the calibration curve, the amount of FSL in the nanosheet immobilized on the nonwoven fabric was calculated by the calibration curve. From the calculated amount of FSL, the number of amphiphilic polypeptides on the nonwoven sheet was calculated. Considering the area occupied by one molecule of FSL, SL, and DSD is ideally 1.96 nm², the area occupied by the peptide nanosheet was calculated. Finally, compared to the surface area of nonwoven fabric characterized by BET measurement, the coating ratio was calculated.

4.9. Immobilization of Anti-CD25 Antibody with the Nonwoven Fabric via Maleimide on the Peptide Nanosheet

HEPES buffer solution (20 mM, pH 8.2, 500 μL), tris(2-carboxyethyl)phosphine hydrochloride sodium (TCEP) solution (0.5 M, pH 7.0, 42 μL), and antimouse CD25 antibody from rats (0.5 mg/mL, 120 μL, Affymetrix) were mixed in a 1.5 mL vial bottle and were reacted at 37 °C for 90 min. Three nonwoven fabric samples, which immobilized via maleimide on the peptide nanosheet, were immersed and degassed at room temperature for 90 min for the reaction of antibody with maleimide. After the reaction, three nonwoven fabrics were equipped with a Mobicol column (Mobitec Corporation) and then Milli-Q water (2 mL) was flowed by a peristaltic pump over 3 min to remove unfixed antibody and buffer solution.

4.10. Immobilization of Anti-CD25 Antibody with the Nonwoven Fabric via Boronic Acid on the Peptide Nanosheet

Three nonwoven fabric samples, which immobilized via boronic acid on the peptide nanosheet, were immersed in 20 mM HEPES buffer solution (pH 8.2, 500 μL) containing antimouse CD25 antibody from rats (0.5 mg / mL, 120 μL, Affymetrix). After gently stirring at 4 °C for 12 h, three nonwoven fabrics were equipped with a Mobicol column (Mobitec Corporation) and then Milli-Q water (2 mL) was flowed by a peristaltic pump over 3 min to remove unfixed antibody and buffer solution.

4.11. Determination of the Amount of Anti-CD25 Antibody Conjugated with the Nonwoven Fabric

First, anti-rat IgG F(ab′)₂ fragment-specific antibody HRP conjugated (1.0 mg/mL, 0.3 μL, Iwai), anti-rat IgG F(ab′)₂ fragment specific antibody (1.7 mg/mL, 70.32 μL, Iswai), and citrate phosphate buffer (CPB, 46.68 μL) were mixed to prepare a secondary antibody solution. The nonwoven fabric samples conjugated with the anti-CD25 antibody were incubated to the blocking buffer (2% BSA in phosphate-buffered saline (PBS), 300 μL for 30 min at room temperature). After the blocking process, the nonwoven fabric samples were added to CPB (90 μL) and secondary antibody solution (10 μL), incubated for 30 min at room temperature, and washed three times with PBS-T (PBS containing 0.05% Tween 20). To 1.0 mL solution of o-phenylenediamine (OPD, 42.0 mg) in CPB (14.0 mL), 35% aq. H₂O₂ (21 μL) was added the nonwoven fabric samples in a 1.5 mL vial tube. After stirring at room temperature for 25 min, 2 N H₂SO₄ (0.5 mL) was added to the vials and then OPD was determined by UV measurements at 492 nm/reference at 620 nm. A calibration curve is prepared using with secondary antibody solution and the amount of anti-CD25 antibody immobilized on the nonwoven fabric is calculated.

4.12. Removal of Tregs in Spleen Cells through the Nonwoven Fabric Filter Column

The cells in the 6–8 week BALB/C mouse spleen were collected and treated with 5 mL of red blood cell lysis buffer (RBC buffer, Thermo Fisher Scientific) for 5 min. After centrifuging the mixture, cells were pelleted, resuspended in PMRI 1640 medium, and then filtered through a 40 μm membrane filter for removing debris. The cell suspension was diluted to contain 2 × 10⁵ cells/5 mL with RPMI medium. Appropriate nonwoven fabrics having a diameter of 6.8 mm were equipped in the 1 mL MoBiTec column, and the column with the fabric was filled with saline (Otsuka Pharmaceutical Co., Ltd). Five milliliters of prepared cell suspension was flowed through the column at a flow rate of 12 mL/h. After flowing the cell suspension, the amount of the cell suspension passed through the column and the density of cells in the suspension for determining the number of cells in the suspension. After replacing RPMI medium to PBS containing 2% FBS buffer (FACS buffer), the cell suspension after and before passed through the column was centrifuged of 300g at 4 °C for 5 min on the 96 well V-bottom plate and stained at 4 °C in with FACS buffer containing Alexa Fluor 700-labeled antimouse CD3 antibody, BV510-labeled CD8 antibody, BV711-labeled CD4 antibody, PE-labeled CD25 antibody, and Fixible Viability Dye eFlourTM 780 for 15 min. After washing with 200 μL of FACS buffer (centrifugation under 300g at 4 °C for 5 min) for two times, the calls on the plate fixed with Foxp3 stating buffer (Thermo Fisher Scientific) at 4 °C for 1 h. After removing the fixation buffer (centrifugation under 800g at 4 °C for 3 min) and washing with 200 μL permeabilization buffer (Perm buffer) (centrifugation under 800g at 4 °C for 2 min), the cells were stained with the PECy7-labeled Foxp3 antibody in permeabilization buffer at 4 °C for 15 min. After washing with 200 μL of permeabilization buffer (centrifugation under 800g at 4 °C for 2 min) for two times and resuspended in FACS buffer, the cells were analyzed by a flow cytometer (BD LSRFORTESSA X-20, BD Biosciences). The removal ratio was calculated as compared to the number of cells before and after flowing through the column.

4.13. Scanning Electron Microscope (SEM) Observation of the Nonwoven Fabric after Filtering the Spleen Cells

After lyophilizing the nonwoven fabric, the fabrics were examined by scanning electron microscopy (VE-8800, Keyence, Japan) at 1.3–1.7 kV and different magnifications. The samples were sputtered with Pd/Au by an MSP-1S magnetron sputter for 90 s and mounted on the SEM stub with electrical tape (3M, USA).

Glossary

Abbreviations

CD4

cluster of differentiation 4

CD8

cluster of differentiation 8

CD25

cluster of differentiation 25

CCL22

CC chemokine ligand 22

CCR4

CC chemokine receptor 4

CTLA-4

cytotoxic T-lymphocyte-associated protein 4

FoxP3

forkhead box P3

IL-2

interleukin 2

IL-10

interleukin 10

TGF-β

transforming growth factor beta

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.9b03494.

• Experimental details, material synthesis, ¹H NMR spectra and mass spectra of peptides, flow cytometry for CD4 and CD8, and the changes of the CD4⁺ cell ratio in the live T cells (PDF)

The authors declare no competing financial interest.